Foresight Games Design Patterns

Foresight games draw on a wide breadth of design patterns, including: board games, role playing,
larp, event planning, and much more. The right set of patterns for your design depends on your
exercise’s goals, the participants’ expertise, and constraints around the event (attendee counts,
space, duration). This guide will provide an array of options you can use in crafting your exercise.

Why games?
Intuition via Mechanics
Empathy via Roleplaying
Creativity via Constraints

Encourage participation
Empower communication
Support workflows

What makes something a foresight game?

Games come in many different forms and we take an inclusive approach toward what gets
included. There’s a spectrum in the design space from clearly-a-game to non-game-workshop.
Ultimately, if it has elements of play, role-playing, worldbuilding, or participants acting to win,
we’re happy to call it a game. A foresight game, in particular, means that play involves grappling
with uncertainties around possible futures. Even if some of the patterns and examples here don’t
fit your definition, they still might help you craft stronger games.

Types of Patterns

This guide contains several categories of foresight game patterns:

●​ Player Activities – what actions players take during the game
●​ Structure Patterns – specific stages of the event, that might include:
○​ Welcome and onboarding
○​ Organizing the participants into different roles and groups
○​ Multiple rounds of participant activity
○​ Facilitator updates
○​ Adjudication
○​ Model of the setting and other forms of state
○​ Debrief
●​ Deployment Patterns – when and with whom to use different types of games
○​ When
 ■​ Brainstorming
■​ Assumption and model testing
■​ Soliciting input
■​ Communicating insights
●​ Building intuition
●​ Generating empathy
○​ With whom
■​ Core team
■​ Experts
■​ Key decision makers
■​ End users
Key Activities
There are a broad variety of activities that you can have participants engage in during the course
of an exercise. For most games, it’s best to limit the number of activities types to just one or two, to
reduce facilitation complexity, participant learning curve, and general cognitive load. This section
contains potential activity types, though the list is endless.

Note: open-ended activities take longer for groups to reach consensus and you should budget time
accordingly. Regardless of how much time you budget, participants will always want more time.
For somewhat open-ended activities done by groups, 10-30 minutes per round of activity time
seems to be the sweet spot, depending on activity complexity.

Open ended interventions

Each round, each participant (or group) has to submit the most important intervention(s) (i.e. most
effective at driving toward their goals) that their Role makes during the time period of the round.
Open-ended prompts create space for a wide variety of answers and generating ideas that
facilitators did not predict. This can be entirely open ended or can be scoped in various ways.

Example: in Threatcast: 2020 we had participants submit two interventions each round. All
interventions had to leverage technology and one had to be practical (low-risk) and one had to be
experimental (high-risk).

However, the blank canvas of an open ended question can intimidate participants. Good support is
critical and examples can go a long way. You can also support players by providing optional building
blocks as inspiration. For instance, in an intervention game you might provide lists of Targets,
Tactics, and Goals so that a player can quickly assemble an intervention by grabbing one item from
each category.

Advice: if you worry about players defaulting to obvious or uninteresting interventions, you can
explicitly list out some answers that aren’t allowed.

Specific questions
Another approach is to ask narrower, specific questions that participants need to answer. If the
participants are organized into groups, then the questions could be tailored to each group.

There are tradeoffs to this approach. A specific question can be more approachable for
participants than an open ended prompt for an intervention; however it also might prevent the
participants from generating valuable insights that fall outside the scope of the questions.
Single-use abilities
In addition to the normal activities, you can also give participants single-use abilities and
dramatically affect the game. For example, in Threatcast 2020, we gave one group the ability to
carry out a massive hack or data breach on another faction, usable only once during the game.

Shopping
In this activity pattern, players or teams get a budget that they can spend on items, abilities, or
investments. The budget can be in the form of real world currency, tokens, or something else all
together. They might purchase items from a fixed menu or describe what they want to buy and
have a facilitator set a price. The item being purchased might give a one-time ability, an ongoing
ability, or might unlock new mechanics. There might be a lag between spending the money and
seeing the effect of their purchase.

Example: in Machine Learning President, players could spend their money on speculative
technology investments that increase their capabilities or at an advertising station where they
could sway blocs of voters.

Advice: consider using poker chips to represent available funds as it is easy for players and
facilitators to handle, track, and trade.

Negotiating
Negotiating can take many forms but involves players making deals with each other or with
facilitators (often in non-player roles).

Negotiating works best when the following elements are in play, to enable win-win deals:
●​ The players have different priorities and value the elements being negotiated differently
●​ Having multiple categories of things that can be negotiated for (e.g. cash, technology,
public support)

Consider having a way to track what deals have been made, so that facilitators can understand
what happened after the game if they were not present during the dealmaking.

This pattern compliments many of the others here, such as negotiating, shopping, and coalition
forming.

Coalition forming
In this pattern, players form alliances to accomplish their goals. This coalition might be explicit,
with a declared alliance, or might be de facto and emergent as players help each other out when
their aims are aligned.
Giving speeches
In this pattern, players give a speech in-character. A game might have a specific time set aside for
speechifying, where play stops and is directed at a podium. Speeches are often an opportunity for
fun!

The speech is an opportunity for players to communicate their actions to others, summarize what
has been decided, or signal intentions. You might structure the game so that an action only
happens if it is said allowed in a speech, potentially simplifying work for the facilitators.

Example: in Machine Learning President, candidates had to give short stump speeches during
rounds. Any policy planks that they endorsed during the speech would then sway voters.

Advice: keep speeches short, perhaps just a minute or two. Consider giving players support in
creating their speech such as potential talking points or key phrases.

Narration and Storytelling

With this pattern, players create and share stories. The players do not need to take on the roles of
their stories’ characters but there could be a frame story (e.g. players are writers on a show
inspired by Black Mirror).

A vast array of formats could be used in the storytelling:

●​ Write a series of headlines to create a story (e.g. Hindsight 2030)
●​ Fill out a template that contains narrative beats (e.g. Premise)
●​ Write or outline a short story
●​ Write journal entries from a character’s perspective

Influencing non-player characters

In this pattern, players seek to influence non-player characters (e.g. played by facilitators or
otherwise modeled). Players might try to influence them through formal mechanics (e.g. explicit
promises, payments, or paid advertising) or through persuasive dialogue with the character. The
influence could be targeted on one specific character (e.g. the CEO of a company) or a broad
demographic (e.g. progressive voters).

Advice: try to have a clear feedback loop where players can quickly see if their attempt at
influence had any impact.

Multiple choice dilemmas

In this pattern, players can choose a response from multiple options. Certain options may be
available or blocked based on player resources or past decisions. Depending on the game, you can
include an “Other” option that lets players come up with their own responses. Alternately, you can
have the responses limited but then ask players to embellish their response with specific details.

Betting
Players place bets on certain outcomes either via an in-game prediction market or financial
exchange (e.g. stock exchange, commodity exchange). The markets can update over the course of
the game and players can see the results of their bets.

Worldbuilding
Worldbuilding is a broad category of activity patterns. Generally, worldbuilding activities are
focused on creating a coherent vision of one specific moment and location in a possible future. This
is likely done out-of-character.

The following are example worldbuilding activities:

●​ Artifact Creation / Design Fiction: players create objects from the time and place they are
exploring (ex: The Thing From the Future)
●​ Institutions: players describe institutions, organizations, and movements that have power
in a possible future.
●​ Character creation: players explore possible futures by creating characters. The character
might be very similar to the player (e.g. selfcasting) or might be randomly generated from
cards or dice. A generated character might then be used with a different pattern (e.g.
Giving Speeches, Narration).
●​ Timelines: players create timelines that lead to a specific future being explored (ex:
Microscope, Hindsight 2030)
●​ Cities: players focus on a specific geography which they build out or update in depth. Cities
are a great way to explore a diverse array of trends.
Structure Patterns
The Structure Patterns are about the different stages of the event. Not every event will have each
type of pattern.

Introduction and Onboarding

Welcome and introduction

●​ Introducing the facilitators
●​ Introducing the participants
●​ Sharing any key information about the space and duration of the event

Goals and Norms

●​ What are the facilitators’ goals for the event
●​ What are the norms for the event:
○​ Privacy about what participants discuss and propose, for example:
■​ No sharing anything
■​ Chatham House Rules
■​ Everything can be shared
○​ Security – do participants need to be careful about what they say, as it may be
reported to bad actors? Are there physical security concerns?
○​ Safety – what else do participants need to know about the venue? Who should they
turn to if there are any issues?
○​ Participation:
■​ Guidance around being good collaborators
■​ Advice on stepping outside of their comfort zone or proposing more
unusual ideas

Game Overview
●​ Game structure and mechanics
○​ Game structure (including duration of different stages)
○​ Participant structure (e.g. roles and groups)
○​ Overview of the model (though this can happen later)
○​ Setting details
●​ Participant instructions and reference materials:
○​ Instructions for key participant activities and actions they can take
○​ Briefing of key information such as goals, assets, approaches, actions (advice:
include more info they they need but make it easily referenceable)
○​ Information about other players’ characters, factions, and key non-player
characters
 ○​ Examples of key activities

Warmup
An opportunity for participants to ease into their assigned roles and groups, such as:
●​ Choosing a slogan for their group
●​ Picking an initial action, perhaps from a list
●​ Giving a short speech
●​ Going through a warmup round (see the Round Structure section for more information)

Participant roles and groups

Taking on roles
While it occasionally makes sense to have attendees participate as themselves (either projecting
themselves into the future or taking on an authoring or worldbuilding role), there can be a
tremendous advantage to having attendees take on roles of the key actors (organizations,
movements, parties, and other actors) of the context that the exercise is exploring. There are a few
reasons for this:
●​ Having one participant (or one group) focused on a key actor means that the actor’s goals
and resources will always be taken into account as the exercise progresses
●​ Key actors, when embodied by players, become a dynamic part of the exercise which forces
other participants to factor them and allows the exercise to examine how they might react
to an array of actions and dynamics
●​ It gives participants an alibi to propose provocative ideas
●​ By taking on a role, the participant can gain empathy and intuition for the key actor they’re
portraying
●​ It’s fun and energizing for participants, when they take on different roles

Types of roles
We’ve run exercises with many different types of roles, including:
●​ A specific politician (or their election campaign staff)
●​ A wing of a political party or an entire party
●​ A social movement
●​ A government
●​ A collection of bad actors and adversaries (including multiple nation states)
●​ A group of aligned business interests

It can be tricky to decide which roles to include and at which level of granularity. When
determining role composition, consider the following questions:
●​ Is each role relevant to the high level goals of this exercise?
 ●​ Does each role have a clear goal that can be successfully pursued during the course of the
exercise?
●​ Does each role have meaningful secondary goals that can still be pursued, even if mid-way
through the exercise the role has no clear path to achieving the primary goal?
●​ Are the roles going to be easy for the attendees to understand?
●​ Would the exercise benefit from having a role split into sub-groups (e.g. splitting a political
party into two roles, representing different wings)?
●​ Is there an appropriate number of roles relative to the number of likely attendees? (see
Individuals vs Groups for more thoughts on the number of roles)

Individuals vs Groups
If you have enough attendees, there are many advantages to organizing them into groups that
collectively play a role (e.g. an organization, political party, campaign, movement) rather than
having roles be played by solely individuals:
●​ Groups can encourage more thorough and diverse ideation than individuals on their own
●​ Groups are resilient to one or two participants that are reluctant, distracted, or who leave
early – this is especially important if the exercise design is dependant on key roles being
and played well
●​ Groups allow for variable attendee count, giving flexibility even up to the start of the event

General advice:
●​ Mix up groups so that participants aren’t with their close colleagues or with people from
the same backgrounds
●​ Encourage participants to take on roles that are different from the role they have at work
or have different ideological or political leanings than the ones they have
●​ We’ve found that the ideal group size is 3-5 participants
●​ While you can assign sub-roles to individuals within a group, we’ve found it perfectly fine
to have all group members being taking the vague sub-role of key decision maker

Selfcasting
Selfcasting is when a player takes a role that is either:
●​ Themselves, but in the future
●​ Someone similar to them today, but in the future (e.g. someone 20 years from now who
matches their current demographics)

Selfcasting can be effective for having a player reach insights that feed into specific changes to
their own life and also to help facilitators tease out insights from a diverse array of participants.

Facilitator controlled roles

If the exercise needs certain key actors to reliably take specific actions, it may make sense to have
the key actor controlled by the facilitators rather than by participants. Some exercises may benefit
from a facilitator playing that role and interacting in-character but it’s often fine for the facilitators
to just mention what the key actor has done.

Rounds and Round Structure

Types of rounds
Depending on the needs of the exercise, you can break a session into rounds in different ways.
Each round could be:
●​ A different period of time (where the time period could be short or long)
●​ Different domains of the context being explored (e.g. political, social, economic)
●​ A focus on different key actors each round
●​ A focus on different types of tactics and interventions each round
●​ Different types of player activities each round

A shorter game might have just one round.

Round structure
A round may include:
●​ An update from the facilitators with new information or instructions, including
consequences from the previous round
●​ The key activity or activities of the round
●​ Participants sharing their output from the activities with each other

Warm up round
It’s often helpful to ease participants into the exercise by giving them a warm up round where they
can explore how the exercise works while the stakes are low. Some thoughts:
●​ You don’t need to call it a warm up round though it may put participants at ease if they
know that their actions in the round won’t be too impactful for the rest of the session.
●​ You can also make this round more approachable by:
○​ Making sure that any impact on game state is minimal compared to the rest of the
game
○​ Focusing the status quo at the beginning of the exercise rather than immediately
having the context in flux
○​ Only including a subset of key actions or mechanics for the participants
○​ Providing pre-selected options for participants to choose, if later they’ll be creating
free-form responses
○​ Giving highly defined short term goals for participants to focus on
Aftermath round
We’ve found it helpful to include an “aftermath” round that explores the time period after the
climax of the exercise. Example: if the exercise explores an election, then consider including a
round that explores what the key actors do in the weeks and months following the election. This
allows participants to explore the short and long term consequences of the actions and results
from the climactic round.

Facilitator Updates
Showing players the consequences of their actions is a critical part of any game. This can be done
via a short intermission in play when all of the players are gathered together and the facilitator
provides an update

Consider including the following:

●​ How much “in-world” time has passed in between updates
●​ Show what elements of the game state has changed and what has changed narratively
●​ Call out how specific player actions led to specific state changes
●​ Introduce additional setting elements as worldbuilding and fodder for future player
activities

Advice: these updates can be more fun if they take place in character, with facilitators playing the
role of newscasters delivering a “top of the news” monologue

Advice: consider preceding the update with players announcing what their key actions were from
the previous round, giving everyone context

Real Time Play

Instead of organizing play into Rounds, you could have play proceed in “real time” with key events
and updates happening at pre-arranged times. This can be effective for a crisis game where you
want to maintain a sense of urgency and time pressure.

Modeling State
For some of our past exercises, we’ve created models that show the state of the setting and update
in response to participant activities. For example, in Threatcast: 2020 we modeled American voter
polling and likely election results, as well as qualitative voter sentiment on other issues.

A toy model can be extremely effective at conveying some of the dynamics of a situation. It also
creates a clear, measurable indicator of the success of participant actions, resulting in a tight
feedback loop between participant activity and state of the exercise.
Victory Points
This pattern draws on European-style boardgames where players can collect Victory Points, often
via a number of different avenues. At the end of the game, the player with the most Victory Points
wins. Victory Points scaffold player activity by giving them a very concrete objective and clear
paths to winning.

Adjudication
Many games benefit from some type of adjudication where player activities are assessed for
success and impact and then used to update game state. Having a tight feedback loop, where
players see the impact of their actions, helps them feel empowered and increases engagement
throughout the game.

Some types of adjudication include:

●​ Deterministic adjudication: if players are choosing from a predefined menu of actions, then
the effect of those actions can be automatically adjudicated by the game model or some
other set of rules.
●​ Stochastic adjudication: the method involves using randomness to determine outcome (e.g.
rolling dice). The probabilities might be fixed or determined prior to resolution by one of
the other adjudication methods.
●​ Expert judges: player actions can be handed off to judges who determine impact or
success. The judges’ decisions can then be fed back into a game model. The experts could
be the facilitators or independent participants.
●​ Peer adjudication: the players’ actions could be adjudicated by the other players or by a
live audience, perhaps with a vote or show of hands.

Debrief
Depending on the objective of your game, the debrief could be as important as the play session,
and could be as long in duration. A well structured debrief allows the facilitators and players to
unpack the events of the play session, surface additional insights, make key learnings explicit, and
capture specific ideas and observations from the players.

The debrief is also an opportunity to set up next steps, both for facilitators and players. If the game
is integrated in a larger workstream, the debrief could be an opportunity to prepare session output
for handoff to the next stage of work.
Part of a debrief could be asking players a structured series of questions. Here are some sample
questions:
●​ Of what we explored today, what would impact your organization or life the most?
●​ If you thought any of the situations discussed today were likely to happen, are there any
specific steps you would take for your organization in response?
●​ What is the biggest opportunity for positive change and what would support that
opportunity?
●​ What is the biggest threat or negative change and what would decrease that risk?
●​ Any other thoughts?
●​ What was most surprising?
Deployment Patterns
See this working doc: When to Deploy Different Types of Games