SS_12_Ch_09 6/6/06 12:27 pm Page 185

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 185

The Role of Notational Analysis in Soccer

Coaching
Nic James
Swansea University, School of Human Sciences, Vivian Tower,
Singleton Park, Swansea, SA2 8PP, UK.
E-mail: n.james@swansea.ac.uk

ABSTRACT
Notational analysis is a technique for producing a permanent record of the
events pertaining to a sporting event and is widely used by sports teams
and individuals of various standards. This paper discusses how notational
analysis can influence the coaching process, giving examples
predominantly taken from published literature related to soccer. The
conclusions derived from these works are critically appraised and
alternative views given. Current issues that are relevant to both researchers
and coaches, such as operational definitions and performance over time,
are discussed using data collected via collaboration with a professional
soccer team. Results are presented in a number of coach-friendly ways
including a novel use of a ‘form chart’ using standardised values for
performance indicators relative to the opposition’s values. Finally, possible
future directions in notational analysis research are briefly considered,
including fuzzy logic and artificial neural networks.

Key Words: Coaching, Notational Analysis, Performance Indicators, Soccer

INTRODUCTION
Hughes and Franks1 state that notational analysis is primarily concerned with the analysis
of movement, technical and tactical evaluation, and statistical compilation. It is a technique
for analysing different aspects of performance through a process that involves producing a
permanent record of the events using codes to represent specific events, thus enabling some
form of analysis to take place. This analysis usually results in a statistical description of
what happened during one event or a series of events, with the complexity of the analysis
and the content of the coding open to choice (but usually determined prior to an event taking
place). Given that one of the main tasks for a coach is also to analyse sports performance,
so that feedback may be given to players and future training sessions planned, it seems
likely that notational analysis would be a useful technique for coaches. To determine
whether this is the case however, it is necessary to demonstrate some added value that
notational analysis brings to the traditional methods (i.e., subjective observations)
employed by coaches. In this respect, Franks and Miller2 demonstrated that international-
level soccer coaches could recollect only 30 per cent of incidents that determined successful
Reviewers: Chris Cushion (Brunel University, UK)
Ian Franks (British Columbia University, Canada)
SS_12_Ch_09 6/6/06 12:27 pm Page 186

186 The Role of Notational Analysis in Soccer Coaching

performance, with significantly better recall of set-piece play than any other situation. This
pivotal research likened coaches’ observations to eyewitness testimony of criminal events
and suggested that reliance on such observations is not only unreliable but also inaccurate.
This evidence has been widely used to promote the need for an objective and reliable
mechanism for recording sports performance, with some form of notational analysis thus
being recommended.
Notational analysis is a technique that is used to derive the information required to answer
a specific question, and the methodology depends on the available equipment. One popular
method, pioneered in 1976 by Reilly and Thomas3, simply involves coding players’
movements into the categories of standing, walking, trotting, running and sprinting. This
information, along with the pitch positions and time spans for each movement, allowed the
calculation of distances travelled and further details regarding work rates to be measured.
Although this particular notation system was liable to some inaccuracies, due to the difficulty
of assigning some movements into a category (such as when trotting becomes running), the
information derived enabled soccer coaches to match training schedules to actual game
demands. High technology solutions, such as heart rate monitors and digital tracking of
players from global position sensors, have more recently been introduced. Essentially, these
solutions give the same information but with more precision. From the coaching perspective
this accuracy may be desirable, essential or perhaps too expensive to implement.
In contrast to the coach, who tends to form an opinion of the frequencies of events based
on recollection, the notational analyst records the details of performance to ensure accuracy.
Arguably the main barrier to adopting a notational analysis approach to performance analysis
is the perceived complexity and consumption of time, due to processes such as coding,
analysing and interpreting the output; all of which cost time and money. However, many
professional sports teams now employ notational analysts under the supervision of the coach.
In order to clarify the role of the notational analyst, illustrate the information gained from
this process, and explore the worth of such endeavours to the coaching process in soccer, this
paper will give examples of conclusions made in published research. Guidelines that
represent good practice will be presented, along with an example of some work currently
being undertaken by the author with a professional soccer team. Quite how technological and
methodological advances will impact on future usage of notational analysis will be
discussed, the current status of notational analysis in sports coaching is summarised.

NOTATIONAL ANALYSIS OF SOCCER – THE BEGINNING

Probably the most influential and certainly the most controversial notational analyst in
Britain was Charles Reep who died in 20024 having devoted over 50 years to analysing
soccer (and other sports) in great detail. Most of his work remains unpublished, but his
legacy remains as the primary advocate of the “long-ball game” or “direct play.” This style
of play has the tactical emphasis on getting the ball into the opposition’s half, in particular
the penalty box, using long passes from defensive and midfield areas. More popular in
England than most other countries, the “long-ball game” is based on the premise that the
more times the ball enters goal scoring areas of the pitch, the more chance there is of scoring.
No team would exclusively use long passes; rather it would use more long passes than a team
that plays a “possession game.” Reep’s first published paper produced the finding that the
structure of soccer is determined by near constants5 and his work is thought to have
influenced many researchers and football coaches to adopt playing strategies based on his
findings. Reep and Benjamin5 compared passing distributions for a large number of
possessions which, when plotted, exhibit a negative binomial distribution (Figure 1).
SS_12_Ch_09 6/6/06 12:27 pm Page 187

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 187

0.5
Proportion (of all Possessions)

0.4

0.3

0.2

0.1

0.0
0 1 2 3 4 5 6 7+
Number of Passes in a Possession
Figure 1. Passing Sequence Distribution for 578 Matches Played Between
1953 and 1967 (Data taken from Reep and Benjamin5)

Small samples do not necessarily follow this distribution (as shown by Reep and
Benjamin for 12 matches played by Arsenal during 1961–2, Table 2, page 582). However,
this detail tends to be overlooked in comparison to the finding that events in soccer matches
are very predictable when large data sets are used. Statisticians would recognise this
phenomenon as the Law of Large Numbers. Not withstanding this observation, Reep and his
colleagues are best known for the findings that: it takes an average of 10 shots to get one
goal5; 50% of goals are scored from possessions that involve one pass or less (zero pass
possessions include penalties and free kicks); 80% of goals are scored from three or less
passes6; and regaining possession in the opponent’s half provides many goal-scoring
opportunities5.
Bate6 has also argued that the more short duration possessions there are, the more
possessions there will be during a game and therefore more chance of making these
possessions into potential goal scoring opportunities. This viewpoint tends to emphasise the
importance of the quantity of possessions in critical areas as opposed to the quality of
possessions i.e., having possession in space is likely to be of better quality (more chance of
doing something positive) than one where the player is marked by one or more players.
Logically, however, situations where no defenders are able to intercept the ball would be
more advantageous than those where defenders are able to challenge for the ball. Clearly,
emphasising chance factors alone does not capture the full extent to which possessions relate
to goal scoring opportunities. Further debate and research is necessary to determine whether
quantity or quality of possessions is the more important determinant of success.
Advocates of “possession football,” which emphasises the quality of possessions in the
critical areas, oppose the long-ball game, with its acceptance of chance factors determining
outcomes. Historically, however, the long-ball game has proved successful. Teams predicted
to not do particularly well by football pundits, usually due to relatively small numbers of
players of the required skill level, have often achieved far better than expected results. The
best known of these is probably the Norwegian national team who have published detailed
accounts of how they analyse their matches in their pursuit of a long-ball strategy in attack7.
SS_12_Ch_09 6/6/06 12:27 pm Page 188

188 The Role of Notational Analysis in Soccer Coaching

Anecdotal reports and views are interesting, but a more scientific approach is needed to
properly evaluate Reep’s work.

A SCIENTIFIC EVALUATION OF REEP’S ANALYSES

Hughes and Franks8 used all matches from the 1990 and 1994 FIFA World Cup finals to assess
whether the findings of Reep and Benjamin5 were still applicable. The results proved very similar
with approximately 80% of goals occurring from possessions containing 4 passes or less (cf. 3 or
less). Nevertheless, Hughes and Franks8 recognised that there were more zero pass possessions
than 1 pass possessions, more 1 pass than 2 pass possessions, and so on. They removed this
inequality by comparing the number of goals scored for each possession length normalized to
1000 possessions (e.g. number of goals scored per 1 pass possession divided by number of 1 pass
possessions multiplied by 1000). As a result of this normalization, the conclusions changed
dramatically. Possession lengths of 3 to 7 passes seemed more likely to produce goals than shorter
and longer length possessions. Clearly the reason for Reep’s finding that 50% of goals are scored
from possessions that involve one pass or less is due to the fact that the majority of possessions
are one pass or less. Taking the values from Hughes and Franks’ paper8, it appears that the ratio
of one pass or less to 2 passes or more possessions is 1.8:1. This may sound an unlikely ratio, but
it can be explained by considering the methods employed in the analysis.
Typically, notational analysts define criteria for allocating events into categories. For
classifying possessions according to the number of passes, it seems sensible to end each
possession when an opponent makes contact with the ball. Consider the following series of
events, however. A possession of 5 prior passes finishes with a cross to the near post that is
intercepted by a defender who only manages to deflect the ball towards the penalty spot and
another attacking player who is thus able to shoot and score. How would this goal be notated?
This depends on the criteria initially defined (i.e., the operational definitions) used by the
notational analyst. ‘End of possession’ could be defined in terms of there being “a touch by an
opponent;” in which case the goal above would be classified as a zero-pass possession. If the
definition of ‘end of possession’ describes a “meaningful” touch by an opponent, however, then
this goal would be described as a six-pass possession. The addition of one word in one
operational definition can have a profound effect on the results of the study. Unfortunately,
details such as these are rarely given in published papers and so caution must be exercised when
interpreting the results of any study that does not explicitly state the operational definitions or
make comparisons between studies when slightly different definitions have been used.
The favoured interpretation of Reep’s findings has been to suggest that the long-ball game
should be adopted. However, as Reep and Benjamin5 stated “an excess of shots by one team
does not mean that, by chance, the other side will not get more goals and thus win the match”
(p. 585). According to Lyons9, Reep has often said that his work has been misinterpreted.
Maybe, therefore, an alternative interpretation is worth presenting. My interpretation is that in
situations where it is difficult to get good quality passes to players in goal threatening positions
it makes sense to try more speculative passes into the goal scoring areas as, by chance, some
goal scoring opportunities will occur. In this respect, Ensum et al.10 found that South Korea
created approximately the same number of shots as Brazil during the 2002 World Cup, but their
inferior shots-to-goals ratio appeared to result from a failure to create good quality shooting
opportunities (possession differences) as opposed to poor shooting ability (skill difference).

FURTHER STUDIES IN SOCCER

Tenga and Larsen11 suggested that Norway played with a “high pressure” tactic for the first
15 minutes of a match; i.e., closing down opponents quickly to try to regain possession and
SS_12_Ch_09 6/6/06 12:27 pm Page 189

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 189

“low-pressure” for the next 15 minutes. In this type of situation if the notational analyst does
not split the match into low- and high-pressure situations, any data collection will be
confounded by this extraneous variable of tactics. One match between Norway and Brazil,
who were hypothesised as adopting a direct and an indirect attacking strategy, respectively,
was analysed to see whether notational analysis techniques would demonstrate this
difference. This would seem a reasonable match to analyse given that these two teams are
well known for their playing styles which are commonly thought to be somewhere near the
two extremes on the direct-indirect continuum of attacking strategy. However, there is
always an inherent danger in only analysing one match because incidents during the match
e.g., an injury to a key player, may cause unusual events to occur which, because of the
limited data set, may produce an unrepresentative picture of the way a team plays. Hence, a
true account of the match may have little value in predicting how a team typically plays.
Tenga and Larsen11 collected 23 attacking and 18 defending variables; and identified 8
variables where relatively large differences existed between the two teams and 8 where there
were relatively small differences. Of greatest surprise, was the almost identical incidence of
fast build-up attacks through midfield. This attack is where the ball is passed along the
ground as opposed to over the heads of opponents (which Norway used almost 3 times as
often as Brazil). However, was the incidence of fast-build up attacks seen in this match a
unique finding? It is possible that Norway and Brazil may never again exhibit the same
frequency of fast attacks through midfield. For example, Norway may have played many
more of these attacks than normal or Brazil played far fewer than normal because of some
one-off incident peculiar to this game.
Recently, researchers have investigated new methods for ascertaining how performances
compare with other performances over time (i.e., performance profiles12) that Jones and
colleagues13 have recently proposed a new methodology for. Performance indicators, defined by
Hughes and Bartlett14 as “a selection, or combination, of action variables that aim to define some
or all aspects of a performance” (p.739) are selected and a systematic record for a team, or teams
over a number of matches is collated. Data collected for each performance indicator during a
match is then standardised against previous matches and presented on one ‘form chart.’
Presenting a team’s match performance relative to previous performances allows the
coach to see clearly what aspects of the games were unusual. To illustrate this procedure, 10
matches for a British professional soccer team were analysed (post-event) using the Noldus
‘Observer Video Pro’ behavioural measurement software package (Noldus Information
Technology15). Thirteen performance indicators were arbitrarily selected from a study by
Taylor et al.16 and the data for the tenth game transformed relative to the previous 9 matches
(see Jones et al.13 for comprehensive details of this). The formula for this transformation is
shown in equation (1)13:

x – Mdn
(
Transformed score = 15* ————— + 50
IQR ) (1)

Where x = the performance indicator value for the 10th match, Mdn = the median and
IQR = the inter-quartile range for the previous 9 matches.
The calculations are relatively simple, but the theory behind the above transformation is a little
more complex (see Jones et al.13, and James et al.17 for full details). The resultant ‘form chart’ is
quite simple to interpret with a standardised performance indicator value of 50 indicating
performance at the same level as previous matches. Standardised values greater than 65 indicate
performance was above the 75th percentile and less than 35 is below the 25th percentile.
SS_12_Ch_09 6/6/06 12:27 pm Page 190

190 The Role of Notational Analysis in Soccer Coaching

100
90
80
Standardised Score

70
60
50
40
30
20
10
0

side

ds
rol
n

red

ded

s
)
)

)
s (%
(%

s (%

(%

(%

(%
ion
e

Car
Tak

ont
Sco

nce

Off
ept
ges

les

ses

les
nce

sse

fC

low
ots

Co
ibb

ack
erc

Pas
als
len

Cro
ara

so
Sh

Int
Dr

Go

ul T

Yel
als
hal

ul
Cle

sse
ul

Go

ssf
ul
al C

ssf
ssf

Lo
ssf
ul

cce

cce
cce
ssf
eri

cce

Su
ul A

Su
cce

Su

Su
Su
ssf

Performance Indicators
cce
Su

Actual Match Median

Performance Indicator
Value (9 Matches)
Successful Aerial Challenges (%) 50% 52%
Successful Clearances (%) 85% 93%
Successful Crosses (%) 25% 29%
Successful Dribbles (%) 76% 73%
Shots Taken 18 13
Goals Scored 4 2
Goals Conceded 3 2
Interceptions 23 15
Losses of Control 10 10
Offside 0 2
Successful Passes (%) 70% 75%
Successful Tackles (%) 58% 75%
Yellow Cards 3 1

Figure 2. Form Chart of the 10th Match Compared Against Performances

from the Previous Nine Matches for a Professional British Soccer Team

As can be seen in the form chart (Figure 2) the team analysed won the match 4-3.
Inspection of the performance indicators suggests a very low number of successful tackles
were made (58% success rate compared to an expected performance of around 75%) and may
have contributed to the high number of goals scored against them. This could thus be deemed
SS_12_Ch_09 6/6/06 12:27 pm Page 191

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 191

an ‘alerting variable,’ such that the coach would need to consider why this occurred and
perhaps focus on this area of the game in training. Other performance indicators that could
signal causes for concern were the high number of yellow cards and the fairly poor
clearances. Both suggest the opposition put the team under some pressure, probably as a
result of the opposition trying to get goals back when losing. On the positive side, the team
created more shots (n = 18) and made more interceptions (n = 23) than usual.
The principle of performance feedback can be at either team level (as above), positional
level (e.g., all midfield players) or at the individual level. The main issue however,
irrespective of who the feedback is aimed at, is recognising that performance naturally varies
between matches and so an unusual value obtained from one match does not necessarily
indicate a poor or exceptionally good performance. It is far better, and more convincing, if
trends are recorded over time. Figure 3 indicates that player D has not been very successful
in his aerial challenges compared to the other centre backs (data taken from Taylor et al.17).

70

60

50
Foul
Percentage

40 Fouled
30 Succsessful
Unsuccessful
20

10

0
Player A Player B Player C Player D
(12 matches) (13 matches) (14 matches) (6 matches)
Figure 3. A Comparison of Aerial Challenge Outcomes for Four Centre
Backs Playing for a Professional British Soccer Team

This performance is over 6 matches and is thought to be a reasonable number of matches

for performance to be considered representative of typical performance, although this
number is dependent upon the typical variability of the performance between matches12.
Indeed every match is a unique event and therefore any collection of matches is deemed to
be a random (stochastic) sample of a population of matches (all matches pertaining to the
team or teams being analysed). Thus any small sample has the potential of not accurately
reflecting the variability inherent in all of the performances of a team or teams.

PRACTICAL OUTCOMES FROM SCIENTIFIC RESEARCH

While scientists are judged by the rigour of their methods, the findings are not necessarily of
practical value to coaches. If the findings are difficult to interpret, then the practical relevance
may be minimal. One good example that crosses the divide between academic work and
SS_12_Ch_09 6/6/06 12:27 pm Page 192

192 The Role of Notational Analysis in Soccer Coaching

practical coaching advice was published in the Soccer Journal in 1989. Partridge and
Franks18, 19 undertook a detailed analysis of crossing opportunities during the 1986 World
Cup Finals. As an example of good practice, operational definitions were given for what was
considered a crossing opportunity and for other relevant features (such as the events that lead
to crosses, player positions, types of crosses and the results of the crosses18). The results
suggested that opportunities to cross were not always acted upon and these missed
opportunities resulted in a loss of possession on 42% of occasions. Further specific advice
was offered to coaches as a consequence of the results from the study. For example, it was
suggested that most effective were crosses played behind defenders that eliminated the
goalkeeper (i.e., too far away from the goalkeeper to make an interception a realistic
prospect). The authors suggested practices for coaches to use in order to improve players’
performances on this aspect of the game.

A PRACTICAL EXAMPLE OF NOTATIONAL ANALYSIS IN

PROFESSIONAL SOCCER
Observing the rule that practical relevance is paramount when presenting notational analysis
support within a coaching set-up, the first task was to identify the performance indicators
required by the coaches. In the following example, which is currently being used by this
paper’s author with a professional soccer team, the following four performance indicators
were chosen: shots, corners, ball entries into a critical scoring area and possession regains in
the opponents half. These performance indicators supplemented other analysis techniques
including further notational analysis. A sample of six consecutive matches was selected as
an exemplar, as this is the typical number of matches used by the team to assess their ‘current
form.’

OPERATIONAL DEFINITIONS
Operational definitions were formulated for each performance indicator. Shots were recorded
when an attempt at scoring a goal was made and classified as either on or off target, but was
not counted if the shot was blocked by a defender such that the ball did not reach the goal
line. Partial blocks (i.e., deflections) were counted. Corners and entries into a critical scoring
area were simply counted for incidence. The critical scoring area was defined as a semi-circle
in the penalty area encompassing the penalty spot (Fig. 4). Since in practice this is not
marked on the pitch, some error in coding was expected (but was assessed by way of inter-
operator reliability). Finally, possessions regained in the opponents’ half were defined as the
occasions when a team regained control of the ball from the opposition’s possession, when
the first touch was located in the opponents half of the pitch. The opposition had to have
control of the ball for a regain to take place. So, for example, if the goalkeeper kicked the
ball into the opponents half and one player from each team challenged for a header and the
defending player made contact this was not deemed to constitute possession. Consequently,
even if an attacker gained the resultant possession this was not coded as a regain. Also, if a
defender kicked the ball into touch the resultant throw was not deemed a possession regain.

VALIDITY AND RELIABILITY

Whenever notational analysis is undertaken there should be an initial query regarding the
extent to which the analyst is coding the events correctly (validity). In other words, when an
analyst codes an event as, say, a shot, was it in fact a shot? Any systematic mistake by the
analyst (e.g., coding a pass as unsuccessful when it actually was successful) reduces the
SS_12_Ch_09 6/6/06 12:27 pm Page 193

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 193

Figure 4. The Critical Scoring Area (Shaded) in Relation to the Penalty Area

validity of the analysis. This type of error occurs typically as a result of misinterpreting the
operational definitions, presuming that these were accurate in the first place. Thus, the
analyst may consistently (i.e., reliably) code events incorrectly. The only way of dealing with
this potential problem is to ensure that the operational definitions are well thought out and
understandable by all analysts and to check all analysts’ codes for a match against a ‘gold
standard’ coding for the match. A ‘gold standard’ coding is deemed to be the actual coding
and therefore as correct as possible. The only way to achieve this degree of accuracy is to
have all interested parties (i.e., coaches and analysts) view the match and agree the correct
coding for all events. This level of detail is not realistic on a day-to-day basis and is therefore
easy to overlook. In practical terms, if the coaches and analysts agree the operational
definitions at the outset then some of the analyses can be checked via recordings of these
matches in a slow and methodical manner. As long as the operational definitions were precise
and free from ambiguity, then this methodical approach should be accurate enough to
produce a ‘gold standard’ coding for the matches involved and hence the validity of the
analysts can be checked. The method for comparing analyst codes with a ‘gold standard’
code is beyond the scope of this paper but can be found elsewhere20–24.

METHOD
All of the coding required for the data presented in the results section took place live via hand
notation although each match was also recorded directly to computer hard disk for further
analysis using the Noldus ‘Observer Video Pro’ behavioural measurement software package
(Noldus Information Technology15) and MatchInsight (ProZone25).

RESULTS AND DISCUSSION

Performance indicators in this study are presented alongside each other as standardised
values relative to the opposition’s values (Fig. 5). The median values (for the six matches)
for each performance indicator are compared against the opposition’s values taken from the
SS_12_Ch_09 6/6/06 12:27 pm Page 194

194 The Role of Notational Analysis in Soccer Coaching

same six matches (see Jones et al.13 for detailed information regarding this procedure). The
overall picture suggests the analysed team has typically outperformed the opposition on these
performance indicators. Actual results for these matches were not particularly good, however
(1 win, 2 draws and 2 defeats), and so this particular chart was used as a motivational tool to
show the players that their performances remained good even though the match results had
not necessarily borne this out.

100
90
80
66.5
Standardised Score

70
60 56.25 57.5 57.5
50
50
40
30
20
10
0
Scoring Area
into Critical
Ball Entries
Comers

Possession
Shots off
Shots on
Target

Regains
Target

Figure 5. Median Scores on Selected Performance Indicators Standardised

Against the Opposition’s Values

Further detailed information regarding the performance indicators was also presented in
summary format using the actual data and on a match-by-match basis, with the data depicted
as shown in Fig. 6, for example. It was clear that match 3 had a very different profile to the
others and reflected the view that the opposition had performed better on this performance
indictor than the analysed team.
In order to track performances during the season, bar charts were used to compare
performances over the preceding two pairs of six matches (see Fig. 7 for an example)
accompanied by pertinent statistics. For the shot comparison in Fig. 7, it was noted that the
analysed team’s on-target percentage had gone up from 42.7% to 51.9% although the
frequency of shots had gone down.

THE CURRENT PREVALENCE OF NOTATIONAL ANALYSIS IN

SOCCER
If interpreting the results from notational analyses is difficult for coaches without relevant
statistical training, which probably accounts for most coaches, then it would not be surprising
if coaches either disregarded notational analysis altogether or otherwise employed specialist
help to undertake this role. In fact, published evidence suggests that notational analysis is
widely used in professional soccer. In 1997, Olsen and Larsen7 reported that nearly all premier-
league soccer teams in Norway used match analysis in one way or another. Given Olsen’s
position as National coach, this view obviously has some authority. More recently Blaze et al.26
SS_12_Ch_09 6/6/06 12:27 pm Page 195

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 195

25 Analysed team
Opposition

20
Number of Entries into Critical Scoring Area

15

10

0
Match 1 Match 2 Match 3 Match 4 Match 5 Match 6
Figure 6. A Comparison of Entries into a Critical Scoring Area on a Match-
by-Match Basis

90 On target
80 Off target
70
Number of Shots

60
50
40
30
20
10
0
Matches Matches Matches Matches
1-6 7-12 1-6 7-12

Analysed team Opposition

Figure 7. A Comparison of Shots by the Analysed Team Compared to the

Opponents for Blocks of 6 Matches
SS_12_Ch_09 6/6/06 12:27 pm Page 196

196 The Role of Notational Analysis in Soccer Coaching

reported that of the 10 English Premier League managers who responded to a questionnaire, 9
used hand or computerised notational analysis with further responses via semi-structured
interviews suggesting that notational analysis was widespread (see also Groom and Cushion27
in the FA publication, Insight, which is produced for professional coaches). Much of this type
of work remains unpublished, as the data and analyses remain secret to all but the authorised
personnel within the club. Thus, the majority of published work comes from academics with an
interest in the subject but who are not necessarily involved in the coaching process.

FUTURE DIRECTIONS OF NOTATIONAL ANALYSIS RESEARCH

One method for arriving at a definite conclusion from complex and ambiguous decision-
making is called fuzzy logic, which involves techniques for data processing and systems
control using a simple rule-based (IF X AND Y, THEN Z) approach. Wiemeyer28 used fuzzy
logic to evaluate soccer players to determine which position best suited their player profiles.
Weimeyer argued that a coach may apply some rules for assigning a player to a tactical
position (e.g., if a player has a good goal-scoring ability, good heading ability and is a good
risk taker, then striker may be the most appropriate position to play). This technique would
thus seem to have a lot to offer with respect to complex decision-making, but in practical
terms the complexity of applying a fuzzy logic computer program requires a great deal of
investment in terms of time and computer expertise. The formulation and management of
rules also requires input from sports experts and the extent to which they would agree on the
relative weightings is questionable. Future developments in this area are only likely from
academic institutions and at the time of writing fuzzy logic approaches have not developed
sufficiently to be directly useful for coaches.
A weakness of fuzzy logic systems is the need for the analyst, probably in conjunction
with coaches or other experts in the sport, to define the rules for group (set) membership.
Although the strength of the method is the ability to handle different rules for different
criteria, the rules still need to be formulated by the scientific team. A computational model
that takes this process one stage further is known as an artificial neural network (ANN).
Developed on the principles of nerve cells and their interactions, this type of model has the
ability to identify new relationships based on previously encountered ones (e.g., Perl29). Like
fuzzy logic this new, at least in the sporting context, technique has great potential but for the
near future is likely to remain an academic research tool rather than one used for coaching.
Indeed, Bartlett30 revisited his 1995 predictions regarding artificial intelligence (fuzzy logic,
ANN and most recently genetic programs) and concluded that the potential remains is high
but actual research output remains low. One of the reasons he gave was the relatively high
cost associated with developing such systems compared to the lower amount of research
money available to academics working in sports research.

CONCLUSION
Technology and research leads the way in terms of development of notation systems. One of
the most technologically advanced systems used in soccer is ProZone 330, which operates by
having specially developed sensors strategically placed around the stadium to enable data
capture which, when analysed using propriety software and unique image-processing
methodologies, produce a very accurate computer simulation of all of the players’
movements throughout the game. At present, the high cost of the most advanced systems is
likely to be prohibitive to all but the wealthiest teams and organizations in soccer. However,
as shown in this paper by the current author’s work with a professional soccer club, the low-
technology solution can be just as valid. The main development over the last 10 years has
SS_12_Ch_09 6/6/06 12:27 pm Page 197

International Journal of Sports Science & Coaching Volume 1 · Number 2 · 2006 197

been the lowering cost of computers and video cameras, leading to the formation of a number
of companies selling specialist software for different types of performance analysis. Over the
next 10 years, it is likely that the process of analysing performances will become significantly
easier. This is likely to be achieved in the software with the use of some form of interactive
help, thus removing the need to have expertise in computer programming and artificial
intelligence. The growing popularity of notational analysis in professional sport has led to the
need for sports scientists to be adequately educated in this area. The development of research
and teaching programmes in universities will strengthen notational analysis as a discipline,
leading to better research by way of methodological advances, more appropriate statistical
procedures and simpler output formats. These advances will in turn be fed back into sports
organisations and companies to the benefit of all prospective notational analysts.

REFERENCES
1. Hughes, M. and Franks, I.M., Literature Review, in: Hughes, M. and Franks, I.M., eds., Notational Analysis
of Sport: Systems for Better Coaching and Performance in Sport, 2nd edn, Routledge, London, 2004, 59–106.
2. Franks, I.M. and Miller, G., Eyewitness Testimony in Sport, Journal of Sport Behavior, 1986, 9, 39–45.
3. Reilly, T. and Thomas, V., A Motion Analysis of Work-Rate in Different Positional Roles in Professional
Football Match-Play, Journal of Human Movement Studies, 1976, 2, 87–97.
4. Pollard, R., Charles Reep (1904–2002): Pioneer of Notational and Performance Analysis in Football, Journal
of Sports Sciences, 2002, 20(10), 853–855.
5. Reep, C. and Benjamin, B., Skill and Chance in Association Football, Journal of the Royal Statistical Society.
Series A (General), 1968, 131(4), 581–585.
6. Bate, R., Football chance: Tactics and Strategy, in: Reilly, T., Lees, A., Davids K. and Murphy W., eds.,
Science and Football, E & FN Spon, London, 1988, 293–301.
7. Olsen, E. and Larsen, Ø., Use of Match Analysis by Coaches, in: Reilly, T., Bangsbo J. and Hughes, M., eds.,
Science and Football III, E & FN Spon, London, 1997, 209–220.
8. Hughes, M. and Franks, I.M., Analysis of Passing Sequences, Shots and Goals in Soccer, Journal of Sports
Sciences, 2005, 23(5), 509–514.
9. Lyons, K., The Long and Direct Road: Charles Reep’s Analysis of Association Football, in: Hughes, M.D.,
ed., Notational Analysis in Sport III, Centre for Performance Analysis, University of Wales Institute, Cardiff,
1997, 75–82.
10. Ensum, J., Pollard, R. and Taylor, S., Applications of Logistic Regression to Shots at Goal in Association
Football, in: Reilly, T., Cabri, J. and Arávjo, D., eds., Science and Football IV, Routledge, London, 2005,
211–218.
11. Tenga, A. and Larsen, Ø., Testing the Validity of Match Analysis to Describe Playing Styles in Football,
International Journal of Performance Analysis in Sport, 2003, 3(2), 90–102.
12. Hughes, M.D., Evans, S. and Wells, J., Establishing Normative Profiles in Performance Analysis,
International Journal of Performance Analysis in Sport, 2001, 1, 1–26.
13. Jones, N.M.P., James, N. and Mellalieu, S.D., The Development of Objective Methodologies for Scoring
Team Performance in Elite Rugby Union, Submitted to Journal of Sports Sciences.
14. Hughes, M.D. and Bartlett, R.M., The Use of Performance Indicators in Performance Analysis, Journal of
Sports Sciences, 2002, 20, 739–754.
15. Noldus Information Technology, The Observer Version 3.0, Base Package for Windows Reference Manual,
Noldus Information Technology, Wageningen, Netherlands, 1995.
16. Taylor, J.B., Mellalieu, S.D. and James, N., Behavioural Comparisons of Positional Demands in Professional
Soccer, International Journal of Performance Analysis in Sport, 2004, 4, 81–97.
17. James, N., Jones, N.M.P. and Mellalieu, S.D., The Development of Position-Specific Performance Indicators
in Professional Rugby Union, Journal of Sports Sciences, 2005, 23(1), 63–72.
SS_12_Ch_09 6/6/06 12:27 pm Page 198

198 The Role of Notational Analysis in Soccer Coaching

18. Partridge, D. and Franks, I.M., A Detailed Analysis of Crossing Opportunities from the 1986 World Cup
(Part 1), Soccer Journal, 1989, May–June, 47–50.
19. Partridge, D. and Franks, I.M., A Detailed Analysis of Crossing Opportunities from the 1986 World Cup
(Part 2), Soccer Journal, 1989, May–June, 45–48.
20. Atkinson, G. and Nevill, A.M., Statistical Methods for Assessing Measurement Error (Reliability) in
Variables Relevant to Sports Medicine, Sports Medicine, 1998, 26, 217–238.
21. Bland, J.M. and Altman, D.G., Statistical Methods for Assessing Agreement between Two Methods of
Clinical Measurement, The Lancet, 1986, 1, 307–310.
22. Hughes, M.D., Cooper, S. and Nevill, A., Analysis Procedures for Non-Parametric Data from Performance
Analysis, International Journal of Performance Analysis in Sport, 2002, 2, 6–20.
23. James, N., Jones, N.M.P. and Hollely, C., Reliability of Selected Performance Analysis Systems in Football
and Rugby, The Netherlands, Amsterdam, Proceedings of the 4th International Conference on Methods and
Techniques in Behavioural Research, 2002, 116–118.
24. Wilson, K. and Barnes, C.A., Reliability and Validity of a Computer Based Notational Analysis System for
Competitive Table Tennis, in: Lees, A., Maynard, I., Hughes, M. and Reilly, T., eds., Science and Racket
Sports II, E & FN Spon, London, 1998, 265–268.
25. http://www.pzfootball.co.uk. (Web page accessed on 27/09/2005).
26. Blaze, A., Atkinson, G., Harwood, C. and Cale, A., Prevalence and Perceptions of Performance Analysis in
the English Premier Association Football League, in: Hughes, M., ed., Performance Analysis of Sport V,
Centre for Performance Analysis, University of Wales Institute, Cardiff, 2004, 79–83.
27. Groom, R. and Cushion, C., Coaches Perceptions of the Use of Video Analysis: A Case Study, Insight, 2004,
7(3), 56–58.
28. Wiemeyer, J., Who Should Play in Which Position in Soccer? Empirical Evidence and Unconventional
Modelling, International Journal of Performance Analysis in Sport, 2003, 3(1), 1–18.
29. Perl, J., Game Analysis and Control by Means of Continuously Learning Networks, International Journal of
Performance Analysis of Sport, 2002, 2, 21–35.
30. Bartlett, R., Artificial Intelligence in Technique Analysis – Past, Present and Future, International Journal
of Performance Analysis in Sport, 2004, 4(2), 4–19.