This is a guest post by Alex Keeney. Alex created Star Spangled Gamblers (RIP), the OG prediction markets podcast, and recently launched , news by prediction market traders for prediction market traders.

Subscribe to Eventual now

If James Talarico wins a Senate seat this November, it would be the ultimate breach of Republicans’ most impregnable fortress: Texas, the Lone Star State, where Democrats haven’t won statewide office since 1994.

But this story is boring.

Everyone already knows it, and no one has much to add. The only thing left between now and Election Day is to find out how real the caricatures of each of these candidates are: James Talarico, the optimistic girly-man running for Senate in America’s most macho state, versus Ken Paxton, who is so scandal-plagued that his own Republican colleagues in Austin impeached him.

Yawn. I’ll see you in November.

The markets have correctly sniffed out that this is a near-50/50 race. And I don’t like trading in markets where my only option to turn a profit is to be right. Better to get ahead of things and make some buys in races where the public hasn’t figured out that the winds are changing.

And in Texas, that place is along the border, all the way from the outskirts of El Paso to the Rio Grande Valley.

Traders would be wise to look at four congressional districts there that Texas Republicans gerrymandered hoping to flip or maintain control of — but where they may come up short. These are:

• TX-15

• TX-28

• TX-34

• TX-23

These districts are all majority Hispanic and have swung to the right in recent cycles.

Trump’s conquest of these historically Democratic bastions fed into a redistricting bacchanal, where GOP operatives got drunk on MAGA Kool-Aid and convinced themselves that the long-prophesied re-alignment of Latinos was now complete. Those assumptions got baked into Texas’s new Congressional maps, where Trump’s strong performance in 2024 became the base case. It may have been a foolish mistake.

“My red flag of caution to the Republicans the day after the election was: this doesn’t mean they’re becoming Republicans. It means that they’re becoming less Democratic,” said Mike Madrid, a Republican strategist and author of The Latino Century.

Subscribe now

In many parts of America, being MAGA is not the same as being Republican. Trump has won the presidency twice — and missed it by a hair in 2020 — by turning low-propensity voters, who more generic candidates have failed to rally. A recent study by the Texas Tribune shows this will probably happen at the border this November.

The Tribune’s data showed Trump winning an impressive 55% of the border vote in 2024 (up from 41% in 2022). But Sen. Ted Cruz carried a more humble 45% that same year, as Democratic incumbents Henry Cuellar and Vincente Gonzalez won re-election.

“We have absolutely missed on the magnitude of Latino shifts to the right over the last few election cycles,” GOP pollster Patrick Ruffini told us. “So I don’t rule out the fact that you might miss on the magnitude of a shift back left.”

“2024 was just such an aberration of election for Republicans and I think we’ll find this to be an aberration of an election for Democrats,” said Brad Johnson, reporter and co-founder of Texas Bullpen.

Translation: expect some Democratic wins.

Share

Then, there is the issue of turnout. In South Texas, it is notoriously low. That’s a problem for Republicans, because their gains have come in favorable years for their side, when turnout boomed compared to averages.

Now, winter is coming and the atmospherics for the GOP are bone-chilling: a D+8 generic ballot, high gas prices, near-record-low approval ratings for Trump, and a very delicate situation along the border, where controversial immigration and tariff policies have a tangible effect. Those are not the pre-conditions you’d expect for another surge in Republican turnout.

New York Times data journalist Nate Cohn has captured this trend in his surveys, which have shown Hispanic voters shifting dramatically to the left.

“South Texas Democrats are going to be able to outrun the partisan fundamentals,” Ruffini told us. We agree. Here is how we see the key races.

TX-15 — Monica De La Cruz (R) vs. Bobby Pulido (D)

Monica De La Cruz won comfortably in 2024, beating Michelle Vallejo by 14 points. But she ran behind Trump, and her opponent this year is a Republican’s worst nightmare.

Bobby Pulido is a Tejano music star with a great campaign schtick: show up at people’s parties, play some music, make the pitch. Brad Johnson says that gives him something in South Texas that “no amount of money can buy.”

Split Ticket forecaster Lakshya Jain told us in August that “I think Pulido is actually a favorite in this seat.”

The limited polling has also shown a much closer race than 2024. And House Majority PAC, a Democratic super PAC, so apply the appropriate propaganda discount, says Pulido is running 15 points better than De La Cruz among 2024 Trump voters. The New York Times’s data shows Hispanics in this district moving 25 points to the Left, which could by itself offset Trump’s 2024 edge.

Backing Pulido to pull the upset is becoming a popular trade and I have joined in on it. A lot of the value is gone at the current 70 cent price, but it remains a good way to get exposure to a large blue wave.

TX-28 — Henry Cuellar (D) vs. Tano Tijerina (R)

In 2024, Democrat Henry Cuellar won by roughly six points even as Trump carried his district by seven. Republicans responded by redrawing TX-28 into a seat Trump would have won by roughly 10. But that doesn’t offset Cuellar’s edge in 2024, especially in the face of a possible wave in November. The markets are very confident that Cuellar holds this seat and I think they’re right. Cuellar is the ultimate survivor, second only to Trump. He’ll beat his opponent, just like he’s beaten FBI probes and obesity.

Here’s the potential spoiler: Cuellar’s opponent, Webb County Judge Tano Tijerina, is also unusually well suited to South Texas. He’s a former Democrat and a Hispanic with his own political base.

TX-34 — Vicente Gonzalez (D) vs. Eric Flores (R)

Right next door to TX-28 is another low-turnout, highly-Hispanic district that has swung Right of late. But don’t expect history to repeat itself in 2026.

Trump carried Rep. Vincente Gonzalez’s old district by 4 points in 2024, while Gonzalez won re-election by 2.6. Again, Republicans have re-drawn this to a Trump +10 constituency, but the assumptions underneath those numbers are highly suspicious, as we’ve already said.

Two Republican-aligned polls have Flores narrowly ahead, while the one independent university poll we’ve found has Gonzalez ahead 45–42. This would align with Gonzalez’s midterm margin in 2022 – albeit in the old district – where he bested Maya Flores by 8.5 points.

All of that is enough for me to think that the markets are right to put TX-34 into the Democratic column in the high 70s.

TX-23 — Brandon Herrera (R) vs. Katy Padilla Stout (D)

This was supposed to be an easy hold for former GOP Rep. Tony Gonzales, but then the world learned that he had been sexually aggressive with his female staffers, including one who committed suicide. Ugly stuff – so he resigned.

But Gonzales’s Republican replacement is bringing scandals of his own.

Brandon Herrera is a successful gun YouTuber with more than a decade of content for Democrats to mine. The oppo book already includes crude jokes about women, veteran suicide, Nazis and the Confederacy. Herrera has defended much of it as dark comedy, but it’s a lot of ammo to hand-over to your opponent, and he’ll be dodging bullets from now until election day.

This district hasn’t elected a Democrat since 2010 and is rated as R+7. It shares some dynamics with the other border districts but you shouldn’t sort it into the same bucket. TX-23 is geographically massive and pulls in voters from the exurbs of San Antonio and El Paso; it’s majority Hispanic but not by nearly the same margins; it does not suffer from low turnout; and Republican victories here have been on-trend since the Obama era.

That said, if you can still get shares in the high-20s or low-30s, it’s an attractive place to trade on a big wave for Democrats.

There is a freight train coming for Republican candidates this November. It’s clear that it will be running right through the Southern Border. The only question is who gets hit.

Get news for traders, by traders. Get Eventual.

Subscribe

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

The Polymarket Institutional Desk offers block trades on U.S. and international exchanges, institutional market maker relationships, and enterprise data services. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

is the President of SemiAnalysis, the Substack-native semiconductor research firm whose analysts hit dozens of technical conferences a year to map every layer of the AI buildout.

The Oracle sat down with O’Laughlin to break down Polymarket’s biggest AI market and discuss why the leading labs may be sitting on their best models.

This interview has been edited for length. All answers are his own.

Subscribe now

How does it make you feel when you see Twitter meme coin accounts suddenly rebranding as AI bottleneck guys? And how is SemiAnalysis different?

Let’s put it this way: they’re semiconductor bottleneck guys, but they couldn’t tell you what a semiconductor was two years ago. They don’t understand the difference between a CW and DW laser.

Me and Dylan were obsessed with this space when no one gave a shit about semiconductors, before the chip shortage. We came to the conclusion that it’s the most important technology ever invented, and that after the end of Moore’s law it would only become more important, because making it would become harder. That’s truly the entire bet of my career. In 2018, no one gave a shit about chips, and everyone was talking about software. Now the world’s totally opposite.

What’s our edge? We go to conferences, dude. 100 technical conferences a year. I’m not talking Dreamforce, where they’re playing Nickelback concerts. I’m talking an IEEE thing where they’re going through the specs of a new chip. We beat the concrete. If you really want to know the future, you’ve got to be deep in the sheets in a way that a casual person who isn’t devoting their life to it can’t get to.

Let’s pull up our big picture AI market: “AI bubble burst”. It resolves YES if three conditions are met from a list: Nvidia off 50% from its high, the semiconductor index down 40%, H100 rentals under $1 for five days, a major AI supplier collapsing, or OpenAI or Anthropic declaring bankruptcy.

I’m going to be honest, these are some garbage yeses. The bankruptcy one is pretty good. But a stock going down 40% from its all time high is kind of pathetic, honestly. That’s just saying the stock market goes down.

You need at least three of the qualifying events, though.

I know, but stocks went down about 30% on the deleveraging in August, and you can argue they had ripped to crazy highs first. Let’s do an intellectual experiment. In the next year, semiconductor stocks go up 10x, and then they go down 40% from the high. And you’re telling me the bubble burst? That’s a pretty shitty contract.

Semiconductors are cyclical, and they have a lot more volatility than the market. In 2022 they went down 45%. In 2025, during the tariff tantrum, they went down 40%. In 2020 they went down 35%, and in 2018, 20%. In 2008, 60%. In 2000, 67%. Honestly, 67% is a really good number. That’s usually a pretty strong bubble burst number.

There are only two stock-decline prongs, so a YES also needs something like the H100 rental condition or a major supplier collapse.

The H100 one is dog. People don’t appreciate depreciation. If you’re buying a 1990 car with a ten year useful life, odds are it doesn’t cost more than the scrap value of the car. At some point that happens to these chips. They physically break down after six or seven years.

The variable cost of running H100s is about 40 cents. But there’s a world where all the new chips coming online are better on a power performance basis, where you’d rather spend two bucks on a B200 than a dollar on an H100. These are depreciating assets. They should go toward zero as new technology comes out that’s way better than the last one. So the problem with the H100 rental prong is a lack of understanding about depreciation. Also, Silicon Data’s index is a little noisy, but that’s me personally.

And “major AI supplier collapse”: Super Micro going down 50% from its all time high? No doubt in my mind that can happen. It might have already happened, actually. Some of these suppliers, like Arista or even ASML, there’s a case where they go down 50% and the bubble doesn’t burst.

Share

Put it in percentages. How likely is it this market resolves YES without a bubble actually bursting?

Nvidia or the SOX goes down 40%, the H100 falls below a dollar for five days on pure depreciation, and one of the majors collapses. To be clear, that’s a soft market. It’s not a fucking strong market. But I wouldn’t call it a bubble burst. That would be a whimper, not a bang.

There is a case where the bubble doesn’t burst and three of these conditions are still met, and that’s like a 20 or 30% chance. It’s more likely that all of these things get met during the quote unquote bubble actually bursting. But you can hit these tail corners, especially with three distinct events, where the spirit of the contract never actually happened.

How would you write contract rules to truly capture an AI bubble collapsing? Like, it’s truly over.

I would probably do five years. I would do OpenAI or Anthropic goes bankrupt. And I would choose something like 67% off the all time high, with no reclaiming of all time highs within five or ten years. That’s a much more painful drawdown. 40% for the SOX is nothing.

Say the AI bear case comes true. AI stays a cool toy, enterprises never adopt it, and we realize we overbuilt like the railroads. Is that how this ends?

I was writing about railroads before it was cool, and people do not appreciate that in the railroad bubble there were three distinct build out phases: one around 1867 that boomed and busted, another around 1877, and pure financial speculation in the 1890s. There’s a chance, maybe more likely than not, that this build out has ramps and drops along the way. There’s a digestion period for a year and a half or two years, it feels like it’s all over, and then a new leg starts working again.

We’re also doing a lot more real shit than ordering a pizza online in 2000. And if the technology is widely diffused, it doesn’t happen in one big boom and bust. It’s a rolling digestion.

Data center moratoriums went from fringe idea to front page this year, and New York enacted a statewide pause in June. Are the moratorium markets priced right, and what would a broad pause do to the cycle?

I don’t have massive conviction on the pricing, but someone on my team tracks this intensely. There is a clear shift: Democrats are running against data centers, Republicans are running for them, and running against is clearly the more popular position. The last election was a CPI referendum, and the data center issue, through energy prices, is a subset of that. There’s a chance of a press release victory: someone campaigns on it, wins, starts the moratorium, and then the powers that be, meaning capitalism, create loopholes, and the original spirit of a hard pause fizzles out.

Here’s the ironic part: I’d argue a moratorium means the AI bubble doesn’t pop. There’s a demand line for AI that gets bigger every year, and a supply line going up every year, and we’re trying to figure out when supply overshoots demand. That’s when the glut begins. If you impose a giant data center moratorium, supply slows down massively, and I don’t think the demand stops. It elongates the cycle. Chip stocks take a breather, but they don’t go down 60 or 70%, because they’re still making money. Meanwhile AI gets better, enterprises keep adopting it, someone makes a new wonder drug, and all of a sudden there’s way more demand than supply again. It self corrects.

A bubble is wild, speculative inflation and then a quick collapse. Electricity took 40 years and inflated the entire time, and that made it not, quote unquote, a bubble. The analogy is you’re at a party, and leverage is your drinks. Two or three drinks and home early, the next day is okay. Twenty beers with the guys until 4 a.m., the next day is not okay. If the government says, hey guys, no more beers, it’s midnight, and you’ve had ten, the next day you’re hungover, but you’re not in the hospital. If politics slows this down, it’s healthier in the longer run.

Share The Oracle by Polymarket

Who’s winning among the big labs?

The phrase the Anthropic CEO likes to use is Cournot competition: multiple suppliers with a relatively fungible product, competing on the supply capacity they build. From that perspective, Anthropic is being rational, and OpenAI is super defensive, trying to catch back up. Their new model is a little smaller, probably cheaper to inference, and they’re more than willing to subsidize it. If Sol and Fable are equivalent models and one is half the price, it makes sense that you’d swap over. It would have been unthinkable last year that OpenAI would not be in the lead, and now they’re not.

[Note: this interview was recorded before the release of OpenAI’s Astra model]

Are we reaching the point where the best models aren’t releasable because they’re too dangerous?

I think that’s going to be one of the big narratives for the rest of the year. If the guys in the lead are not playing as hard as they can, how is that bullish at all? Maybe they have a better model, but they’re not going to release it. The government wants to review releases for safety reasons. And on the China versus US side, if you put out a new model, people try to distill it day one, so you give them a version you think is better for anti distillation. They’re willing to hold their punches because these guys aren’t suffering. Anthropic is essentially break even right now. People are having a lot of drama around the ARR deceleration, but part of that is because they’re not releasing new models. If you’re growing at a tremendous rate, break even, and in the lead, why press when you can train a better model and always have something in your back pocket?

Apple sat out the model race entirely. Smart?

It’s going to hurt them in the very long run. The biggest winner of the last cycle often misses the trends of the next, and that’s Apple. Once upon a time it was just IBM, and we don’t talk about IBM like we used to. If you just don’t play, you lose. I believe it’s more likely than not that Apple becomes less relevant. Apple is a much more financialized company than people appreciate. They buy back 100% of their free cash flow. It’s an equity bond. They make a crap ton of cash every year, use it to buy back shares, and have nothing left over to invent new things.

Some people say the 10 year Treasury rising is the real AGI indicator. Do you buy that? Are you a paperclip guy?

I’m not a paperclip guy, but half of my company is. What it’s clearly showing is a demand for capital that is real, and it is moving the big number, dude. The Fed doesn’t control the long end of the curve, and the 10 year and 30 year moves are in real terms, not nominal. Someone has massive conviction they can borrow capital and make a higher return than in the past, and the someone is the hyperscalers. I’ve been calling them gigascalers, because they’re the third largest issuer of bonds in the entire world: the United States, China, gigascalers. You’re watching capitalism’s speed limit say, if you’re going to do that, you’re going to have to pay more. And the issuers are saying, yeah, we think we’ll see the return.

What AI polymarkets do you want to see next?

AI for healthcare is going to be a big one. Just more “will AI cure a cancer by X year.” That would be cool, because I think it might.

Follow Doug’s work at SemiAnalysis on Substack, and on X @semianalysis_

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Historian and geopolitical analyst Walter Russell Mead is a distinguished fellow at the Hudson Institute, the Wall Street Journal’s Global View columnist, and the author of Special Providence: American Foreign Policy and How It Changed the World. His “Jacksonian America” thesis predicted the rise of Trump-style populism fifteen years early.

He caught up with The Oracle to discuss Ukraine endgame scenarios, the key drivers of Trump’s Iran strategy, and why Putin is watching the French polls.

The following interview has been edited for length. All answers are his own.

Subscribe now

How much should someone in the world of geopolitics be trying to make forecasts?

Especially in wartime, if you go back and look at the history of punditry, what you’ll find is that people are almost always overwhelmingly wrong at every moment of a war. In the U.S. Civil War, both sides at the beginning think the Battle of Bull Run is going to be it, that it’s going to be 90 days. Half of Washington is out there with picnic baskets to watch. Then it turns out both sides are wrong. Think of the Russia-Ukraine war, where momentum has swung repeatedly from one side to the other, or the Iran war, the same thing.

In war, people really want to know what’s going to happen next. But in fact, in war, uncertainty typically rises, and so you have that emotional mismatch between a desperate need to know what’s coming next and the extreme difficulty of actually being able to predict.

War pushes people out to the frontier of human possibility. It’s a great accelerator of progress, because you’re doing everything you can think of doing, marshaling all the forces you know how to pull together, to achieve an objective against an enemy who is doing the same thing. Both of you are changing as the war goes on. Both sides’ capabilities change in different ways. They might realize: I thought taking your capital was the way to win the war. I’ve taken your capital, and I haven’t won the war.

I was reading through your recent op-eds, and one of your phrases stuck with me: the “slop of war.” Is that a pithy way to encapsulate this idea, that in a war you have two sides doing their damnedest to change the situation and you really don’t know what’s happening?

For me, it’s a way of referring to the immense amount of intermixed information and pseudo-information about the course of a conflict that erupts. The sheer volume of punditry about a war, the sheer volume of coverage, the statements by both sides, by official governments, then the statements by insiders who claim they know what Trump is really thinking or what the Iranians are really up to. This whole immense mass of what calls itself information actually makes it harder to see what’s going on.

In the old days, gunpowder made so much smoke that even a few minutes into a big battle, nobody could see what was happening anymore. The fog of war is a metaphor for that kind of uncertainty over the course of a conflict. But I think now we’ve reached a new level of misinformation, disinformation, earnest but often erroneous information, and somewhere in there maybe some nuggets of truth, all swirling together in a way that makes it very difficult to say anything clear.

To the extent that you can say anything meaningful about what’s happening, it’s helpful to look at the structural reality. For example, one thing that has led a lot of people to misjudge, and let’s not even talk about the ultimate outcome of the Iran war, which we don’t yet know. Maybe somebody on Polymarket does. I don’t. But what people misunderstood was how committed Trump was to it, and how do you understand that about him? People were not able to look at the war through Trump’s eyes and understand what he thinks, what the situation looks like to him.

So how does Trump think about the Iran war?

Going in, few people noticed that Trump has been consistently very hawkish on Iran, going back really to the Iranian Revolution in 1979. And Trump has a history of sticking to things. When Trump has believed something for a long time, like protectionism is good, trade deficits are bad, he keeps coming back to these points through all kinds of circumstances. So on Iran, people underestimated the degree to which that is one of the core things he has had in his mind for a very long time: that Iran are bad guys and need to be dealt with.

Then what was his calculation going into the war? As far as I can tell, he seems to have thought, first: who knows, I might get lucky and win. So the initial decapitation strikes and so on. It might get me a Venezuela, and what fun that would be. I’d have done something no one else could do, and parades, and I love that idea.

But if that didn’t work, okay, then we get into a very chaotic situation, and Trump actually thrives in chaos. So when the original attacks don’t lead to regime change, a lot of observers concluded: oh boy, Trump’s in trouble now, it’s chaos in the Gulf. No. He thinks he has an advantage and he’ll be able to figure something out. Who knows what it’ll be? He doesn’t know what it will be, whether there’s a rabbit to pull out of a hat or some development. He just thinks he’s better at getting through this kind of situation than most people are. That’s number one.

Number two, he thinks, worst case, he’s really good at spinning things. He can hold on to at least a critical mass of his base so that he doesn’t get convicted in an impeachment trial or something like that. So his assessment of the downsides was different from the assessment a lot of people thought he was bringing to it.

Then, what’s his problem in this chaotic zone? There are two things you want to avoid if you’re him. One is sending a lot of ground troops into a long, complex campaign, and any ground troops if you can. The other is the kind of global economic crisis that would put irresistible pressure on you to fold. So he’s been navigating between these two things, going between Scylla and Charybdis. A lot of the craziness people attributed to him at various points of the war, the TACO motif and other things, are really about: no, he’s trying to prosecute the war without causing some kind of economic crisis. Hey, I slow down, I take sanctions off for a while, I’m still readying myself for more economic pressure.

And I think you can add a third that’s negative: he doesn’t want an agreement that looks anything like the JCPOA that Barack Obama had. So there are three negatives for Trump that are defining his course.

On one hand I’m hearing you say that Trump is very improvisational, but also that confronting Iran is one of the core things he’s believed for decades. Is there a contradiction there?

I don’t think it’s either-or. He’s playing several games here. One of them is that he is trying to hold power as long as he can, and he wants to control the Republican succession. What he does not want is some kind of open primary in the Republican field where Trump’s endorsement is one of many factors but not the decisive one. He wants to be able to say: it’s you that’s going to be the nominee. And he doesn’t want either Vance or Rubio to take it for granted. I don’t think it’s impossible he picks somebody at the end who’s neither Vance nor Rubio, as a way of expressing dominance and retaining power. You’ll notice some of the candidates he has supported in Senate races and other things, like Lindsey Graham’s sister, are people who wouldn’t be there without him. So there’s that game.

He also doesn’t want the lame-duck clock to start ticking. To do that, you want them dancing around you, desperate for your endorsement, while you keep your hand completely free. I think he’s got that now.

On Iran, do you think we could see another tactical pause? A few weeks?

First of all, two weeks is a long time in this war. That would make me cautious about putting down a yes vote for anything, certainly for any date. I don’t know if you’ve heard the line they teach you in pundit school: it’s always good to give a date, it’s always good to give a number, but it’s never good to give a date and a number. Something smart pundits try to remember.

Then you have to think, what causes the U.S. to either strike or not strike? What is Trump’s desired state? It is that he is adding economic pressure onto Iran without Iran being able to screw the rest of the world. The Strait is largely closed to Iranian oil and largely open to other oil. That’s what Trump would really like to see here. And if Iran is pushing Trump too far away from that goal, he becomes more likely to take military action to restore the balance he wants. But on the other hand, if Trump is getting too close to his goal, then Iran is likely to take some kind of action to try to change the military direction, in a way that Trump would then likely have to answer with some kind of military strike.

I want to pick up on your “pundit school” comment. If we woke up tomorrow and every pundit had a Polymarket score next to their byline showing how correct they’ve been on predictions, would that be a good thing, a bad thing, or just noise?

To some degree, if you believe as I do that the future is something people don’t fully understand and can’t fully predict, then it’s not a mark of wisdom in a pundit to lay down a lot of markers. There might be exceptional cases or moments where you would be able to say something definite. But in general, I think you can help people think. You can do, in a sense, what we’re trying to do here, which is helping people who must bet on the outcome to think more clearly about what’s going on.

But say you’re an oil company that needs to decide whether to send ships through the Strait of Hormuz. Isn’t it better to have a reasonably accurate probability of one getting sunk than someone telling you that the future is unknowable?

Those of us who are oil companies, or an insurance company, have one set of needs, and often what they need is not information. Real information is nice, but what they need is a CYA method, so that even if I’m wrong, I followed the right procedures, and so I did nothing wrong, even though, alas, the ship is at the bottom of the sea. That is very often the corporate and governmental decision-making area people are actually working in, and the incentives they’re actually working with. So any kind of data, whether good or bad, as long as other people think it’s kind of good, is useful to decision makers. Not necessarily in terms of making better decisions, but in being able to stay in the jobs where they get to make these decisions and get compensated for it.

Let’s pull up the Republican 2028 nominee odds. It sounds like you are anticipating more volatility, rather than a heavy favorite continuing to dominate?

You have to ask what Trump wants in his successor, and I think he wants three things. Number one, he wants it to clearly be his choice. Number two, he wants his successor to win. I don’t think he’s going to deliberately pick a loser to spite people, unless, well, I could see him doing that in certain circumstances, if he thinks it’s all but done and he wants to take the Republican Party down with him or something. I could imagine it. But ultimately, he wants his successor to promise that he and his family will be protected from investigations and retribution after he leaves the White House.

Are there any names that spring to mind from the lower tier who could be real contenders?

He would want somebody who maybe doesn’t have much of a political base of their own, who couldn’t raise money for a national race without a lot of help from him, but somebody who in some way would have some kind of appeal. Would he look for a minority candidate of some kind? In Trump’s mind, probably the greatest Republican convention ever would be one where nobody knows who it’s going to be, and Trump does a surprise announcement that is really surprising on the day of the convention.

Let’s go to the Ukraine war. How do you think about where things stand now?

At the start, I did not think Ukraine would crumble in a hurry from the sudden attack. I felt there was real resistance there. Interestingly, geographically and historically, the most nationalist parts of Ukraine are in the farther west, around Lviv, and those would be the hardest places for Russia to conquer. So knocking Ukraine out in one blow, I thought, was less likely than people were anticipating. I also thought, and I still think, that if the West really got its act together, it could inflict a lot of pain on Russia and provide a lot more support to Ukraine than we have.

The trick is that this has been, since very early, a war of attrition, where both sides are taking damage, and Russia is the bigger country with the bigger economy and the bigger population. In a sheer, ordinary war of attrition between two countries isolated alone on a planet, Russia probably ultimately wins some grinding, horrible victory. But Ukraine’s allies and potential allies have a lot more to offer in terms of support than Russia’s allies do. I would say so far the West has done a pretty lousy job of providing support for Ukraine over a long-term war.

At the same time, Russia has been making significant progress in getting more of the capability of its allies on board to help it fight the war, getting more help out of China and North Korea as the war goes on. But on the other hand, support from China is a kind of poisoned chalice for Russia, which really doesn’t want to be dominated either by the West or by China. Even if Putin wins in Ukraine, if the cost is that Russia has become so indebted and so dependent on China, and has meanwhile lost so much of its position in places like Central Asia to China, Russia could win the war in Ukraine and lose its position as a great power, and be sort of quasi-eaten by China. So there are several different ways Russia could lose the war. It’s harder to see a path to a real victory in a slow war for Russia.

What headlines would make you revise and think, okay, now we’re headed toward a resolution to this war, one way or another? And conversely, what would lead you to say the war of attrition is grinding against Ukraine in a way that means they lose?

The biggest sign of impending peace would be headlines that say “coup in Moscow,” or Putin overthrown. The way for the war to end is either for Putin to realize it’s costing so much for Russia that he has to get out of it, or for a critical mass of Russians other than Putin to reach that conclusion and act. That would tell you the logjam had broken.

On the other side, if you saw the Russians accelerating their territorial gains. Right now, when you look at the rate of Russian advance in the Donbas, it looks like it would be years or even decades before they could actually achieve the goals Putin is currently setting forth. So Russian military breakthroughs on different parts of the front, even if they aren’t so deep and enormous that Ukraine looks about to be conquered, but where Ukrainian forces are no longer able to hold a stalemate, that would be a sign something significant had changed.

I imagine when Putin is looking at whether he can continue, whether he still has a hope of winning, a big sign for him is the polls showing Marine Le Pen is likely to win the presidential election in France next spring, where a real splinter in the Western alliance could take place, with Merz and Macron both getting weaker politically by the day. The heart-of-Europe support is not great. So I would look at those as indicators.

You mentioned the midterms. If we assume the Democrats win the House and the Senate is a coin flip, how do you see the latter half of Trump’s term playing out if the midterms go the way most people think?

I would just say, remember that we are in the Trump era, and a week of Trump time is like a month of regular time. So the midterm is still farther in the future than the calendar might tell you. But okay, set that aside.

If he loses both houses, it’s more of a problem. My sense, unless the Democrats get really smart this time: the House of Representatives is a terrible body. Of all the elements of the American government, it has the longest track record of losing power and coherence. You get these Congressional investigations, and even when they’ve got a really good case, they often fall short, partly because every Congressperson gets in the lights, and this is their moment, and they just have to grandstand. They’re not thinking, how do we craft an investigation, how do we really build a case? They’re thinking: how do I get my five minutes? So House investigations usually fail to move the needle on anything.

Senate investigations can be more effective. Not always, but they can be, and the Senate still has greater gravitas than the House. So getting investigated by the Senate is a more serious problem for Trump. There’s also the problem that he won’t be able to confirm officials in many cases, which is going to change the balance of power inside the Trump administration a little bit.

If the Democrats win the House, do you think it’s a lock he gets impeached a third time?

And maybe a fourth or a fifth.

In a scenario where Trump loses both houses of Congress, will there be any meaningful war powers attempts to curtail his freedom of action in Iran?

There will be efforts, but I’m not sure how successful they’ll be. It also depends on how big the Democratic majority is in the House and in the Senate. How does Fetterman vote, and a couple of other senators, on some of these things? You often lose a couple on contentious votes. Trump often loses a couple of Republicans on difficult votes. So that gets very tricky.

I also think the greatest thing, in a way, that could happen to Trump would be for Iran to launch a level of attack that would allow him to get a declaration of war through Congress before the end of this Congress, that is, before the new Congress is seated in January. Because with a declaration of war, his powers expand dramatically, and a lot of the legal constraints that chafe him would disappear. I don’t know whether the Iranians understand that this might not be the smartest thing they could do.

But as long as we’re talking about wild cards and things that could really change the mix: suppose Trump gets a declaration of war against Iran. That makes the war powers stuff completely off the table. I mean, Abraham Lincoln had hundreds of newspaper editors arrested because of what was in their newspapers. The precedents are amazing. Woodrow Wilson put people in jail for sending anti-war messages through the mail. So that could change a lot of things. Trump can’t just pull that rabbit out of a hat, but I have to think that if the opportunity arose, it would not break his heart.

Are you following the Israeli election closely?

Not too closely, if only because it’s all going to depend on maneuverings. And look, right now the odds look heavily against Bibi, but Bibi is very good at pulling miraculous victories out of difficult situations. The last few elections before October 7 were deadlocked squeakers. Bibi, in spite of having more enemies in Israel and outside of Israel than any Israeli prime minister, is also the longest-serving prime minister in the history of Israel, and that is a very, very difficult job to get and to hold. So I’m waiting to see.

Then there are the questions of the Arab votes. Will the opposition to Bibi be content with depending on the Arab votes? How does the prospect that an Israeli government depends on Arab support affect the way voters behave at the time of the election? There are a million things going on here. Will Trump decide that he wants to give Bibi some kind of a present before the election? Or will events in the Gulf lead to a more muscular American presence, attacks in the Gulf, in ways that actually benefit Bibi at the polls?

When you say give Bibi a present, what are you thinking of?

Suppose the U.S. decides to really go in and try to get the nuclear stuff, or whatever it is, in Iran. Or simply resumes fighting, or gives a statement of support for Israel against Hamas and against Hezbollah. There are lots of ways, some of them fairly inexpensive, that Trump could make Bibi’s situation stronger in the run-up to the election. But will he do it? And if he does, how much will it move the needle? I can’t tell you.

Personally, I’m surprised that traders are quite optimistic that Israel and Iran will not have a direct exchange of fire through the end of the year. They see that as quite likely. Do you agree?

That’s actually not how I look at it. What Israel is doing now is advancing against Hezbollah and managing the Gaza situation. And it’s Iran that isn’t attacking Israel, more than Israel that isn’t attacking Iran, if you see what I mean. The balance of forces in the Gulf is such that, so far, Iran has stood by fairly passively as Israel has significantly tightened the screws on Hezbollah. And this probably has some kind of factor in U.S. thinking about what’s going on in the Gulf and the relative strength of Iran: at least at the moment, Iran feels it doesn’t have the ability to try to protect Hezbollah by attacking Israel. So why would Israel feel a need to change that status quo by attacking Iran?

Modern prediction markets grew out of DARPA forecasting experiments. Are you bullish on them?

Betting markets have a history of thousands of years, and predictions and so on. It’s just something people have been doing going back to the dawn of recorded history, and probably well before then. I bet you Thor hits the mammoth before Ulrich does, or whatever it is. It’s human nature. Human intelligence to some degree develops around what’s going to happen next. We’re really focused on that: how do I keep the leopard from eating my kids, how do I find food? We’re focused on making predictions about the future. But it remains the case that we’re not necessarily that good at predicting the future. That’s not going to make us stop trying, though, which is probably a bullish sign for betting markets.

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

In many ways, the 2026 NFL season is defined by the one that preceded it: it’s the Year After the Weird Year.

2025 was marked by such unpredictability — the three-time defending AFC champion Kansas City Chiefs missed the playoffs entirely, while the two actual Super Bowl teams (Seattle and New England) were both below 2% to win before the season — that this year will revolve as much as anything else around which of those surprises represented genuine change, and which will simply regress away toward the established NFL order we thought we knew.

Take Wednesday’s Seahawks-Patriots season opener. The Seahawks won 13-10, but neither team went into the game any higher than 8% to win the championship, with Seattle checking in as a distant No. 2 in the NFC title odds even before QB Sam Darnold was injured, and New England arriving tied for sixth in the preseason AFC odds even before WR A.J. Brown was injured. (Yes, the game was marked by injuries as much as performances on the field.) That puts the pressure on both Big Game teams from a year ago to prove they can do it all over again — and they’re not the only ones with a lot to prove.

Even with a potential “superteam” in the mix — for all the good it did them on Thursday — traders are staring at a title race with far more questions than answers. So as the 2026 season gets underway, here’s what the markets know — or just think they know — about the NFL going into the year, starting with maybe the only team being asked to live up to expectations rather than defying them.

Are the Rams really a superteam?

The closest to a sure thing in this NFL season is generally considered to be the Los Angeles Rams, who were (at worst) the second-best team in the league last year behind Seattle and spent the offseason loading up on big names like edge-rusher Myles Garrett and cornerback Trent McDuffie — plus luring their own future Hall of Fame lineman, Aaron Donald, out of retirement. Because they probably had the most uncomplicated story to tell, the Rams spent the summer rising above the fray to establish themselves as title favorites with nearly double the preseason odds of any other team:

Certainly on paper, L.A. probably has the most talent of anybody — both ESPN and Pro Football Focus ranked their lineups as the best in football, and they have stars at just about every position. Amidst a sea of question-marks, the Rams’ pitch is basically, “don’t overthink this.” And that may be right.

The downside of that theory, however, was on display in the Rams’ opener against San Francisco in Melbourne Thursday night, where they came out flat and could neither move the ball nor stop the Niners’ balanced attack. Many will write it off due to the historic travel distances involved in getting to Australia, and L.A.’s now-questionable decision to arrive Down Under just in time for the game (when the opponent had been there for weeks). There are even people making the case for this being some kind of 4D chess to help the Rams weeks down the line.

But maybe L.A. also just isn’t quite as far ahead of the pack as everyone was hyping them up to be.

And besides, no NFL team is really a superteam as we typically understand it from other sports. A team like the Rams, at 17% in preseason, would be third in the NBA title race, and would have trailed the star-studded crosstown Dodgers in MLB every single day of the ‘26 baseball season. That’s more of an apples-to-oranges comparison than anything scientific, but it illustrates how tenuous the superteam concept is in football: Even the closest thing the sport has to one of those is still an 80%+ underdog against the field.

Which version shows up?

Aside from L.A., most of the other leading contenders are basically staking their case on one particular snapshot in time being more salient than another.

In the “2025 was real” camp, we have teams like the Seahawks, Patriots, Broncos, Texans, Bears, Chargers and Jaguars — those who took leaps or otherwise defied expectations to be among the final group of teams standing.

Not all of these teams are equal in terms of regression red-flags looking ahead to 2026: the Bears, for instance, have more factors suggesting a backslide — a sky-high turnover margin last year (which tends to fall back to earth quickly), a mismatch between actual and expected (based on point differential) records, a big uptick in schedule difficulty this year — than, say, the Texans. But all were 2025 playoff squads who met one or multiple of the following qualifications for concern:

• Major year-over-year improvement in record (which tends to be given back some the next year).

• Heavy reliance on the defensive side of the ball (which regresses to the mean between seasons more strongly than offenses).

• Mismatches between actual record and Pythagorean Expectation (i.e., the record a team “should” have had based on its scoring margin, with neutral luck in close games).

• Outlier turnover margins (which have very little year-to-year correlation, despite strongly influencing NFL team success).

• Above-average health, in terms of man-games lost (unusually injury-free teams tend to not stay that way for very long).

• Changes from facing notably easy schedules in 2025 to notably hard ones in 2026 (which obviously makes sustained winning more difficult).

The more of these factors that a team runs afoul of, the more likely they are to suffer a rude awakening the following season — but so many ‘25 contenders fit those profiles that sorting out who takes a tumble will be more difficult than usual.

Meanwhile, we have the teams who’ll argue that 2025 was an outlier and their previous form was more indicative of true quality: the Chiefs, Lions, Ravens, Bengals, 49ers, Cowboys, Eagles and Packers.

The magnitude of underperformance, and reasons for it, vary in each case. The Chiefs saw all of their luck from previous years flip around on them with a vengeance — and then QB Patrick Mahomes was lost with an ACL injury, for good measure — while others such as the Lions are dependent more on full-roster health improvements. Others still, like the Cowboys, Bengals and Packers, are defined by the potential for bad defenses to regress while maintaining strong offenses:

Taken altogether, the 2026 Super Bowl race may depend on who can either stave off regression — or harness it in their favor — than the usual offseason improvements via comings, goings and other transactions.

Will Buffalo earn its wings?

As you might have noticed in the chart above, the Bills are an oddball even compared with their fellow contenders. They did not trigger any of our multi-colored flags — the closest they came was a change from second-easiest in schedule strength to 14th-hardest — and on the surface, they had a similar 2025 (going 12-5) to each of the previous handful of seasons. (Buffalo has sandwiched itself between 10 and 13 wins for seven consecutive seasons, an impressive run of consistency.)

But the Bills nonetheless deserve highlighting because of what hasn’t happened for them yet, despite all of their regular-season winning: a Super Bowl berth.

Buffalo has made the playoffs seven straight years dating back to 2019, winning eight playoff games in that span — third-best in the NFL behind Kansas City (16) and San Francisco (9), both of whom made multiple Super Bowls during that period. But the Bills have only made two AFC title games, losing both to K.C., while bowing out in the Divisional Round four of the past five years.

QB Josh Allen remains one of the league’s best QBs — he’s finished Top-5 in MVP voting four straight years — and the Bills’ offense has ranked Top-5 in Expected Points Added (EPA) per play for six years in a row, helping make Buffalo one of the most consistent contenders in the NFL recently. Traders clearly think they remain an upper-echelon team after swapping coaches from Sean McDermott to Joe Brady (just look at our leaguewide odds chart from above).

In other words, there’s no good reason why the Bills haven’t been able to get their breakthrough yet… or, more importantly, why this wouldn’t be the year it finally comes together. But until we actually see it happen, Buffalo’s ceiling remains unproven — and with Allen heading into his age-30 season, time is starting to run out on the Bills’ chance to capitalize on this current window.

Divisions of labor

Not all paths to the playoffs are created equal. Some teams are strong enough to survive whether they win or lose their division; others are essentially competing for one available route, whether through winning a weaker division, or having the bailout option of the wild card from a stronger division.

To measure which is which, we took the preseason Polymarket playoff odds and decomposed them into what share comes from a team’s chance to win its division versus make the wild-card. The teams with the biggest relative gaps between those shares — compared with the average team — are the teams most dependent on either winning their division or making the wild-card to punch a postseason ticket:

By this measure, the Bills are the team with the most lopsided split between their divisional-path share (57%) and wild-card share (19%). (The remaining share, 25%, is the probability Buffalo misses the playoffs outright.) In other words, if Allen and company are going to make the playoffs, they’re most likely going to do it as AFC East champs, even though they claimed the wild card last season. Others in that category include the Tampa Bay Buccaneers (NFC South), Houston Texans (AFC South), New Orleans Saints (NFC South) and Baltimore Ravens (AFC North).

On the other side of things, the 49ers went into Week 1 as the team most dependent on a wild-card, which speaks to the strength of the NFC West — which accommodated the Seahawks, Rams and Niners a year ago and could easily do so again in 2026. (Especially considering how dominant San Fran looked in Australia.) Beyond the Niners, the Pittsburgh Steelers (AFC North), Tennessee Titans (AFC South), Cleveland Browns (AFC North) and Las Vegas Raiders (AFC West) are also most disproportionately reliant on a wild-card if they are to make the postseason.

Which path is better? Obviously it’s better for your seeding to win a division — but teams like San Francisco also have a good luxury in not being completely division-or-bust.

MVP = AFC QB Royal Rumble

Much like when we looked at the World Cup Golden Boot market in the context of each team’s odds to go deep in the tournament — more games equals more chances to score — there is value in thinking about the conditional probabilities around the MVP chase odds.

Going into the season, our favorites are almost all AFC quarterbacks: Jackson, Justin Herbert, Allen and Joe Burrow, with Maye and Mahomes not far behind.

The only NFC QBs to crack the Top 7 are Dak Prescott and Caleb Williams — not the defending MVP, Matthew Stafford, despite his Rams being the “superteam” we discussed earlier. What gives? Some of that might be banking on voter fatigue: Aaron Rodgers in 2020-21 was the only back-to-back MVP winner since the 2010s (and only the fifth player to do that ever). Some might also just be about the difficulty spreading credit on a team with so much talent — but there is a real mismatch between team quality and MVP odds in Stafford’s case.

The same might have been said for Sam Darnold, whose Seahawks were rivals and foils for Stafford’s Rams all year last season, but his injury Wednesday might make the point moot. Williams is the opposite — he will likely be a strong candidate if the Bears make the playoffs, but traders are not at all sure that will happen (46% playoff odds). In terms of combining playoff certainty and MVP potential, it’s hard to do better than Jackson and/or Allen, adding another dimension to the longstanding rivalry between the two members of the 2018 NFL Draft class.

And what of non-QBs? Each of the Top 18 names on the odds sheet are QBs, and no non-QB has even 1% odds to win the award. (That’s because only one non-QB — RB Adrian Peterson in 2012 — has claimed the award since Shaun Alexander and LaDainian Tomlinson won it back-to-back in 2005-06.) But maybe this is the year that all changes — even if the odds are far higher that a QB from the AFC just walks away with the hardware.

And if the first two games are any indication, the NFL is going to give us answers to all of our preseason questions — and then some — faster than we can even process them. We’ve already seen the defending champs survive an ugly, injury-filled opener, plus the presumed superteam get flattened halfway around the world. In an NFL season built around figuring out which version of reality we should actually believe, that all feels like a pretty fitting start.

On September 3, Polymarket launched Perps, its perpetual futures exchange. Perps let traders go long or short individual stocks, indices, crypto assets and commodities with up to 20x leverage and no expiration date from the same balance they use for their polymarket.com prediction trades.

After the launch, we caught up with Xtine Fang, Polymarket Perps Lead, to talk what’s next for the perps product and do a little whale watching.

This interview has been edited for length. All answers are her own.

Share

Why does Polymarket need perps? What’s the open space in the market?

The idea is to have a single house of all finance. We have this killer feature where you can trade stock and crypto perps because we began as a prediction market. We can offer a place for users to trade everything in one place, without having to withdraw funds to another exchange and go through a lot of transaction fees.

The other big differentiator is speed of listing. We listen to what users want, and we list very quickly. Previously it was semiconductor names, then the photonics names, a lot of crypto beta names, as well as crypto stocks like Circle. We’re very much catering to what the user wants. We also have Asia equities like CXMT, which is extremely topical, as well as stuff like Alibaba.

What’s being traded the most so far?

The top ones are still the crypto majors like BTC and ETH, especially over the past week where we saw the market rebounding. But previously, when BTC was stuck in the $62k-66k range, we saw a lot more volatility in the Korea semi names like SK Hynix and Samsung, and other semi names like Micron.

Who are the top perps traders right now? What do we know about how they are trading?

Since it’s a DeFi exchange, all the trading data is open to the public, and there’s a leaderboard where you can see all the top traders.

We have one guy, trader2121 who’s up almost $500k. I tried to email him and get in contact but he never replies. Maybe he’s scared that I’m a CIA secret agent.

The top perps wallet, ShadowPixel47 looks like a large market maker firm, with dozens of trades per day. But they have also had a large Long ETH / short BTC position that has performed very well since late August.

Anything you see the best traders do differently?

The smartest traders generally trade event driven stuff very well: betting into FOMC, positioning into earnings like SanDisk, taking leverage off after the huge semi rally, and then shorting at the right time. I can probably pull up some data for you.

One tool that could be useful for this is Polymarket Calendar. They made a calendar, and for each day it shows which events you can trade. We also have an earnings calendar for public companies on Polymarket.

Subscribe now

Are you noticing people do perps trades along with a corresponding Polymarket event contract? What’s the crossover in user activity between perps traders and OG Polymarket traders?

That is the ultimate end goal, successful cross selling to our Polymarket users. I haven’t seen it yet because perps is so new. But that’s the end goal, and it’s something a lot of our smarter traders have brought up.

Anything surprising you about how traders are using the product?

The more surprising thing was I thought people were no longer interested in trading crypto, but there were still a lot of people asking me to list UNI, AAVE, and Morpho, all those names, even when crypto wasn’t doing very well. You have this diehard crypto native segment that trades almost solely crypto. Even when we’re in a roaring bull market for AI and semi stocks, they’re not trading that, because they think that’s not their edge.

What’s the max leverage you can get? How is it set for the different assets?

It’s different based on each contract. On BTC the slider goes to 20x. Then we go to stocks: Micron’s leverage is 10x. Generally the RWA stuff is around 10x and the crypto offerings are around 20x.

How do the trading fees work with perps?

Our fees are very friendly right now. We’re so new and we want to be competitive, so our fees are better than Hyperliquid, maker and taker.

The way it works: every fill has a maker and a taker. If you place a limit order that rests in the book, you’re the maker, you’re adding liquidity. If you hit an order that’s already sitting there, you’re the taker, you’re removing it. Takers pay more because they consume liquidity.

The fee is charged on the notional of the trade, price times quantity, and it’s paid in pUSD. Everyone starts at 0.04% taker and 0.0125% maker, so a $10,000 taker fill costs you four dollars. From there it’s tiered by your trailing 30-day volume, and tiers get re-evaluated every day. The rates keep dropping as you trade more, and at the very top tier, $1 billion in volume, makers actually get a rebate. We pay you half a basis point to quote. Accounts from the beta are also on the top-tier schedule for now, regardless of volume.

We had the public launch last week. How has the volume been?

I’ll send you some charts.

I know there’s been a lot of requests for leverage on things like “Democrats to win the House” and the traditional events contracts. Have you thought about that? Is that something perps could do in the future?

The genuine answer is I’m not sure, because that’s more on the prediction market side. For the perps side that I’m in charge of, we want to be more like the traditional perp exchanges, where we list the traditional assets and the cryptos.

A unique offering we do have is cross margin, where a single balance collateralizes positions across multiple markets. A lot of the assets that are isolated margin on Hyperliquid, we have cross. Take CXMT, the recently listed ChangXin Memory, a very popular Chinese semi name that IPO’d recently. Other venues would only offer something like that as isolated. We have it cross.

On any market, you pick a market or limit order, choose isolated or cross, and then there’s the leverage slider: how much money you have on your balance, if you want to full port that at 10x and slide it out.

What are you seeing generally in the market? Is crypto winter over?

Let me put it this way. It’s currently still a very fast money driven bounce, and we’re squeezing pretty hard to make it more sustainable. Let’s see if we can keep BTC in the $80s for a while, and if Trump doesn’t say anything silly. Before Polymarket I was at a crypto liquid fund, so I know a lot of funds were literally shorting in the hole. They were shorting BTC at the $62-$67k levels, and they were really caught off guard by this rally. A lot of short covering is sending us higher, especially on the BTC and ETH front. And I think interest has generally come back to crypto. There were a couple of centralized exchanges, tier three exchanges, that were going to shut down in the past month because there was no trading volume. Now at least one, BitMart, wants to resume operations.

What about interest rates? Shouldn’t we have Treasuries, the 10 year, the 30 year? That’s been such a macro narrative.

We could. But I like to think of our current user base as younger degenerates who trade based on following FinTwit, or Reddit, following Citrini, that kind of thing. Interest rate products are another world that doesn’t really interest them that much. They would rather follow an SK Hynix bull post and just long SK Hynix at 10x.

Hugo Martingale asked me to ask when we’re listing Fartcoin?

Fartcoin, that’s a good one. Fartcoin is Sol beta, but not a lot of people are trading Fartcoin anymore. People treat Pump as the Sol beta now. Pump has a lot more traction, so we have Pump. I looked at volume on Fartcoin. Previously, whenever the market recovered, Fartcoin would bounce so hard, but now that’s not the case. I think the market just lost interest in Fartcoin.

Share The Oracle by Polymarket

Polymarket Perps tldr:

Leverage. Up to 20x on major crypto pairs and around 10x on most equities. Max leverage is set per asset.

Cross-margin. A single balance can collateralize multiple perps positions, including assets that other venues restrict to isolated margin.

Fees. Fees are charged on a volume-tiered structure by trailing 30-day volume and calculated daily. Taker fees start at 0.0400% and fall to 0.0200% at the top tier. Makers start at 0.0125%, and the top tier earns a 0.0050% rebate. Full fees info.

Exchange. Central limit order book (CLOB) with funding payments that anchor contract prices to the underlying index. Margin and liquidation run on a mark price that tracks an asset’s price across multiple venues. All trading collateral is in pUSD, Polymarket’s native stablecoin.

Always on. Markets trade around the clock. Price feeds shift across trading sessions (regular trading hours, overnight, weekends).

Transparency. Every account’s positions and trade history are viewable on its public profile, the same way other DeFi perps exchanges work.

Builders. An API supports trading bots, market making systems, portfolio dashboards, and risk monitors.

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

The Polymarket Institutional Desk offers block trades on U.S. and international exchanges, institutional market maker relationships, and enterprise data services. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

Elon Musk has a long history of rolling up his companies. X.com and Paypal (2000), Solar City and Tesla (2016). xAI and X.com (2025) and SpaceX and xAI in February of this year.

For nearly a decade, Elon has insisted that there was “not a strong rationale” to merge Tesla with SpaceX. But in November we started to hear a different tune. “My companies are, surprisingly in some ways, trending towards convergence” Elon posted in November of 2025.

Will 2027 be the year of the Elon giga-merger?

The quest for AI compute has been the main driver of this “convergence.”

Elon understands the problem that memory output is growing 20% annually but demand is growing at 200%. He can’t buy chips fast enough and the only way out is to leverage the resources of his other controlled entities.

To run AI at scale, three things are needed: power, silicon and a model. SpaceX has the model and GPUs while Tesla holds the energy storage and designs inference chips.

6th August 2026, we got news of the closest collaboration yet between the two companies: the Terafab chip plant, to which both firms committed $16.8b. Yet The Tesla-SpaceX merger odds on Polymarket barely moved on the news.

What moves the odds?

Before the SpaceX IPO, movement in the odds were mainly driven by speculation. Post SpaceX IPO, the odds were driven by the relative performance of Tesla’s and SpaceX’s share prices. A stronger Tesla makes the acquisition more expensive for SpaceX. A weaker SpaceX valuation reduces the attractiveness of using its stock as a currency given it dilutes existing SpaceX shareholders.

All in on AI

Both earnings transcripts show the firms are reorganizing around AI. Tesla is decommissioning both Model S and Model X lines at Fremont for Optimus. These humanoid robots rely on AI chips to function. SpaceX has already spent $28.5b on capex in the first half of 2026, most of it AI-related.

SpaceX’s earnings report showed that it had ~$93.5b of cash and revenue growing 92% YoY although still loss-making. On the other hand, Tesla is profitable with ~$1.11b net income but has negative free cash flow of $1.09b driven by 142% growth in capex. If they combined their balance sheet, the single entity could fund AI capex internally rather than each trying to raise capital separately.

Moreover, the market has underpriced how Elon operates and his intense desire to build up the required AI capacity. Currently, SpaceX is at 1.4GW of compute and during the Q2 earnings call, Elon suggested a range estimate of up to 10GW compute (approximately 7 times higher) by year end 2027.

Interdependency of the two companies is rising. In the April - June quarter, SpaceX purchased $295m of Megapack high-capacity batteries from Tesla as SpaceX scales its data centre operations. This purchase value already makes up around 60% of Tesla’s 2025 Megapack sales.

In Q2 2026, two customers of SpaceX made up ~38% of consolidated revenue compared to a single client taking up ~17% a year earlier. The revenue from Client B is entirely related to the AI segment. Client B was below the 10% threshold required for reporting but reached 19.5% in the 2nd quarter, indicating the growing importance of AI revenues. A merged entity will carry Tesla’s revenue base and will own Tesla’s AI chip designs, Optimus, and battery hardware that SpaceX has been buying aggressively from Tesla.

The main risk to a possible merger is China. Tesla’s Shanghai plant can produce more than 950k Model 3/Y per year. Combining both entities could expose SpaceX (which holds massive Pentagon contracts for satellite launches) to national security entanglements. This would be a key regulatory hurdle, but it is a deal execution risk rather than an announcement risk considering spinoffs could be a remedy: this polymarket resolves based on an official announcement from SpaceX or Tesla.

Leading indicator

Since the SpaceX IPO, Polymarket merger odds, and the SpaceX / Tesla share price ratio moved together with a strong correlation of +0.71 across 27 trading days. On July 22nd, as Elon ruled out merger talks, that correlation fell to -0.12. The share price ratio has since recovered but the year-end merger odds have not.

There is a good reason why the December 2026 contract stopped repricing: the market has only four months left to resolve and Polymarket traders believe there is not much room for that to show up in the price due to time decay.

However 2027 is a blank canvas. The four contracts (Jan, Feb, June and Dec 2027 strikes) started trading on August 15th. Based on the strong correlation we saw earlier in the year, my view is that the share price ratio can act as a good leading indicator because equities reprice faster than the implied merger odds.

Between Elon’s past rollup history, and the increasing signs of a merger on the horizon, an interesting way to play this would be a relative value trade if there’s any further dislocation between the Polymarket merger odds and the SpaceX/ Tesla share price ratio.

Related polymarkets:

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Alexander Muse aka (@amuse on 𝕏) is one of the largest political accounts on the American right: half a million followers on 𝕏, a daily op-ed habit spread across (at least) three Substacks, and sources inside the administration, Congress, and the Texas statehouse.

Elon Musk has replied to his posts more than 25 times by PolitiFact’s outraged count, Karoline Leavitt and Donald Trump Jr. are among his followers. Combined, his channels reach around ten million Americans a month.

We caught up in late August to discuss the Texas Senate race and who could be voted off Trump Island next.

This interview has been edited for length. All answers are his own.

Subscribe now

Let’s establish first who you are, and where you sit in Republican politics.

I’ve been on 𝕏 since 2007. In a previous life I was a tech entrepreneur, and in the startup world being a conservative was the kiss of death, so I just didn’t tell anyone where I stood. After January 6 I was banned. Elon bought 𝕏 and restored my account, and I grew it from maybe 200,000 followers to half a million pretty quickly. Now I do 𝕏 full time: daily op-eds, a newsletter I send to GOP donors. And people from the administration, from Congress, from state politics start contacting you. They give you leads, they give you scoops, and you build a bigger and bigger base of sources. I don’t consider myself a journalist. I’m a commentator with opinions on policy, and I try to share them in a format that’s easily consumed by policymakers. One in a hundred pieces I write suddenly makes a difference.

Can you think of a story where you’ve been really ahead of the curve, either from your own sourcing or your political instincts?

The cabinet selection process, between the election and January 20. I had really good sources in the room, and there was a relatively small group of people gaming out who was going to be picked for each role. What shocked me was the political consultants, the people who run the campaigns and the super PACs, and the outsized influence they have on picking who runs the country. I watched consultants torpedoing their own clients.

Why would someone attack their own client?

Say you’re on the short list for Attorney General of the United States, and you’re currently the attorney general of some state. Your consulting firm wants you to run for Senate in the midterms instead, because a consultant makes a huge percentage of the spend on a Senate race. So they go out and say our own client isn’t confirmable, he’s got all these problems. They’ve done the opposition research, so they’re using their own oppo against their own candidate. The sausage is pretty ugly.

Let’s talk Trump approval. What is driving his approval with the base versus with these donors you talk with?

There’s a hard base for Trump he’s never really going to fall below, a could-do-no-wrong group, probably 30 to 35 percent of the GOP. Within his own party he’s more popular than Obama or Biden were at the same point, and your party is most frustrated with you right before the midterms. Among the donor class, he’s never been more popular. The money is flowing in faster than you can count it. On the overall number, you’re lucky if you keep it above 45; 45 is a huge win.

Share

Nate Silver is the resolution source for the Trump approval polymarkets. Do you think Trump’s approval is basically capped on the downside from his hardcore base?

I think so, and nothing can change until the midterms. Iran is not going to do a deal before then; they’re of the opinion the midterms give them the most leverage. But here’s where you could be pricing the market wrong. Say November 6 comes and we don’t lose both chambers. The war ends favorably, the GDP numbers are great, the AI and data center spend keeps going, and his approval could easily go back over 50 by the end of the year, maybe January or February if the Iranians continue to be annoying.

If the Republicans lose the Senate and the House, what do you expect the Democratic Congress to do? They’ll cut off funding?

The War Powers resolutions pass, and there are plenty of Republican votes for them; that becomes a Supreme Court fight dragging into 2027. And almost a certainty: starting in January you have the first impeachment proceedings, then hearings on every single business transaction his children have been involved in, Trump coin, Epstein, every imaginable thing.

Today, Polymarket is showing an 88 % chance for the Dems to win the House, but only 63% that Trump will get impeached before his term ends. If you think it’s basically a lock that he gets impeached if Dems win the House, why isn’t impeachment higher?

It’s underpriced. That should be at least 88. And honestly the 88 is probably too low; it probably needs to be 92. There’s room to be had in both those markets.

Let’s talk about the Senate. You’re very close to the Texas race, which is basically 50-50 today.

The chances of the Republicans holding the Senate are much greater than this market reflects. Take Texas as the microcosm. The polls showing Talarico with a three-to-five-point lead are run by Democrat activists, and the crosstabs are complete BS. Almost assuredly, Ken Paxton is up five points in Texas, and he will win by at least three, maybe four.

And look at your own history. In the primary runoff, the public polling had Cornyn winning or barely losing. Your market had Paxton 62-36 right before he won by 27. The market got it right in a really good way, and the pollsters did not. Ken has never lost a statewide race, he ran against a guy who outspent him ten to one running every scandal on TV all day, and all the bad news about him has been out for years. The market should be at least 75, maybe 80. There’s money being left on the table.

Let’s pull up the Real Clear Politics average. Which of these pollsters are the bullshit ones?

Texas Public Opinion Research (TPOR) is completely fake. Those are literally fundraisers for the Democrats, and it forms the basis for everything else: the Fox poll pulls on it, other pollsters set their conditions on these things. That one huge poll is what flipped the public polling. Ken had been ahead ever since the runoff. I wrote a whole piece for you guys on why that pollster is complete BS.

What the fake polls are doing is oversampling. They’re assuming Republicans are peeling off, Cornyn loyalists staying home. That’s not happening. And they’re not taking into account the Black Democrat vote in Texas, a very religious voting bloc that is not excited about Talarico’s theology, what amounts to a Christianity that doesn’t believe in the Trinity or the basic tenets of the faith. Go watch the town halls in predominantly Black communities: he’s getting shouted down by his own constituents. Some of that may be residual resentment over his defeat of Jasmine Crockett, but it’s also a rejection of his politics. It doesn’t turn them into Republican voters, but it keeps a lot of them home.

Share The Oracle by Polymarket

Last time we spoke, it was before Tulsi was out, and you’d intimated that more people were heading for the exits. Who will be out before the end of the year?

Trump was reluctant to make changes because of all the turmoil in the first term. The person he was least thrilled with was Pam Bondi. I don’t think she was his choice; that was his chief of staff’s choice after the Matt Gaetz debacle, and he went along with it because he trusts his team. Once the Kristi Noem termination happened, the band-aid had been ripped off, and everyone was up for grabs. From that moment, Pam was in the process of being replaced.

What about Lutnick? What puts him in the higher likelihood to be out group?

Howard Lutnick at 30% is the Epstein stuff; I guarantee it. There are MAGA people calling for his head constantly because of the statements he made about his involvement. But I don’t think that carries with Trump, because Trump knows the truth on all that. And Lutnick is doing a great job. The Intel deal was what, a twenty-billion-dollar check to the Treasury? Free money. Trump’s a dealmaker; he likes that. He’s not going anywhere.

Susie Wiles at 24 percent could be underpriced. She’s struggling with breast cancer right now, and she is actively campaigning for Marco Rubio for the 2028 nomination, which puts her in conflict with the administration and with J.D. Vance. You’ve seen your own market moving from Rubio to Vance; the more you see J.D., the less power Susie has. As she goes through her treatments, it may be a real convenient time for her to move to the next phase of her career.

On Rubio: a lot of us have been shockingly impressed. He’s carried himself with a level of honor and loyalty I did not know he possessed. But he doesn’t agree with a lot of the policies he’s implementing, and the moment he’s in charge, he does things differently.

What about Hegseth? Is he in trouble because of the “running out of missiles” narrative?

Hegseth is one hiccup from being out at any point in time. He’s doing exactly what the president hired him to do, and doing it exceptionally well, but a lot of it can be interpreted really badly by our own side, particularly in the Senate. If it were only Trump’s decision, he’s happy. And the missile-gap reporting is disingenuous: stockpiles are allocated per theater, and the Middle East arsenal being extinguished doesn’t mean we’re out of missiles. They moved stockpiles from Europe; you can watch the cargo flights on FlightAware. Trump was not confused about that.

How do you size up how the U.S. is doing so far in the Iran war?

It depends on what you set the condition of victory to. Militarily, we have basically defeated them. Their entire naval fleet has been destroyed, almost every mobile launcher, and they’re sitting on 10 to 15 percent of their previous drone and missile capability with no ability to construct new ones, because we’ve degraded their manufacturing base. From a capability perspective, this is total victory.

However, the Strait of Hormuz is a threat you’re never going to be able to fix. Say Iran surrendered tomorrow. There’s still some former IRGC commander running around with a speedboat and a shoulder-fired rocket, and he can cut traffic through Hormuz 75 or 80 percent on a weekend, because he hit one ship and everybody stops. What Trump has done, and I think Iran striking its neighbors was a huge miscalculation that made it possible, is bring the whole region under an American security umbrella with a focus on energy. There are twenty-two billion dollars of pipeline projects now to divert 70 percent of the oil around the Strait, underground, in the next 18 to 24 months. U.S. companies are going to make billions making the Strait of Hormuz a non-issue.

Regime change?

We killed the guy in charge, the new guy is just as crazy, and the next one will be just as crazy. If that was the goal, we lost that game.

If you had told me that the Strait of Hormuz was going to be closed for months and asked what the price of oil was going to be, I would not have guessed what the price is right now. Is it demand destruction? Is there secret oil being pumped out somewhere that we don’t know about? Is it just the SPR?

We’re moving a lot of oil. Tankers come through every day; they just turn off their AIS. We’ve seized a lot of oil. We took over the Venezuelan oil market and production there is climbing. And we’re producing a boatload here. Prices are high enough to go get all the oil sitting in the Permian, so every single rig in the national fleet is out right now. The amount of oil we get out of that in the next year is going to be unprecedented.

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

The Polymarket Institutional Desk offers block trades on U.S. and international exchanges, institutional market maker relationships, and enterprise data services. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

By

The Fed’s annual Jackson Hole Summit is a time honored tradition in the central bank community. Customarily, the Fed Chair delivers a speech, which is closely watched by markets as it is often a tone setter for the final three FOMC meetings of the year.

This year will be Chair Warsh’s first speech and it is coming at a very interesting time. Not only is the new Chair finding his footing in the job, but it is a time of great debate at the Fed. At the July meeting, the central bank decided to hold interest rates steady.

This was meaningful as the market priced in close to a 40% chance that the central bank would hike rates by 25bps, and 3 members of the FOMC dissented against the decision to keep rates on hold, the highest number of dissents since September 2016. So not only is this Warsh’s first Jackson Hole, but it comes at a time when the committee is at odds with how to proceed in the current economic backdrop. This tension came across quite clearly in the minutes of the July meeting that we received last week.

Traditional markets are debating whether tomorrow’s speech will be hawkish or dovish. Will it move the pricing of the September meeting, currently priced at 9bps or 35%?

What is interesting to me now is, on Polymarket we can now trade the speech itself. I.e. Instead of looking for which side of the “family fight” Warsh is on, or if there is a tip at all for the September meeting, we can trade the actual words he is going to say. What is so interesting about this to me is, Warsh is famously not a big believer in forward guidance, as we have seen with his first two FOMC press conferences. This means there is a high chance that the answers the bond market is looking for are not actually present in Warsh’s first Jackson Hole speech.

Nevertheless, Polymarket’s “What will Kevin Warsh say during his speech at Jackson Hole?” has some very interesting options to trade.

These are the type of markets we want to bring awareness to on Polymarket Institutional as they are fundamentally differentiated from traditional markets. And between block trades and organic growth, we expect the depth of these markets to expand.

For macro traders that are laser focused on what Warsh will say tomorrow, these markets offer a very good venue to trade on that directly.

With that said, some of the options listed are:

• Task forces 5+ times

• Stable 5+times

• Too late

• Productivity

• Independence

• Capex

• Integrity

Using AI, I built a tracker to help me with the concepts and words that Warsh has used most in his press conferences, congressional testimony and public appearances as Fed chair.

One trade that screens as interesting based on this, and empirically feels right based on watching Warsh’s press conferences closely, is that “task forces” 5+ times at 38% feels too low. Jackson Hole is a venue to discuss big picture topics, and Warsh has come in with a big reform initiative that is highlighted by the work of the task forces he has appointed. I expect a big theme of the speech to be the early findings of these task forces but also what he hopes to accomplish with them in general. I think 5+x below 50% is good EV for something that from a thematic perspective will likely be the main part of his speech.

In terms of markets like September hike, which is a very close call right now on Polymarket, I think there are two key things to be looking at.

1. A big tension on the committee right now is the debate between the spot data and the balance of risks. The hawks are of the view that the balance of risks around inflation has not improved in recent months, while the doves point to more dovish data on inflation and employment over the last 6-8 weeks. If Warsh gives any indication where on this debate he is, that will be meaningful for the September meeting odds.

2. Post the July meeting, there has been a lot of pressure on Warsh to actually show the key metrics he is looking at in a way that is different from “forward guidance.” Jackson Hole is a good venue for him to list a few, and the ones he lists will be a sign of where his reaction function is at the moment.

It is Warsh’s first as chair and it is coming at a time of big debate on the FOMC. An interesting angle to play this from is trading on what he is going to say on Polymarket instead of relying on the speech being “impactful” in traditional market terms.

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Werner K. Zagrebbi is a board-certified Superforecaster (TM) who writes about politics and economics at Right Rationalism.

GMU professor Robin Hanson arguably invented modern prediction markets. Here’s part two of his interview with The Oracle (read part one here!).

Subscribe now

Your logarithmic market scoring rule from 2003 is the ancestor of every automated market maker in existence, including the crypto ones moving billions of dollars. And Polymarket runs an order book, not your subsidized market maker. How should people trade it?

There are different trading strategies. The simplest: ask what’s true about the world, compare that to the price, and ask where you think the price is wrong. Then ask, what’s the chance other people don’t know this thing that I know? You shouldn’t just compare the world to the price, because maybe the price is based on things other people know that you don’t. Look for pieces of information you have that other people are not likely to have, so it’s an addition to the price, and trade on that.

Another strategy, very popular, is to look for deviations from random walks and move prices toward a random walk. A third is to be a market maker, which is what Polymarket and others pay people to do: pay attention to the market microstructure, sit on both sides with buy and sell offers, and take profit on the difference. My simple automated market maker was just giving up on that. I designed a market maker that doesn’t lose too much money, not a clever one, because if I made it clever, people more clever than it would take advantage of it. These professional market makers are smart enough to be clever.

And the actual finance game is this: if you set a really strong academic reluctance to believe, by the time you’re finally convinced there’s a pattern, somebody else will have used it up and there will be no profits left for you. So professionals have to be more aggressive, which means they’re wrong a substantial percentage of the time, and then it’s all about judgment. When finance companies hire people, it’s all about whether they trust their judgment to decide when there’s enough evidence to act. That’s what we’re all betting on.

Would we have a better world of ideas if more places gave status to contrarianism?

I doubt it. But I think I have found a secret to being a contrarian who’s usually right. The world is full of intellectual silos. The people in each field roughly know the newspaper version of other fields, and they learn the expert version of their own field. And often the expert versions of different fields are just in conflict. You only learn that if you study multiple fields at the expert level. When you find the conflict, there are usually not many ways to resolve it, and usually you can tell who’s right: who has dug into more detail, who has more solid theory. If you embrace that resolution, you are contrarian with respect to the field you’re saying is wrong. And your edge is that they refuse to listen to the other field. When a field hears its results conflict with another field, they deny the conflict. So this is how you can be consistently right and contrarian. There’s intellectual progress just waiting: combine different fields, notice when they’re in conflict, resolve the conflict. That’s something LLMs have a real potential for in the coming years.

This may be kind of annoying, but how would you use this to trade? Because you probably could, right?

Later on, when we have prediction markets on many more topics, you could go look for these intersections and trade on them. Unfortunately, the current range of prediction markets is pretty narrow. But if the people trading in a particular market are all from one discipline, you might ask: what other discipline has expertise relevant to this, and has it been consulted? You’re typically asking: the market has some information in it. Where is that from? And what information could I find that’s not in the market? Look for unusual sources of information that other people in the market haven’t been looking at. That’s one way to have an edge.

Joe Weisenthal wants to know: how important is it that a forecaster can show their work? Broader version: do prediction markets incentivize you to hoard your information instead of publicizing it? You used to get a status reward for writing the article. Now you can just get a payout for silence.

Prediction markets most directly give you incentives to trade, and that’s not a direct incentive to explain your reasons. But the simplest way to make money on information is: you have a piece of information, you trade on it, you reveal the information, the price comes to accept it, and now you can undo your trade and go on to another trade. The major cost of trading is the opportunity cost of the assets you trade. To minimize that, you want the shortest possible time between when you trade on the information and when you get to undo the trade because everybody else agrees. So mostly you want to reveal, not hoard.

The exception is when there’s a big distance between your piece of information and any market you can trade in. Then you’re tempted to hoard it. The more directly you can trade on the information, the more you want to just reveal it and get on to the next thing. This is a reason to like combinatorial markets: a stronger, more direct relationship between the information you have and the markets you trade in. And all information has a duration: the expected time before other people get it. The longer you have it exclusively, the more money you can make, but the longer you have to lock up your money too. So what you really want is to trade as much as you can on your information, reveal it, and be done.

Could you explain combinatorial markets?

An ordinary prediction market gives a probability or expected value for some particular event. In principle, what you really want is a full joint probability distribution over all possible events. That’s what statisticians use, what game theorists use, what decision theorists use. These markets are only giving you marginal distributions, not all the correlations and connections between everything. Hedge funds generate their own joint distribution in the background, their model of the world, and use it to decide where to trade in these particular markets. But it’s actually surprisingly possible to just have a market that represents a full joint distribution over everything.

The key thing is that the full joint distribution has an exponential explosion in the number of parameters. So you need a reduced-form representation. One common representation is a Bayesian network. Roughly ten years ago, I and some colleagues here at Mason made a combinatorial prediction market with over a thousand variables connected in a Bayesian network. It was called SciCast, and DAGGRE before that. And a key theorem I proved early on: if you have a set of base markets with some liquidity, and you introduce all possible combinations of them, how much more liquidity do you need for all the combinations? The answer is zero.

How do you not need any more liquidity to run this market times that market?

Think about the worst case. When you fund a market maker, the worst-case loss is the price going from the middle all the way to an extreme and never coming back. Say you have N markets with the same liquidity: your worst case is N times that. Now make it a full combinatorial market. The worst thing that can happen is you go to the corner of the combinatorial space, which is exactly the corner where each individual market has gone to its extreme. And that turns out to be exactly the same total loss. The problem of combinatorial markets is the computation, not the liquidity.

So a fund with a full model of the world could just trade the whole thing directly.

Once the markets are combinatorial themselves, you can take your combinatorial belief and just directly go trade on it. That’ll be a thing. I do think in the long run there will be combinatorial financial markets, and typical hedge fund practices would change a lot in the face of them.

Let’s play overrated or underrated: play money prediction markets.

They were overrated, but now they’re probably correctly rated. Once Polymarket and others showed up, the demand for play money markets went way down. Which is roughly correct.

Sports as the gateway drug to prediction markets.

Sports betting is one of those things where, once they had legal permission, they went for it because they knew there were a lot of customers there. It’ll lower infrastructure costs and set legal precedents and make customers familiar with things, all of which is good for the other things prediction markets will want to do. It’s not my priority, but I don’t mind it, and it will help the things I’m working for.

The GMU econ intellectual culture.

I’m too close to that to rate it. Mason is a collection of contrarians who have somewhat different contrarian takes. When I’ve been contrarian on issue X, other people at Mason say: gee, we don’t want to be associated with X, that’s not how we’re contrarian. A coalition of contrarians each wants the others to be less contrarian, to give them room to be more contrarian. Most contrarians’ stance is: those guys really are contrarians, I’m really not. The world is going to come over to agree with me, but not with them. And as soon as the world recognizes me, I’m leaving them.

Base rates.

It would be hard for them to be overrated. We can’t rate them highly enough. Most of the time, base rates are what gives you your answer. Most of the time, people are trying to add some correction to a base rate, and they’re overcorrecting. Just the base rate, guys.

One of the central questions about contemporary prediction markets: why is there so little real hedging yet? Is it culture or something structural?

A few decades ago, regulators said: we don’t want to allow prediction markets because they’re too small to support hedging. Markets need to be big to hedge, and if they’re big they need heavy regulatory oversight. That’s why the world just did not legalize prediction markets in the early part of my life.

I do see lots of opportunity for increasing hedging, but I still think the total demand for hedging in the world isn’t that large. And here’s the thing: jobs are a form of insurance. If you as an employee were just paid your actual productivity every week, you would face a very uncertain and often negative payment. When we give you a job, that’s insurance. The company is taking on your risk and you get paid a salary. It’s probably the biggest form of insurance we have in our world, and it doesn’t show up in the insurance markets. It’s not called insurance.

And famously, people face a lot of big risks they don’t want to insure. Career risk. Marriage risk. Risks about whether they’ll be proud of their kids. When life insurance first rose up in the late 1800s, it was kind of a random marketing trick: you’re an irresponsible husband if you don’t have life insurance to cover your family if you should die. It was sold as responsibility.

Well, it’s not hard to figure out. There’s a massive first-mover status disadvantage to being the guy who hedges his divorce risk.

Signaling is often at odds with insurance. People want to signal confidence, and they want to signal that they have prospects. That’s often a good explanation for why people don’t insure. But then all the insurance we do have, you wonder: why didn’t signaling block those? Why aren’t you ashamed to be an employee, not showing confidence in your own productivity?

That one seems clear: in most industries you can’t assess productivity that easily.

Do you expect AI superforecasters to go dramatically past the human frontier?

I don’t actually much care. Prediction markets should be robust to participants. The cheaper participants can be accurate, the more the price of information falls, and the main attraction of this institution is that it plausibly has a lower price of information than the alternatives. But if AI has an advantage here, it probably also has an advantage elsewhere. So it’s not obviously going to make this institution better than the alternatives. It’s going to make them all work better.

You’ve said you pick your battles where adoption of prediction markets is likeliest. I’ve thought of a new application: the futarchic interview. A market on each question, conditional on this one being asked, how many likes on Substack?

To get scale, I would want to advise a whole process rather than each interview. Marc Stiegler had a science fiction novel [Earthweb] where an alien ship is attacking Earth, we sneak some people aboard, and we have betting markets on Earth telling the team on the ship where to go left or right and what to do next. That’s the model: betting markets advising a team making individual choices. The hard part is that any one interview isn’t worth enough to have a big audience for it. You’d want something with high enough attention that people could specialize in advising it, like markets advising the choices in a high-level chess game.

But it’s not one of the first applications I’d advise. What I look at first in the world is: where are we making decisions where other people nearby say, those are bad decisions? People grumbling that whoever’s making the decision has no idea what they’re doing. That grumbling is a hint that there’s a failure to aggregate information. I don’t know that we hear that about interviews. We should smell a problem. Something should stink a bit.

In some sense you’re complimenting me. It’s not as bad as CEO hiring.

Actually, when people do podcast interviews with me, I typically tell them: you’re not challenging me enough. You’re not disagreeing with me enough. This is boring. They have their list of questions to get through, and we’re not arguing over things.

What’s your most underrated idea? You should have a view. Who’s in a better position to judge?

In my mind, the most underrated thing is what I’m doing right now, cultural drift, because I’m an obsessive person, focused on what I’m obsessed with at the moment. But I’d say my best idea is futarchy, decision markets. That just obviously has the biggest potential to make the world better.

Is anyone actually running one?

There are crypto companies being run with decision markets now. So far they’ve done great. I’m excited for that and I try to help them.

Final words of wisdom? Anything you want to say to our readers?

When I was young, at this project trying to make the World Wide Web, we succeeded in influencing the World Wide Web, and I was just a junior person on the project. But what I saw was that the people who had actually taken a chance, had a vision, and succeeded in implementing it and making the web happen, they didn’t win much in the world. They weren’t famous. They didn’t make money. They just had the satisfaction of knowing that they made a big difference. And if my old friend Hal Finney made Bitcoin, he also did not make money on it. Maybe he’ll get fame, maybe not. But he had the satisfaction of knowing that he had a big impact.

I was warned early in life that it’s actually more possible than you might think to have a big impact, mainly because the people who have the big impact don’t actually get that much out of it. That’s the trade-off. It’s easier than you might think to have a big impact on the world, exactly because you might not get much from it personally.

To what extent is your whole intellectual project trying to let people internalize more of the gains for making an impact?

People who make bigger, better things happen should get more gains from it. Prediction markets are certainly a part of this, a way to try to make that happen. But they have to happen in a world that doesn’t yet reward that so much. That’s my high-level advice. But in any case, I can be proud of what’s happened, and of having had an influence.

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

The Polymarket Institutional Desk offers block trades on US and international exchanges, institutional market maker relationships, and enterprise data services. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

On July 31 at a Camp David meeting, Scott Bessent had only one item written down on his to-do list: “Buy Japanese Yen (JPY) $5-10 bil.”

One day earlier, the yen had just hit its weakest level against the dollar since 1986, and Washington and Tokyo responded with the first coordinated currency intervention since the 1998 Asian Financial Crisis. This intervention saw USD/JPY pull back from 164 to 155. In the two weeks since, the $88 billion effort looks misspent, as the yen has slid back above 159.

Polymarket traders are now considering the risk that Bessent was trying desperately to avoid: could the largest funding trade in the world be starting to unwind?

Borrowed time

The yen carry trade is simple: borrow cheap yen, convert to dollars, buy higher yielding assets, pocket the difference. Japan’s policy rate is 1%; the Fed’s upper bound is 3.75%. Traders have gorged on levering up this trade as the yen continued to weaken.

But the setup is precarious: when the yen strengthens, borrowing gets more expensive and leveraged traders get margin called. Covering means selling dollar assets and buying yen back, which strengthens the yen further and forces more selling, creating a vicious cycle.

The yen has steadily weakened since 2011 under Japan’s ballooning debt load and the widest rate gap with the US in the developed world. The Iran war and higher oil prices have accelerated the Yen’s decline, since Japan imports nearly all of its energy.

Polymarket traders have priced the ongoing slide in the past month. When the USD/JPY rate hit its peak of 164, odds of it topping 165 by year-end ran to around 77%. The US and Japan intervention knocked those odds to 30% in two days. But just two weeks later, they are back to 43%. Traders essentially shrugged off the intervention as merely buying time to avoid confronting an inevitable further decline in the yen.

Trigger warning

This is not the first round of jitters about a carry trade unwind; there was a similar episode in the summer of 2024.

When the Bank of Japan hiked to 25bp on July 31st, and jobs data on Aug 2 pulled US yields lower, the yen surged ~6% in just a few days. Margin calls went out on August 5th, crashing the Nikkei 12.4%, its worst decline in nearly 40 years and dragging U.S. stocks down with it.

The trigger in 2024 was the BoJ hiking rates 25bp off a floor of zero. Today, a BoJ hike is the market's base case. While traders priced a September hike only at 8% a few weeks ago, the intervention spiked the odds and sped up the repricing, now trading at 68%.

A September BoJ hike would take Japan’s policy rate to 1.25%, the third hike since December, and a pace not seen since 1989. With the Fed also expected to hold rates in September (75% odds), the carry trade spread would further compress.

Wall Street shrugs?

The stock market has not yet been thrown off by yen carry risk.

As we saw in 2024, equities are the first casualty of a carry trade unwind. Today the S&P sits around 7,800, and traders give 34% odds it tops 8,200 by December. In late July, the equities forecasts held steady even as the yen intervention collided with the Situational Awareness margin call. As both stresses faded, the market resumed its climb, even as the yen resumed sliding.

So while yen stress and AI stress hit the same index as leveraged players sold off, there is still optimism in the short-term resilience of the S&P. The tail risk that the S&P touches 6,200 by December has a 17% chance. If a September hike sends the yen sharply higher, the first signal may be a repricing of the S&P forward strip.

Treasuries impact

But a carry trade unwind would also have second-order effects beyond stocks.

Japan is the largest foreign holder of US Treasuries. By selling Treasuries to buy yen, Japan pushes US yields higher. Odds of the 10-year yield hitting 4.8% before 2027 reached near 70% the week of the intervention. Notably, Bessent funded his part of the intervention by selling Euros, knowing that the US Treasury market couldn’t absorb any more selling.

The yen intervention was supposed to protect the Treasury market. But every round of yen defense means Japan is selling more Treasuries to prop up its currency.

As the 10-year Treasury forecast creeps higher, Polymarket traders are forecasting the slow exit of the most reliable buyer of US debt.

The risk for the equity market is that Japan’s exit pushes US interest rates higher, squeezing the equity risk premium that thinly supports the stock market’s valuation.

September showdown

Polymarket forecasts for BoJ rates and US debt see a carry trade squeeze coming. Stock traders not so much.

These narratives will collide at the September meeting. A BoJ hike, if it snaps the yen sharply higher, would hit equities first and trigger cascading margin calls in a replay of the summer of 2024. A hold buys the carry trade another quarter, but how much lower can the BoJ let the yen slide?

Anton Wagner is studying Applied Math and Philosophy at Harvard University and is an intern on the Growth and Partnerships team at Polymarket.

Related Polymarkets:

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Werner K. Zagrebbi is a board-certified Superforecaster™ who writes about politics and economics at Right Rationalism.

If you take the trip from DC down into that intricate amphibious Northern Virginia region, there could often be seen sauntering across Fairfax a very interesting old man, like a wizened sage—portly—with a glint in his blue eyes and lightness on his feet testifying to a glorious twinkhood lost.

In the words of his colleague Bryan Caplan: “When the typical economist tells me about his latest research, my standard reaction is ‘Eh, maybe.’ Then I forget about it. When Robin Hanson tells me about his latest research, my standard reaction is ‘No way! Impossible!’ Then I think about it for years.”

It could be argued Hanson originated modern prediction markets: If he gets the Nobel Prize (Manifold put his odds at 20%), it’s expected to be for prediction markets.

I came across his work in his capacity as one of the first niche internet microcelebrity bloggers, and I saw in him an intellectual hero worth idolizing. Beginning in High School, I often took the pilgrimage to George Mason’s campus, and faced him over games of Dungeons and Dragons. This month, I went back with a recorder.

Here’s our conversation, condensed and edited for clarity.

Share

Thanks a lot for doing this, Professor Hanson.

Well, you know, prediction markets just boomed in the last few years, and all of a sudden lots of people want to talk to me. That’ll probably go away when it becomes less of a novelty. But at the moment, it’s a novelty.

Why don’t companies run these markets internally? They need the information, and they have the money.

Once you realize you want the people who value the information to pay for it, the obvious first place you’d go is companies. They need the information, they’ve got money, and their decisions are valuable. That’s what people tried to do. In the aughts there was a lot of effort to get companies to do these markets. And what we learned is this: most managers present themselves to the world as scientific decision makers. They’ve got a spreadsheet, they’re trying to fill in the numbers so they can make the best decision for the company. Under that model of who they are, they should want these markets. But that’s not who they are. They’re actually politicians. They have an alliance and they’re trying to support it. They’re trying to recruit allies to support decisions, and they don’t want some random neutral source weighing in that messes up their ability to control the narrative.

Is this what got you interested in the economics of status?

My longer-term history is: I learned the usual points of view in economics as a student, and then I noticed lots of puzzles, things that didn’t make much sense. This is one of them, but it’s only one of many puzzling things about the world. My book The Elephant in the Brain is the realization: what if people are just wrong about their motives? Then you can explain a lot of things. And I think that works here too. Managers’ motives aren’t actually to be the scientific decision maker filling in the spreadsheet.

I have this hypothetical of the autist in the C-suite. You put a really smart person there who understands the company really well, and every time a subject comes up they just spew out lots of relevant, important things about it. They would not be allowed to sit there, because other people can’t rely on them to support their agendas. They would be disruptive. But that’s what a prediction market is. Something that speaks the truth without any sense of who wants to hear it.

Isn’t that kind of the self-image of McKinsey?

The most common scenario with consulting is that somebody wants to make a decision but doesn’t have enough internal support, so they hire McKinsey to be the outside support for what they’ve already pre-decided to do. McKinsey isn’t actually doing that much original analysis. They’re offering their support. It’s credible in part because there’s only so far they’re willing to go. If you bet wrong that McKinsey will support what you want, you lose out. But that is the main thing they do.

Subscribe now

You built the Policy Analysis Market at DARPA, markets on geopolitical instability to inform policy. Two senators held a press conference calling it a federal terrorism casino, and it died. What was that day like?

There was this guy Poindexter, of Iran-Contra fame, very famous, who was in the hierarchy above us at DARPA. Some reporters said: how can we find dirt on him? So they looked at projects under him, and they found our project. To get him fired. He wasn’t fired for Iran-Contra or whatever. He didn’t have much influence over the project, but he was nominally in charge of it, so that was the basis.

They made sure to have their press conference when the DARPA PR person was unavailable. It was a Monday morning, and by the next morning the Secretary of Defense said our project was killed. They did not have much time in 24 hours to assess the situation. They killed it on the basis of: we don’t want to be associated with this bad publicity. And they did not check whether the accusations were even correct.

The other people on the project just wanted to hunker down and hide. I was the one person willing to talk to the press, so I went through several months of talking to the press, and I learned a lot about how these sorts of issues play out.

The power of bad publicity is really strong in our culture.

There’s also the power of accusing norm violations. If you’re associated with an accused norm violation and you say, let me get back to you on that, I need to look into it, that looks bad. You’re supposed to have an automatic response. A norm violation is supposed to be so obvious, and your position on it so obvious, that you don’t need time to react. This is of course not true. In reality almost all norm violations have details you might not understand, and we don’t always agree on what the norms are. But we have this norm that we DO agree on the norms and that their application is obvious. That’s why, when you’re accused, you don’t have time to look into things.

Should it be made again?

It did happen, ten years later. IARPA, a related but different government organization, funded prediction markets, and my team’s pitch was: we’ll do the combinatorics. That was SciCast, roughly a thousand markets, starting with geopolitical events in the East and moving to science and technology. We did it ten years later without much publicity, because Poindexter wasn’t in the hierarchy above us.

Was this useful for the government? Maybe they have a more secret version now.

For a while, maybe ten years, the CIA had internal prediction markets, and they were useful to them. But they were only allowed to exist because of one rule. The coin of the realm in the US intelligence community is white papers. White papers cite other white papers. What you aspire to do in that world is write white papers that get cited. So the rule was: the markets can’t be mentioned in white papers. They were allowed to exist as long as they didn’t get credit. Individuals could participate, look at the prices, let them influence their opinions. But in the white paper they presented themselves as having thought of all these things themselves.

And then that norm ended, and that killed it?

I don’t know what killed it. My guess is that eventually people higher up heard about it, and when they see a white paper they say: what does the market say? Now it’s getting credit. You can’t let that happen.

What should this lesson tell us about the prospects of prediction markets as an institution in broader society?

Institutional innovation is just more difficult, because you’re threatening people’s status and power. People are much more okay with innovation that happens in a lab: if the lab succeeds, they change the production process of a product, and the whole organization stays the same, except now they’ve got more money. Innovation of institutions, of who gets rewarded and what we believe, is much more threatening. So you should expect it to be harder to get experiments started, harder to get them to last, harder to produce innovation. And you can’t patent institutional structures. It’s a basic fact about the world: we are much more eager for innovations in computation and physical devices than in innovations of social practices. The biggest social innovation of the last few decades was social media, which was presented as a technology. People were okay with introducing it as a technology, and then it had social implications as a side effect.

You were very early to a lot of social phenomena. The web, AI, prediction markets. Is there anything you’d add to the list?

The world of people I was hanging out with in Silicon Valley in the late 80s. I left in ‘84 to go out to Silicon Valley, and I went away in ‘93 when I went to grad school. In that world there were lots of big ideas for the future that people were throwing around, and I was trying to contribute to that. In retrospect we could say I added prediction markets to that world, and that was successful. Other people were doing other things. Hal Finney, of Bitcoin fame, was one of my friends in that world. He’s my best guess for at least part of Satoshi. Eric Drexler, who many people thought was more likely to be the harbinger of the future, for nanotech. That turned out not to be such a thing yet. There were people trying to make social media at the time. There was cryonics, which I got into eventually.

Share The Oracle by Polymarket

Prediction markets as they exist now, what percentage of that vision do we have? They do reference Polymarket prices in WSJ now.

When I first came to the subject, I was imagining: let’s just have legal markets on the questions we all argue about in public policy and politics and the media, and that would be enough. That’s the initial vision, and I think it’s the vision most people start with when they come to the subject. Every few years more people come to the subject, and that’s where they start.

There are three things I think people should eventually come to realize. First, the world has far more topics that matter than the topics that show up in the news. Second, the markets we’ve had so far are mainly a side effect of traders who are there for risk hedging or entertainment reasons. What we want is to switch to a model where the people who want information pay for it. Then we could unleash a much bigger market. And third, decision markets: if we want to make things relevant for decisions, we make the markets right next to a decision.

But even on the first milestone, look at the number of news articles that show up in a year and the number of markets that show up in a year. We’re still less than one percent. We have a long way to go even to reach that milestone, and that milestone is still much smaller than the bigger long-term vision.

So is it, in a weak social science-y way, inevitable that we eventually get more futarchy, because the firms that do it outcompete the ones that don’t? Maybe the AI firms of 15 years from now?

That’s the key open question. Years ago I wrote a blog post listing 21 ways firms seem inefficient. You might think that for each of them competition will eventually force the more efficient thing. But it’s an open question. I still hope, and there’s a chance that’s worth pursuing even if it’s not obvious it’ll happen. If you know the literature on innovation, the world just has far less innovation than would be possible. We’re proud that we innovate much faster than people did three centuries ago, and that’s true. But compared to what seems possible, we’re still way below.

Now, before the web existed, you worked at Xanadu, Ted Nelson’s legendary hypertext project: the people who invented the link and spent decades trying to build a smarter internet before the internet. I’ve heard you say prediction markets are in some sense part of the same intellectual project. To cleanse the world of ideas.

That’s where I got my inspiration. Other people had this idea of the World Wide Web, and I thought that was cool. I went to Silicon Valley in part to work with them on that idea. They had a vision: they thought making criticism easier to find would help conversation. And I had doubts. I thought, is that really going to work? It probably has helped, but it’s not nirvana. That’s when I asked, well, what else could we do? And that’s when I started to think about prediction markets. In that context of big, grand things we could do with new technology to change the world.

Random thing I’m curious about: does LessWrong, the forum the rationalist movement grew up on, owe anything to Xanadu?

The LessWrong community has a lot in common with the Xanadu community. Similar sorts of people, a generation apart, trying to do similar sorts of things. The Xanadu people thought of themselves as thinking rationally about which technologies would be beneficial and how, and as better able than other people to see possibilities. That’s what the rationalists have seen themselves as, too. I don’t know that one caused the other, but whatever forces caused one probably caused the other.

The only org chart overlap was you.

There were other people associated with Xanadu who were also in that community. Still around, still having influence.

The FOOM debate, 2008. Famous, sort of. Eliezer said one AI project suddenly explodes past everyone. You said growth will be broader, slower, more like normal economic growth. Seventeen years and one revolution later, I think both of you could claim partial credit. Could you imagine four cells: what did you get right, what did Eliezer get right, what did you each get wrong? Ems, the whole-brain emulations your book The Age of Em predicted would come before AI, do seem wrong, right?

People inspired by Eliezer are still predicting FOOM [explosive acceleration in AI ability]. Soon, to happen in a few years. They say it hasn’t happened yet, but it’s about to happen. So to settle this bet, we just wait a few years and see if it happens. Ask me again in ten years. Then I’m going to look pretty right. They will not have seen the FOOM happen in the next ten years. Of course, they may say it’s still coming then.

On ems: if AGI shows up in the next five years, then it shows up before ems, most likely. There are people working on ems right now. But obviously a lot more attention has gone to the usual kind of AI. There are many orders of magnitude more money in AGI than in ems.

Okay. I won’t give you the L on that one yet.

I see Eliezer’s argument as: it’s possible for a very basic algorithmic change to have this huge impact, where it allows the machine in the basement to take over the world on the weekend. And if that’s true, we need to prepare by giving all the AIs good values through that transition. My critique was that the sudden explosion is not very plausible. It’ll happen slower, we’ll have more chances to react, and it will be less concentrated, spread across the whole world economy, not just one machine in the basement. He has been saying lately: see, Robin, you were wrong, because look how algorithms matter. And I say: we do have algorithms that matter, but they matter when paired with lots of data. The big ventures take pretty widely known algorithms and combine them with specialized implementations and data to be the leaders in these industries. The difference between the leaders is often data advantages, hardware advantages, market position advantages. Not just algorithm advantages. Algorithms matter, but not in the way that a machine takes over the world in a weekend matters.

If you re-ran the debate today with bets, are there sub-claims where you could find a price?

There have been forecasting efforts over the last decade where groups tried to get people to forecast different aspects of AI predictions, and then asked whether that helped resolve differences. It apparently did not. When people break the argument into pieces and argue over the pieces, that hasn’t reduced the difference between the groups. And in a way it did for other topics. There was one study with several topics, and on the other topics, breaking things down and engaging the details brought the groups together. Not on AI. They had subject matter experts and general smart superforecasters engage on the details, and on the other topics they came closer to each other. On AI, the superforecasters said: no, we don’t believe it. Sorry. So the AI debate is somewhat different from these other debates. The technical details just don’t make a big difference

Generally, who should have status as a good forecaster? The actual superforecaster people, Polymarket traders who do well, or literally you?

Our world is too much structured by status as the default organizing principle of institutions. If you don’t have a prediction market and there’s some topic we’re uncertain about, the usual thing is: who’s highest status here? What do they say? That’s how we handle hard questions. I would rather replace that. I want status to matter less as the way we run parts of our world.

In my world as an intellectual, I hear people getting status for doing impressive things. And I go: no, I want to give status to people who made an impact, who actually made a difference, not just did something impressive. A lot of academic work is impressive but not very useful. People tell me I’m smart, and I go: I wasn’t trying to be smart, I was trying to accomplish something. If you’re impressed with what I’ve accomplished, I’ll own that. I deserve the status when I did something that mattered. But I don’t want to hand out status for potential, or impressiveness, or way with words, or the way he holds himself and commands authority. I want actual accomplishment. I’ve met Elon Musk. He’s not that impressive as a person. For real. But damn it, he deserves status, for what he’s done in the world. I don’t know who deserves more. So hand out status for actually doing stuff, not for how someone holds their head when they’re standing next to you, or talks fast, or uses a big vocabulary. In a prediction market, people who trade and make money because they moved the price in the right direction: that’s a real accomplishment. Give them status.

If you’re reading this, go give a high five to the winning Polymarket trader near you.

The one who’s made money.

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

The Polymarket Institutional Desk offers block trades on U.S. and international exchanges, institutional market maker relationships, and enterprise data services. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

On July 30, margin calls forced AI hedge fund Situational Awareness to sell its $16 billion public-equity book to Citadel after the fund cruised into July’s AI selloff at a reported 4X leverage.

This article examines how this episode played out on Polymarket, in particular the “AI bubble burst in 2026” contract, which is a near-perfect proxy for the market forces that no doubt made Leopold’s pre-wedding investment “colloquium” far more stimulating.

But were Polymarket traders simply responding to the news headlines? Or anticipating them?

Drawdown gauge

Polymarket’s AI bubble burst contract resolves YES if at least three of six conditions occur within a 90-day window before year-end.

The criteria are a mixture of equity drawdowns, compute pricing, and AI lab corporate events (eg bankruptcy) that would result from significant market pressures:

These are the exact moves that would also push a highly leveraged AI fund like Situational Awareness to its breaking point.

By comparing the movement of this AI stress index to the news flow around Leopold’s fund, we see that this Polymarket contract anticipated many of the key developments in the saga.

The above chart plots the Polymarket AI burst odds alongside the VanEck Semiconductor ETF, a widely cited proxy for the AI trade.

Notice that the Polymarket contract didn't move higher until the Situational Awareness stress emerged in late July, while the VanEck Semiconductor ETF had been falling all month.

In this instance, the Polymarket contract arguably reflected the acute stress around the fund unwind more directly than one of the most-watched AI proxy ETFs.

What moved first?

Prior to the Financial Times announcement on the evening of July 29 that Situational Awareness was seeking capital, Polymarket traders had already anticipated that the fund was in trouble.

For the two weeks leading up to the margin call, the market sat in a narrow 15-to-17% range, even as the semiconductor index was already rolling over. Then, on July 27, traders broke this range, bidding the odds to 20%. By the afternoon of July 29 the market had repriced to 22%; the FT’s report that evening only took it the rest of the way, to its 28% peak.

Right now, the AI bubble burst market is priced at 12%, aligning more with its pre-Situational odds.

Another market also seemed to anticipate this event. NVIDIA's odds on Polymarket of being the world's largest company at the end of August fell from 78% on July 24 to 36% on July 30. This anticipated the dip in the AI hardware equities that Leopold was so highly leveraged on.

Containment or contagion?

Within a week, Polymarket’s AI stress contracts had round-tripped the entire move and kept going. By August 6, the AI bubble market sat near 15%, below the range it held before the margin calls. The largest AI-fund blow-up of the cycle just happened, and the priced odds of an AI bubble burst are lower than before. Instead of treating this sale as a forewarning about the future of AI, Polymarket traders seem to be pricing this event as a risk isolated to a single overextended fund.

NVIDIA’s end-August dominance odds, which had cratered to 36% on the day of the sale, are back above 93%, higher than before the unwind. Even the broad US recession contract barely twitched, holding at 10–12% throughout. The polymarket on whether the fund itself winds down is priced at just 5% by year-end.

On August 5, Situational Awareness announced it was closing a $400 million private investment in a Sequoia-backed company, on top of $100 million it had put into the same company a month earlier.

Polymarket traders reacted early

Across a handful of Polymarket contracts, you could watch as systemic risks to the AI trade were repriced in real time. Equities markets repriced the AI complex in the days before the margin call, jolted when the sale broke, and have climbed steadily back since, capped by the fund’s own return to investing. But while equities gyrated wildly as traders hunted Leopold’s troubled trades, Polymarket moves were more modest, never pricing the AI bust scenario above 30%.

For both retail traders and institutional desks, this case study shows the value of looking at event contracts as a contra-indicator to index proxies to sort the signal from the noise during periods of market dislocation.

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

is the chief political analyst for Israel’s top rated Channel 12 news outlet, and a columnist for Israel Hayom. In English, he writes It’s Noon in Israel on Substack. The Oracle caught up with Segal to discuss why he thinks the October 27 Israeli election is mispriced, the state of the Trump-Netanyahu relationship, dark horse coalition scenarios, and how the Iran war will end.

This interview has been edited for length. All answers are his own.

Subscribe now

Netanyahu has dissolved parliament ahead of the October 27 election. Polymarket has Netanyahu, the ultimate political survivor, at just one-in-three to be prime minister after the election. Does that surprise you?

First, an introduction crucial for Polymarket investors.

Unlike in the US or UK, an Israeli election might end in a draw. It can also end in a rotation government, one figure serving as prime minister for two years and another for the next two. Or in an outright majority.

Here’s the thing. Netanyahu, as the incumbent, can enjoy a hung parliament. If there is no clear winner, he stays in office in what we call a transition government. So any chance of a draw adds chances for Netanyahu, the only outcome that can come of it. And if a rotation government is possible, that also increases Netanyahu’s odds, as he’s the leader of the right wing.

I look at Polymarket every day, and I’m quite surprised by Netanyahu’s low chances, and by the very low chances given to Avigdor Lieberman.

In Israel, the decision on the next prime minister is in the hands of the parliament, not the people. It is neither a beauty contest nor a popularity contest. In 2021, Naftali Bennett became prime minister with 5% of the popular vote. The lesson is that in Israel, kingmakers can become the kings themselves.

Share

How could Lieberman become prime minister?

Eisenkot is the front-runner in the opposition, leading every poll since April or May. But Avigdor Lieberman can make his way to the prime minister’s office, because he is not a natural part of the opposition. He is a settler living in a settlement. He is not exposed to pressure from center-left news outlets. His political base is Russian immigrants. He is a maverick.

Second: he is 68, so he knows this is his last round. And third, he’s a kingmaker. He belongs to the right on diplomacy and the Palestinian question, but to the opposition on domestic issues, liberalism, the economy. So he can come to Eisenkot the day after the election and say, “I’m one of only three people in Israel ever to serve as defense minister, treasury minister, and minister of foreign affairs. Give me the chance to be prime minister first.” What can Eisenkot do? The only guy you have to convince after the election, if not the voters, is Lieberman himself.

So I think Lieberman is heavily underrated and written off. He is somewhere around 4%, 4.7% on Polymarket. I would give him something like 20%.

Also: we have two types of polls. In the mainstream media, Netanyahu is heading toward his worst defeat in 20 years, 51-52 seats for his bloc out of 120. But according to Channel 14 and Channel 24, more affiliated with the right wing, he is heading toward the best victory of his career. Statistics cannot explain that gap. Margins explain 3% or 4%, not 12% or 13%.

And I want to give you the magic number: 55. We have 10 Arab Knesset members who do not take part in coalitions: they do not sympathize with the Zionist state, and the Zionist parties do not want them. Post-October 7, there is almost no legitimacy, especially in the center, to base a coalition on non-Zionist, sometimes anti-Zionist, Arab parties. So if Netanyahu gets 55 seats, he does not lose to Eisenkot in terms of Zionist seats.

The average of the polls has Netanyahu around 52 and a half, and we know the shy voter phenomenon, the inability to really poll right-wing voters, from the US, UK, and Israel. So even trusting the mainstream polls, Netanyahu is two and a half seats short of ending this election without Eisenkot being able to form a stable coalition. That leaves two options: a new election, meaning Netanyahu stays prime minister, or a rotation government, in which Netanyahu most likely goes first.

Remember: 61 seats are enough for a victory speech, but 59 does not force you to concede. That’s what Americans fail to understand about Israeli politics. One can get away with 52, 53, 55, 57 seats. The sitting prime minister deserves a premium simply for sitting as prime minister.

So I would give Netanyahu 45%, Eisenkot 35%, and Lieberman 20%. Netanyahu is still under 50% because of some 100 consecutive very bad polls. Eisenkot is still the one with the best chances in the center-left. But Lieberman is there, and Netanyahu has a higher chance than on Polymarket now.

As we record this, the US headlines say ‘Netanyahu visits DC amid tensions with Trump’. Are they really tense? Or is this a media invention?

Tension did exist. But as a general rule, the relationship is always better than meets the eye in the press. It is based on four to seven people from the two sides of the pond: Netanyahu, Ron Dermer, Trump, Kushner, Witkoff, and maybe Susie Wiles. Seven out of seven belong to the right, while most of the media covers it from a center-left point of view. That means lack of information, and a natural bias towards saying the situation is worse.

The relationship has cooled a bit since March, for two reasons. The war was less successful than hoped, and this administration has become more closed, with decisions taken by fewer people, so it’s harder for the Israelis to penetrate. But I would not write off Netanyahu’s chances to talk Trump into something.

Big picture, what does Israel want out of Iran? Did they already achieve their goals with getting rid of Khamenei?

No, not at all. There is no shortage of lunatic fundamentalist leaders in Iran. There are two goals.

The deliberate one is the nuke. Specifically, the 440 kilograms of uranium enriched to 60%. Israel and the US destroyed the nuclear facilities, but as long as Iran holds this uranium in underground bunkers, it has the theoretical option to relaunch nuclear production in a way Israel is not aware of, or unable to respond to. So extracting the enriched uranium, one way or the other, is the spoken goal of the war.

And if the Iranian regime collapses, the next regime is necessarily way friendlier. That’s the unspoken goal, because it’s way more difficult to extract regimes than to extract enriched uranium.

Odds that Iran agrees to end enrichment or surrender their existing stockpiles are way down.

To be honest, I wouldn’t give it even 1%. Unlike North Korea, for the Iranian regime the nuclear program is not only a means to ensure survival. It’s almost a religious principle. Iran is not only a country, but an idea: to spread the Islamic revolution, first within Iran, then the Middle East, then the world. Giving it up in return for lifting sanctions would mean the formative idea behind the revolution is dead. They’ll word it in different ways, but they’ll never agree to stop. Unless there is a regime change, of course.

So how does it end?

For Trump, there are three options. One is extracting the uranium with boots on the ground. Second, an agreement. And third, regime change.

Israel does not believe in an agreement, because we know our customers. Once Iran has the first Muslim nuclear bomb, it is no longer vulnerable. It is on the world stage. So for Israel there are only two options. The failure of the Epic Fury operation came from Israel wanting to extract the enriched uranium, whereas Trump said, “Let’s wait, we can reach an agreement that will give us the uranium. Why bother?”

In my mind, a ground operation to seize the uranium is a 1%, crazy idea. How likely do you put it?

It was on the table at the start. There are no big secrets: we all know where the uranium is, and we know the abilities of the US and Israel. Former defense minister Yoav Gallant, himself a former commando commander, said that while it was possible three, four months ago, he is not sure it still is, because the Iranians have prepared many obstacles on the ground.

I still think it’s feasible. The main obstacle is psychological, for the Republican Party and this very Republican president. Sending American combatants to Iranian soil is boots on the ground, even if temporary, even if a few hundred people rather than hundreds of thousands like in Iraq. For a president whose raison d’être in 2015 was to be the first Republican ever to speak out loud against the Iraq War, it was too much. The obstacles are more political than military.

Does the war just continue in phases, fight, pause, fight, pause, until something happens with the uranium?

We keep talking about it in a binary way: war or no war. We should adjust to the idea that it’s going to be non-binary, something in between.

And there is something else: the Iranian economy. The assessment in Israel three months ago was that without an agreement, the regime collapses by the end of 2026. At the end of 2025, millions of protesters stormed the streets of Tehran, Qom, and Bushehr over a devastating economic situation. It has only worsened since. The damage from the Epic Fury operation alone is estimated around $400 billion.

One might say the barrier of fear went up after the massacre of some 7,000 to 32,000 protesters. But fear and having nothing to lose are competing in Iran as we speak. Say the one-month MOU postponed the collapse to January 31, 2027. Watch the streets of Tehran closely.

For Israel, no war, no deal is the perfect solution: no casualties, international flights, and the regime slowly fading away. But Israel is not an actor in world energy prices. President Trump is, and for him that means fights over the Strait of Hormuz.

So you’re more bullish than the market on regime change. What’s your number by the end of the year?

I would give it 30%. There are two options: top-down or bottom-up. We have the image of the velvet revolutions in Eastern Europe, but even the Islamic Revolution in 1979 began with mass protests and ended with a faction of fundamentalists taking the regime. We now know that Israel and the US work simultaneously top-down, with former President Ahmadinejad, and bottom-up through propaganda warfare. The main idea is that protests would lead parts of the regime to acknowledge the gloomy reality and come to terms with the United States.

Israel is not being targeted and not shooting in this round of fighting. What could pull Israel back in?

Three options. One, an Israeli-American decision to attack Iran’s energy facilities. Two, the same decision on the uranium. Three, the regime feeling its end is near and seeking legitimacy by attacking the Zionists. The first two are more probable than the third.

Polymarket recently listed markets on Qeshm, Hormuz, and Kharg islands. Will Iran keep control over Hormuz, regardless of what happens with the regime?

No matter which regime is in Iran, the Strait of Hormuz is going to be international. The world has seen the effects of taking over a relatively small strait, and no one is going to let it happen again. This is the conclusion of every US ally in the region, mainly the UAE, Saudi Arabia, and of course Qatar, whose economies took amazing damage.

Two processes are coming. One is alternative routes, be it Jordan-Israel-Mediterranean, Syria-Turkey-Mediterranean, or Syria-Turkey-Europe. And second, even a new regime in Iran will go through what Venezuela went through: you can keep your sovereignty, but your oil is controlled by us.

Could a naval operation unblock Hormuz?

I think it is considered. I’m not an expert here, but I keep hearing rumors about it. Such an operation would not include Israel, which has no interests in the Strait of Hormuz, unlike the uranium question. It could happen even without any Israeli knowledge.

Follow Amit Segal on X and on his Substack, It’s Noon in Israel.

Follow all Israel Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

By Andrew Asani*

Today, odds on Polymarket for the Houthis to close Bab el-Mandeb by 30 September 2026 stand at just 14%.

In my last piece for The Oracle, I argued that the Houthis were unlikely to close off this critical Red Sea chokepoint due to their increasing financial dependence on Saudi Arabia.

Things have changed.

Share

Based on my recent conversations with several Houthi insiders, I believe Polymarket is significantly underestimating the odds for a Bab el-Mandeb closure today.

Despite the potential risks, closing the waterway remains the cheapest and most effective way for the Houthis to achieve their primary aim of forcing the US and its allies to de-escalate against Iran, at a time when most other options have been exhausted.

The Houthis support de-escalation as US-Israeli strikes on Iran have severely weakened the logistics infrastructure of the country’s Islamic Revolutionary Guard Corps (IRGC), the hardline paramilitary that oversees Tehran’s proxy network and serves as the Houthi’s top arms and fuel supplier, financier and patron. Without IRGC support, the balance of power within Yemen would shift in favor of the Saudi-backed Yemeni government, as the Houthis are currently too weak to confront their domestic enemies by themselves.

The Houthis previously sat out prior rounds of US-Iranian warfare in June 2025 and February-April 2026 as the stakes then were not as high. Following a ceasefire on 02 April 2022 that ended the 7-year civil war between the Houthis and Saudi-backed Yemeni government, Riyadh quietly began providing the Houthis monthly payments of roughly $100m to ensure quiet on its border. The Houthis became dependent on these transfers and in June 2025 were reluctant to launch attacks against US and allied forces that could provide Saudi Arabia a pretext to cut off payments.

At the time, the Houthis were caught between two rival patrons: Saudi Arabia, which they relied on for financing, and Iran, their main weapons supplier.

However, the expansion of US airstrikes in Iran to include energy and civilian infrastructure along with steps taken by the Trump administration signaling intent to seize key Iranian islands and even carry out a ground invasion—which the White House has instructed the Pentagon to develop plans for—have shifted the Houthis’ calculus and made the group more responsive to Iranian demands for assistance.

A member of the Houthi Political Bureau told me that Iranian operatives have deployed across Yemen and are driving the current escalation in the Red Sea, adding that the Houthis are following IRGC requests because continued aid is contingent on meeting certain Iranian directives.

That said, should the United States continue to prosecute the war and remain undeterred by Iranian counter measures the Houthis will likely resume attacks on commercial shipping as a measure to help protect Iran and ensure continued access to IRGC patronage.

A relative of the Houthi’s Jihad Council—the group’s top decision making body—claimed that in the event of an expanded US-Israeli campaign in Iran, the Houthis would escalate their attacks to include shipping in Bab el-Mandeb as deterrence to prevent further US strikes and the collapse of IRGC logistics networks.

That the Houthis would do so appears plausible in light of reports that IRGC operatives in Yemen have begun preparations to develop a Red Sea “authority”—similar to Iran’s newly created Persian Gulf Strait Authority —to extract tolls from ships in the Red Sea.

Shuttering the Bab could trigger the opposite response: a vicious US counter campaign which the Houthis would not be able to withstand. However, opinion polls showing overwhelming opposition to the war within the United States and reports claiming that American interceptors and munitions stockpiles have been depleted below levels previously thought cast doubt on such a prospect.

That said, closing Bab el-Mandeb remains a low-cost method for the Houthis to force the United States to de-escalate against Iran and pressure Saudi Arabia to hold off support for anti-Houthi forces that have recently increased their attacks on the group. Beginning on 20 July 2026, the Houthis attempted to deter Saudi-backed anti-Houthi forces in Yemen by imposing a blockade on Saudi ports and conducting symbolic strikes on energy infrastructure.

However, these strikes failed to achieve the desired effect, instead triggering violent Saudi and Yemeni government reprisals, the first major act of hostilities between both sides since the 02 April 2022 ceasefire. The Houthis have responded by deploying large-scale reinforcements across the country, including to the southwest coastal province of Ta’iz, overlooking the strategic waterway.

Subscribe now

Cracking at the Seams

My senior Houthi contacts describe a group unable to afford renewed full-scale war with Saudi-backed Yemeni forces, following renewed pressure placed on the group by the second Trump administration. The 2025 US-UK Operation Rough Rider and subsequent Israeli strikes degraded Houthi leadership, weapons stockpiles, and port infrastructure, prompting the Houthis to reduce their Red Sea shipping attacks—especially after the October 2025 Israeli-Hamas ceasefire—and cutting the group’s revenue.

Expanded US sanctions further squeezed the group’s income from customs and limited their access to fuel needed for missile production. Financial strains in early 2026 triggered several waves of mass defections and three leadership purges from April-June 2026 over alleged Saudi espionage. Tribal unrest, including July 2026 clashes with Daham tribesmen in Jawf governorate over land disputes, further destabilized Houthi control and prompted defections.

These pressures have made the Houthis more receptive to Iranian escalation calls, hoping for greater support. On 17 June 2026, the US provided Iran a temporary sanctions waiver to sell petroleum products, enabling Tehran to net roughly $5bn-6bn before US-Iran hostilities resumed on July 11. During this period, Iranian funds flowed to Hezbollah and the Houthis, including at least two Iranian arms shipments intercepted by Yemeni government forces on 29 June and 14 July 2026—the first in 2026—that contained heavy drone/missile manufacturing equipment.

Leave a comment

What To Watch

After more than a year of electing to sit out fighting between US and Iran due to fears of antagonizing Riyadh, it now appears the Houthis are leaning closer towards Tehran. As the group becomes more desperate, it will continue to do so. However, the Houthis still seek to avoid the type of escalation in Iran and Yemen that can degrade the IRGC and Houthis themselves enough to embolden Saudi Arabia and invite a resumption of full-scale war.

That said, for the foreseeable future, Houthi attacks against Saudi Arabia will mirror US escalation in Iran itself. Should the United States follow through with plans to seize strategic Iranian islands, prepare a ground invasion or expand its target list to include a wider range of civilian and energy infrastructure, the Houthis will likely have little choice but to play their final card and close Bab al-Mandab.

The best sources to follow for updates include the daily “Iran Report” by the American Enterprise Institute’s Critical Threats project, which does a great job at cataloguing every kinetic strike in the region. For minute to minute updates, great X accounts to follow include @OSINTdefender, the Yemeni @SanaaCenter think tank, and Times of Israel correspondent @manniefabian.

* Andrew Asani is the pseudonym of an OSINT-HUMINT researcher who tracks extremist patronage networks across the Middle East, including Syria, Yemen, North Sinai, Libya and Sudan. He advises clients performing strategic advisory and litigation support, and his work has been featured in numerous academic journals and news outlets.

Browse all Yemen Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

Polymarket’s Institutional Desk is facilitating block trades for JetBlue bankruptcy event contracts on both our U.S. and international exchanges. Contact Brooke Rizzetto at institutional@polymarket.com for onboarding.

By

Warren Buffett famously said that airlines have devoured capital since the Wright Brothers:

Since the 1970s, the US airline industry has produced almost 160 bankruptcies. The problem with airlines from an equity or credit perspective is that they are thin margin businesses with high fixed costs and volatile variable costs. The business model is very dependent on scale. Which is why, for these past few decades, the airline industry typically cannibalizes itself, as the acquirer aims for increased scale and the acquired is in financial trouble from not having it.

JetBlue, over the last couple of years, tried to execute this rollup strategy. Up until last year, they put in a good-faith effort to buy Spirit Airlines. That was struck down by antitrust regulators and Spirit went on to declare bankruptcy in 2026.

This failed move by JetBlue has left them in a very awkward place. The financial markets are saying this with 5y CDS on JetBlue bonds (chart below) trading near 1000bps, but the left tail on Polymarket is still quite cheap. Which is interesting in light of a very tricky backdrop for JetBlue.

Without the Spirit acquisition, JetBlue is stuck in between the big guys (Delta, United) and the budget airlines. They are too big to be a discounter but they are not big enough to compete with the network economics (pricing, routing etc.) of the big guys. And the big guys have been able to move up the pricing power curve with premium traffic.

This imbalance can be seen in the recent labor talks going on at JetBlue. United and Delta crews just landed deals that were 34% bigger than what JetBlue is currently offering its crews. That shows that JetBlue cannot match the market.

The financial picture right now is hairy for JetBlue. They have $9.3B in outstanding debt, with most of the fleet already pledged as collateral and a CCC credit rating. Their interest costs alone are running at $800m a year. They have very little room to maneuver, and it doesn’t seem like a sale is viable now.

Within this very tight operating window, the headwinds to the business more tactically have increased in recent months. Jet fuel prices have surged following the war in the Middle East. They have negative pressure from Pratt and Whitney engine issues disproportionately impacting JetBlue’s Airbus-heavy fleet. And, they are the most exposed large carrier to fall weather disruptions, as they have a heavy Florida/Caribbean fleet concentration.

This could bring forward some interesting tactics around managing their debt position. It is possible that declaring chapter 11 early, could make it less likely that they have to declare chapter 7 later on. This would keep the company operating instead of it being forced to liquidate like what happened to Spirit.

An interesting trade on Polymarket right now is that this left tail of JetBlue, that financial markets are pricing quite seriously with CDS near 1000bps (10% notional per year to insure against default), is only modestly priced in prediction markets. In “Which airlines will announce bankruptcy by December 31, 2026?” “Yes” for JetBlue is around 4%. That seems like an interesting opportunity in a relative value sense to traditional credit markets but also given the left tail dynamics in the business over the second half of the year.

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

Polymarket’s Institutional Desk is scaling up. For information on block trades, market making, and commercial data sales on either our U.S. or international exchanges, contact Brooke Rizzetto at institutional@polymarket.com.

By

An interesting trade to me in the macro space, that can’t really be expressed in financial markets at the moment, is that I think it is highly likely that Lagarde leaves her post as President of the ECB before the end of the year. To be more specific, I think she announces her intention to leave her post before the end of the calendar year.

Something that entered the ECB halls earlier this year was the idea that Lagarde could leave her post before her term ends in October of 2027. The reason behind this was twofold:

First, I think Lagarde probably wants to create optionality for Macron before the volatile French Presidential election in April of next year. Lagarde would likely be more comfortable with Merz and Macron being the influential voices in deciding the ECB’s next president vs. whoever in France will come after Macron in the spring of next year. Lagarde is a down the center defender of the European Union in a traditional sense, I think the chances of a less orthodox French President deciding her successor, gives her pause.

Second, there is also a chance that Lagarde actually wants to be part of the French Presidential election herself. Lagarde is a master politician, and has evolved from French finance minister to head of the IMF, to now president of the ECB. It is unlikely that she will directly run, but I think the bar to her wanting to be part of the conversation is very real. She is very much a Carney type figure who went from famous central bank governor to Prime Minister of Canada last year.

So where do things stand now?

Post the start of the Iran war, Lagarde, when asked about leaving early, said, “a captain doesn’t leave the ship mid-storm.” This saw the odds of her leaving before the end of 2026 on Polymarket drop from 30% to 17%.

However, in recent weeks, this trade has come back to life with Lagarde saying in an interview with Reuters that an early exit was “possible.” This saw the market move from 20% to close to 60%.

And then came the ECB press conference yesterday following their July meeting. Lagarde, when asked if she was going to leave, said “you will not see the back of me before 2027.” This saw the market price the odds of her leaving by 2026 plunge to 25%. Where I think it is now a very good trade. Lagarde said something interesting, she said she wouldn’t leave “if” there are clouds on the horizon (a reference to her captain comments in March), and she left it very open to leaving after 2027. This leaves plenty of room for her to announce she is leaving before 2027, even if she actually only vacates the seat in 2027.

I think the market post this comment, still is underestimating the likelihood that Lagarde announces her intent to leave her post before the end of 2026. I think the two reasons related to the French election in April have become more relevant, not less. The acute stage of the crisis in the Strait of Hormuz is less pronounced now, and while it does not mean the ECB is done raising rates, the level of urgency has come down quite considerably from her March “captain” comments. The market currently views this trade as a 25% chance, but I think conditional on economic volatility remaining in the range of the current status quo (2-3 more hikes priced for this year), the odds of Lagarde announcing her intention to leave her post early by the end of the year should be closer to 70-80%. And it is all centered around the French elections in April.

With Le Pen the current favorite on Polymarket to replace Macron, and the RN polling ~35% ahead of a fractured left and center heading into the 2027 presidential race, French political risk remains. This is something that Lagarde seems particularly mindful of.

I think a key part of this trade is the language in how the contract will resolve:

An announcement of Lagarde’s resignation/removal before this market’s end date will immediately resolve this market to “Yes”, regardless of when the announced resignation/removal goes into effect.

Lagarde will want optionality for the French elections while giving the ECB the proper time to make this adjustment. This sets up for her to announce she is leaving in Q4 and actually leave in Q1 of 2027. The announcement is enough for the market to resolve “yes.”

Overall: To me, based on how relevant the French elections are for her in April of next year, I think the odds that she decides to leave the ECB early are much closer to 80%. And only that low because I do believe that she will not announce her departure during a time of heightened volatility, which given the recent dynamics in energy markets, is still a very real risk.

Leave a comment

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

🇺🇸 Midterms are LIVE on Polymarket USA.

New users get a $20 bonus with $10 deposit.

Download now with code “ORACLE” to claim your bonus.

Werner K. Zagrebbi is a board-certified Superforecaster™ who writes about politics and economics at Right Rationalism.

Graham Platner dropped out of the Maine Senate race on July 10th.

This probably isn’t news to you. You’ve surely tired of the panoply of takes that have been bandied about after his withdrawal: Was that symbol on his pectoral associated with Imperial Germany? Is it cringe to be on Kik after the age of 25?

But there is an important question left unanswered by all these takes: was this a good decision for his party?And after this kind of scandal hit their nominee, wouldn’t you expect the Democrats to be less likely to win in Maine? I certainly would. And yet, on the same news, the Democrats’ chance of winning that Senate seat went up, from 54% to 63% in a couple hours - days before Platner bowed out of the race.

THIS is the real Platner story!

Platner vs. Arrow

GMU’s Tyler Cowen went so far as to rank Kenneth Arrow as the fourth greatest modern economist, after Hayek, Keynes, and Milton Friedman. Arrow’s most important contribution was to “build up economics from a theory of ‘state prices,’” according to Cowen.

What does this mean? We can think of every possible world as a separate “state.” A state price is what you would pay today for a dollar that arrives in exactly one of those states and no other. Naturally, the more likely you consider that state, the more you will pay for a dollar that “lives there.” Thus each possible future carries its own price, and those prices behave like probabilities.

This kind of thinking may come naturally to Polymarketeers, but its implications are not necessarily intuitive.

Subscribe now

Consider Sergeant Platner. The price of “Democrats win the seat” is essentially a weighted average of two possible worlds: Democrats’ odds of winning the Maine Senate seat if Platner goes, and their odds of doing so if he stays, each weighted conditional on the likelihood of it occurring. The Platner accusations, in this reading, is an “exogenous shock” that moved the weights. Said Alex Tabarrok of Marginal Revolution:

The market is signaling that Platner reduces the Democrats’ chances of victory. We can be more precise. If an 87-point increase in the probability of dropping out gets you 9 points of winning, then a 100% chance of dropping out implies a gain of 9/0.87 ≈ 10.3 percentage points. Thus the market’s best estimate is that Platner is reducing the Democrats’ chance of winning by about 10 percentage points (compared to an unknown replacement). That’s a pretty big number!

It was a pretty big number! And thus Platner was probably right to drop out: That decision, at the time, made the universe where a Democrat is Maine’s senator 10% more likely. The scandal hurt Democrats, but Platner’s withdrawal helped them more.

Share The Oracle by Polymarket

Watch The Unlisted Market

The lesson for traders here is to keep an eye out for the unlisted market. A visible contract is often but one leg of a hidden state-price bundle, and when news changes a price, adjacent contracts don’t always update in the same order. Perhaps write out the synthetic before you trade:

In this case, that would be Democrats’ overall chance of winning = (Their chance of winning if Platner leaves) × (The chance that he leaves) + (Their chance of winning if Platner stays) × (The chance that he stays).

Endeavor to ask, “Have the displayed prices caught up with the implied conditional prices?”

Every Dem name in the race to succeed Platner as Maine Senate nominee has a live probability of being nominated on Polymarket. Their sum is a synthetic “Democratic nominee is X” bundle. If it prices at less than the “Dems win” market divided by a plausible general-election win rate, someone is mispriced.

You can use this trick elsewhere. JD Vance’s chance of being out as VP by December 31, 2026 is about 7%, for instance, and his chance of winning the Republican nomination in 2028 is 41%, so his implied nomination probability conditional on remaining VP is 0.41 / 0.93 ≈ 44%. Rubio is at 28% in the same market, and if the VP-exit market rises by 20 points on some news, but Rubio doesn’t rise 8 points, there is an arbitrage.

The total-cuts-in-2026 polymarket prices the odds of no interest rate cuts at 81%, 1 cut at 15%, and 2 or more at 4%, for expected cuts of about 0.25; the Jul–Oct decision-path market prices Pause-Pause-Pause at 47%, and the two describe the same underlying object, so summing the path probabilities should recover the total-cuts distribution.

Zohran Mamdani out as mayor before 2027 is 5.5%, and his rent-freeze market is 92%, so the rent-freeze conditional on him remaining in office is about 97%. Every one of his policy contracts (millionaire tax, $30 wage, city grocery store) is a compound bet of P(Mamdani still in office) times P(policy passes given he is), so a shock to the tenure market should reduce every policy market by the same factor.

Or consider Israel. The Likud party is priced at 63.5% to win the next Knesset election, while Netanyahu himself is at 34.5% to be the next Prime Minister. But if Likud wins, its leader is who forms a coalition, and the two markets describe the same underlying object. Divide one by the other and you get the market’s implied probability that Netanyahu is the one still leading Likud when the votes are counted: 0.345 / 0.635 ≈ 54%. No Polymarket contract asks that question directly, but the two that do trade can together get you to a price. On any news that resolves it (a corruption verdict, a coalition fight, a health scare), both markets will move together.

You should also take advantage of spanned information in other ways when you can. If you’re a Democrat who wants to know how to donate to races in which it will be maximally effective, Polymarket is what you should be watching.

It’s easy to underestimate the beauty and promise of this idea: Conditional markets offer rigorous evaluation before policies are even enacted. Robin Hanson might go as far as suggesting we use them to decide what the government does, but there are a whole host of other use cases short of that: Central bankers already closely watch market-implied expectations through things like CME FedWatch, and people have proposed going further with NGDP prediction markets or decision markets for monetary policy. And research funders could look to markets for scientific replicability, or to more exotic experiments like Scott Alexander’s impact markets.

Certainly once we know what we want, we should more often ask the Arrow question: What is the market price of the world where such a thing happens, compared to the world where it doesn’t?

Leave a comment

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

This is a cross-post from Polymarket Institutional Research. To receive future issues, sign up here.

Polymarket’s Institutional Desk is scaling up. For information on block trades, market making, and commercial data sales on either our U.S. or international exchanges, contact Brooke Rizzetto at institutional@polymarket.com.

Pre-war, around 20 million barrels of oil flowed through the Strait of Hormuz daily. Polymarket traders now price a 62% chance for Hormuz normalization by year-end, defined as greater than 60 vessel transits recorded by IMF Portwatch on a 7-day moving average basis.

The latest 7-day moving average was 32, just over half the normalization threshold.

Over the past 30 days, year-end reopening odds have trended down, as traders grow skeptical of peace headlines and major differences remain over transit fees (Iran? U.S.?), and continued strikes on commercial vessels transiting the Omani lane.

What’s not priced in?

The chart above shows the realized Brent-WTI spread while the points from September to December are the forward futures curve. The break is where the historical data ends and market implied futures pricing begins. (Note: calculated daily from Brent-WTI settlement prices; conflict averages calculated since 28th February. December spreads are calculated from CME quotes.)

The spread between WTI and Brent oil benchmarks is widely seen as a barometer of Hormuz stress, as Brent is more heavily impacted by Gulf crude prices, while WTI is mainly U.S.-based and has little exposure to Hormuz transit risk.

The Brent-WTI spread shot up from ~$4.39 pre-conflict to over $25 during the war’s peak in April.

On Polymarket, the Hormuz reopening curve gives a live read on where the Brent-WTI spread should be. If the Strait reopens, the spread should revert to its ~$4.39 pre-conflict level. If the conflict drags on, it can stay near the ~$8.59 average since the war began.

Polymarket’s odds can be used to estimate the fair spread. At Polymarket’s 62% reopen and 38% prolonged closure, the fair spread is calculated to be ~$5.99.

Fair spread = [0.62 x $4.39 (pre-conflict spread)] + [0.38 x $8.59 (prolonged conflict)] = ~$5.99

Today, the December Brent-WTI spread trades at ~$4.90, well below the calculated fair spread. Should the year-end reopening odds continue to drop, the logical trade is to long the Brent-WTI spread near $4.90, using ~$5.99 as the fair-value anchor, with more upside should the reopening odds fall further.

All Iran Polymarkets

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Werner K Zagrebbi is a board-certified Superforecaster™ who writes about politics and economics at Right Rationalism.

Nina Totenberg is NPR’s legal affairs correspondent. At 82, her resume includes breaking Anita Hill’s harassment allegations against Clarence Thomas, reporting on secret Court deliberations during Watergate, and exposing that Douglas Ginsburg had smoked marijuana, which ended his chances at nomination.

And so at 10:51 AM ET on Tuesday, June 30th, when National Public Radio published a story headlined “Justice Samuel Alito Retires”, the markets took her seriously. The polymarket on Alito announcing his retirement by December 31, 2026 was trading at 37.5%. Nine minutes later it had shot up to 88%.

Sixteen minutes after publication, however, NPR retracted the story. By evening, $552,265 of trading volume later, the contract had settled at 29%.

Subscribe now

Did the market behave well?

Domer (aka ImJustKen), one of Polymarket’s most followed sharps, certainly handled it well. His wallet sold 11,314 shares of YES at $0.902 at 11:04 AM, four minutes after the peak and three minutes before NPR pulled the story. He then bought 18,690 shares of NO at $0.579 twenty-five minutes later—for a profit of ~$10k.

Two things had to happen in the four minutes between NPR’s publication and Domer’s sell. First, someone had to look at the article and notice it was structurally broken. Second, someone had to check the actual resolution source, which is the Court itself.

Both were public in real time. On Bluesky, six minutes after publication, the legal commentator David Lat asked whether SCOTUSblog had received a release from the Court’s Public Information Office. Mark Walsh, physically in the Court press room, answered that it had not. That thread was live at 10:57 AM. Ken sold at 11:04 AM.

Leak Bait?

A careful reader of the Totenberg article contained more warning signs. Take a look—the caption to the leading photo says that Alito “retired Friday,” while the first paragraph claims his retirement was “announced Tuesday” (which—at the time of publication—was today).

It’s grammatically possible that NPR meant “Alito had already retired Friday, and the Court announced it Tuesday,” but given the context it’s implausible.1

The article also quotes a Yale Law professor as saying that Alito “will forever be remembered” as the justice who “took sown Roe v. Wade” instead of “took down”—which is not the kind of copyedit that usually makes it through NPR.2

Many have argued that this story was an “embargo breach” or that Alito planted the story to catch a leaker. I think it’s worth taking off the tinfoil hat on this: A real embargoed story would probably not have an egregious contradiction between the lede and leading photo’s caption. If NPR had been given advance notice by the Court or some political insider, the basic timeline would not read like it had been assembled from three incompatible drafts.3

Share

“Rookie Mistake”

But it’s hard to fault the rest of the market for missing these clues.

Nina Totenberg is a highly respected 82-year old journalist. And as has noted, The Media Very Rarely Lies. Crazy, I know, but it’s true! In fact, you have to go pretty far back in history to find a precedent for a mistake of this magnitude in a mainstream outlet4—a frequentist would certainly buy up.

And Supreme Court reporters, in particular, have traditionally commanded a lot of power and respect. America’s favorite economist Tom Sowell even made the case that they have a massively outsized impact on legal outcomes: It’s well documented, after all, that while “Supreme Court justices swim slowly, they tend to swim to the left” — ie, they tend to vote more with the left over their tenures. Desire for favorable press coverage from the likes of NYT’s Linda Greenhouse, Sowell argued, may well be what’s going on, under a model he called the “Greenhouse Effect.”5

Wake Up Call

Polymarket has a separate market, Supreme Court vacancy in 2026?, which applies to any justice creating an opening. It ran up alongside the Alito contract, peaked at 60.5% around 3 PM, then went back down, but only partway. As of this writing it trades at 41.5%, up 17.5 percentage points on the week.

If you back out the Alito-specific probability at 27.5% under a rough-independence assumption, the market is implicitly pricing ~18% chance of a 2026 vacancy created by someone other than Alito, which is notable—that premium was near zero seven days ago. So what’s going on?

Well, as Fisher Black taught us, noise traders are an effective subsidy to smarter traders; they give them a reason to correct the price.

The any-vacancy market has $18,576 in lifetime volume. In the entire week before the Totenberg story, it did $37 in trading and seven trades—essentially dormant. In the sixteen minutes the NPR story was live, it did $5,019 across seventeen trades. The false NPR signal traded a hundred times more volume, minute-for-minute, than the market had seen all week. The scare gave the market a jolt of noise that woke it up: In a flash, it became worth investing the effort to discover the right price.

In mid-term calendar years since 1980, at least one justice has retired or died six of eleven times, a 55% base rate. Pre-Totenberg, the any-vacancy market was already at 28.5%—below that. Certainly the post-Totenberg price is more consistent with these base rates, and with the brand-new Clarence Thomas retirement contract.

Share The Oracle by Polymarket

What to Watch

Supreme court Justice is a dangerous profession: Three died in office. Of the seven who retired, six announced between March and July, with the announcement clustering at the close of the term. The seventh, Breyer, was leaked by CNN in late January, and formalized on the following morning.

In any case, I’d watch any communication out of Thomas’s chambers, clerk-hire announcements for OT26 (Justices who plan to retire stop hiring), and retirement timing signals in July, when Justices are wont to drop these announcements at the close of a completed term. The info on hires disperses entirely informally, but if you were to make a bot David Lat of Original Jurisdiction is the place that’s known for reporting earliest. The indefatigable Scotus Blog also should have an inkling.

And—theoretically—such considerations would be priced into markets on live cases. Polymarket has an active market on whether the Court accepts the sports event contract case, currently pricing 50.5% for December 31 resolution. Retirement expectations flow into case-acceptance expectations. Watch the pair.

These contracts are functionally binary options. Their prices naturally decay toward zero as expiration approaches—but that decay isn’t smooth. Discrete calendar events like end-of-term rulings and clerk-hire announcements create predictable moments of potential repricing. A trader who models these markets as pure theta is leaving alpha on the table for someone who knows when the term ends, so keep it in mind.

The Greenhouse Effect, if it exists, took decades to move a Justice by a few ideological points. The Totenberg Effect, if we want to call it that, took sixteen minutes to move a Polymarket contract by fifty, and to effectively subsidize another to a much more correct new price.

Maybe Tom Sowell is right and the reporters are just that powerful. If nothing else: It’s worth checking if the first page has errors before you trade off events: Tell your bots!

Follow Supreme Court Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.

Today we are launching Polymarket Institutional Research, a publication exploring the intersection of Polymarket with the global financial system.

The first issue is being cross-published here on The Oracle. To receive future issues, sign up here.

If you are a financial institution looking to engage with Polymarket, our Institutional Desk is scaling up. For information on block trades, market making, and commercial data sales on either our U.S. or international exchanges, contact Brooke Rizzetto at institutional@polymarket.com.

Anthropic Right Tail Underpriced?

By

It is rare for single companies in the $35T U.S. economy to become themselves a macro variable. Anthropic is one now. Their ARR growth, capex spend, and model impact on productivity all have multipliers on the U.S. economy.

An interesting trade to me is looking at the strip of the “Will Anthropic valuation hit _ by December 31st” Polymarket contract.

The interesting thing is where the market prices the convexity point and how the market prices the right tail in the context of the recent run that Anthropic’s valuation has been on.

Valuation has tripled just in the last six months and the revenue has backed it up. Not only have we gotten stories about ARR going from $9B to $45B, but Anthropic showed that they are already profitable this year despite massive capex intensity.

Fable Inflection?

Last week, the U.S. government reversed their ban of Fable 5, which has cross-cutting impacts on the forward valuation curve. On the one hand, this development could be bearish Anthropic’s valuation based on the fact that it will be harder to release new models going forward.

On the other hand, it is itself evidence that the model is a genuine capability leap rather than an incremental release, which is exactly what a step-function valuation jump needs. The other factor here is, as these models improve from a very high baseline, it also raises the chances of regulatory capture which would be very beneficial for Anthropic’s valuation in a world of increased model competition.

In this context, the strip has a strange shape. There’s a hard cliff exactly at the $1.5T mark, then the upper tail flattens, which in the context of the recent growth and where SpaceX went public, is surprising to me.

This reads as a mixture of two distributions: a tight cluster around the consensus round print, then a smoother “something else happens” regime above $1.5T. Given the trajectory of growth and what the current risk sentiment is in the AI trade, thinking that Anthropic doesn’t add another 50-75% of its valuation this year feels like a 50-50 at worst to me given the current non linear growth in ARR. Especially with an IPO kicker likely coming towards the end of the year.

In that context, Anthropic at a valuation of $1.75T or above by year end at 40% seems too low. And the convexity point where the market hands off from mode to tail seems too close to the spot valuation, even if that number is +50% from here.

Get Polymarket Institutional Research

Related Polymarkets

Disclaimer

Nothing in The Oracle is financial, investment, legal or any other type of professional advice. All odds are time-sensitive and subject to change. Anything provided in any newsletter is for informational purposes only and is not meant to be an endorsement of any type of activity or any particular market or product. Terms of Service on polymarket.com prohibit US persons and persons from certain other jurisdictions from using Polymarket to trade, although data and information is viewable globally.