(Mis)judgments in Football Analysis (1/3) – MH
Daniel Kahneman was one of the leading psychologists of our time. In his book Thinking, Fast and Slow, he presented extensive insights from his many years of research. This article explores how Kahneman’s findings, along with other insights from behavioral science, can be applied to (strategic and tactical) football analysis.
Kahneman became well known in the field of behavioral economics. Together with his colleague Amos Tversky, he succeeded in combining insights from psychology with those of economics. In 2002, he became the first psychologist to be awarded the Nobel Prize in Economic Sciences.
His research challenges the fundamental assumption that humans make rational decisions and that cognitive distortions are primarily driven by emotions. Instead, he argued that systematic thinking errors significantly influence our judgment and decision-making. These systematic errors, known as “biases,” occur in predictable ways under certain conditions.
Kahneman’s work has been applied in fields such as medical diagnosis, the legal system, intelligence analysis, philosophy, statistics, and military strategy. It is therefore a logical step to apply his research to sports analysis, where strategic decisions also play a crucial role.
“I suppose that every author has in mind a situation in which readers of their work might benefit from reading it. I imagine the coffee machine at the office, where colleagues exchange opinions and gossip. My hope is to enrich the vocabulary people use when they talk about the judgments and decisions of others (…).” – Daniel Kahneman; Thinking, Fast and Slow
A mere list of cognitive biases is not sufficient to recognize and avoid distortions in one’s own or others’ decisions. Only through a deeper understanding of their origins and mechanisms can the vocabulary around judgment and decision-making be enriched in Kahneman’s sense. Therefore, this article sheds light on the underlying processes and mechanisms of judgment and decision errors.
We begin by examining the scientifically established dual-process model. Following that, the mechanisms of WYSIATI, substitution, and priming will be explored in the context of football. In addition, the most relevant heuristics and cognitive biases will be applied to football analysis. Finally, we will address individual factors within analysis that contribute to the emergence of bias.
This article is only the first part of a series, which will span at least three essays. The series will examine flawed judgments in football analysis and their causes within a scientific framework informed by behavioral research.
One point should be made clear from the outset: bias (and noise) are present wherever decisions are made, and thus affect even expert judgments within their domain.
This article is based on a German translation of Daniel Kahneman’s work. Accordingly, the quotes cited here may not exactly match the original English wording.
System 1 und System 2
To explain human thinking, Kahneman introduces two hypothetical agents: System 1 and System 2. This model provides a framework for understanding how thought processes work. System 1 is responsible for fast thinking—it operates automatically, intuitively, and with little effort. System 2, on the other hand, governs slow thinking—it works analytically, consciously, and requires effort.
These two systems interact continuously whenever we are awake. System 1 runs constantly and automatically, while System 2 operates in an energy-saving mode, only stepping in occasionally. As a result, System 2 rarely takes control. Typically, System 2 adopts the impressions and intuitions generated by System 1 and accepts its intuitive responses.
The division of labor between the two systems is highly efficient. From an evolutionary perspective, this can be explained by the need to react quickly to threats while also engaging in deliberate, long-term planning. Thus, the automatic System 1 developed primarily for threat detection, while the energy-intensive System 2 evolved to support complex reasoning—though it is less frequently engaged in full.
However, it is particularly System 1 that is susceptible to cognitive biases. Due to its energy efficiency, it is limited in logic and statistical reasoning. It often tends to substitute complex questions with simpler ones. System 2, in turn, tends to adopt these flawed outputs, as it does not always monitor or correct the conclusions generated by System 1.
The following simple and intuitive thinking task illustrates the interplay between System 1 and System 2:
Team A has 62% possession and wins a match 1-0. Team B has 38% possession and loses. Which team was more dominant?
Of course, Team A was more dominant. At least, that’s the intuitive conclusion System 1 leads most people to. However, it neglects the fact that possession is not necessarily a reliable measure of dominance, and that match results are very imprecise indicators of performance. Despite lacking detailed data, System 1 assembles the question into a coherent picture. System 2 supports System 1’s incorrect solution, even though with minimal effort it could have reached the correct conclusion. However, System 2 is not consciously engaged.
From this, it follows that people rely on mental shortcuts—so-called heuristics—to quickly solve complex problems. Heuristics are usually very helpful due to their efficiency but can sometimes lead to errors in judgment—known as biases. This is where the relevance for football analysis arises, as will be shown in the following section.
“A heuristic is, technically defined, a simple procedure that helps us find adequate, though often imperfect, answers to difficult questions.” – Thinking, Fast and Slow
WYSIATI
First, I have linked a YouTube video that you should watch:
Even if you are already familiar with the video, its insights are remarkable. On the one hand, it can be concluded that focusing on one task can make us blind to other processes. To see and orient ourselves, we need a certain level of attention. However, the main insight of the video is that we cannot believe we have missed something so obvious. For those who did not see the gorilla, the player leaving the scene, or the curtain changing colors, it is unimaginable not to have noticed such obvious events.
“We can be blind to the obvious, and we are also blind to our blindness.” – Thinking, Fast and Slow
Applied to our football analysis, this once again demonstrates how prone we are to errors. How can we direct our attention to all levels of football tactics when, at the slightest distraction, we don’t even see a gorilla? We are only just beginning to discover many additional dimensions of the game as reference points, alongside spatial considerations in positional play. It is not only individual situations that are overlooked due to lack of attention, but possibly entire dimensions—and we may not even be aware of this.
We only realize our mistakes when we can no longer analyze a team with the tools we have. We follow our own subjective beliefs in analysis and overlook other logical conclusions that could have been drawn equally well from the existing data. This problem becomes particularly interesting when assessing tactical risk.
Recall the earlier question where you were asked to judge a team’s dominance based on possession percentages and the match result. System 1 constructed the best story it could from high possession and a win. What was not done, however, was likely questioning what is necessary to authentically assess a team’s dominance. We are blind to what we do not pay attention to.
Hasty conclusions based on limited data form the basis of many heuristics. Kahneman refers to these conclusions as “What You See Is All There Is” (WYSIATI).
“It is easier to fit everything you know into a coherent pattern when you know very little.” – Thinking, Fast and Slow
This central mechanism explains why consistency of information, rather than its completeness, is what primarily shapes our beliefs. Unfortunately, we also interpret information accordingly through our coherence-seeking System 1. Kahneman explains that System 1 is completely insensitive to the quality and quantity of information from which impressions and intuitions arise. System 1 looks for coherent information, while a lazy System 2 fails to check it.
“They did not want more information that might spoil their story. WYSIATI.” – Thinking, Fast and Slow
Attribute Substitution
To start, I have prepared a question that should first be answered:
How successful will Relationism be compared to Positionism in the next ten years in the top 5 leagues?
The answer will follow at the end of this section…
To interpret actions and processes, a complex, multidimensional understanding of the game is necessary. In analysis, we face difficult questions that can be answered on multiple levels. Often, there is not just one correct analysis, but several possibilities which together create an accurate picture.
“I know that I know nothing.” (cf. Plato, Apology 22d: “For I was conscious that I knew practically nothing…” )
Through reflection and the constant questioning of our observations and conclusions, we come closer to the goal of better understanding the game. The ability to accept that there are always new perspectives and interpretations fosters a deeper, more nuanced understanding. Unfortunately, when faced with complex matters, humans intuitively tend to fall into different patterns of thinking.
Kahneman explains that even with complex questions, we rarely seem overwhelmed. We are able to provide answers to questions we do not fully understand. But why is that?
Heuristics, as explained above, can be defined as mental shortcuts. Unfortunately, complex questions must also be answered in an energy-efficient way. Thus, we tend to substitute complex questions with simpler ones. This concept underlying many heuristics is called substitution. Often, we do not even notice that we have replaced a difficult question. However, we then use the answer to the simpler question to respond to the complex one. System 1 strives for the simplest possible answer to a question, while System 2, due to its inertia, does not verify the answer.
“You are not overwhelmed, you don’t have to exert much effort, and you may not even notice that you have not answered the question that was asked of you.” – Thinking, Fast and Slow
The concept of substitution is particularly pronounced when emotions are involved in the actual target question. Emotional attitudes amplify the effect. In the case of the so-called affect heuristic (judgments based on emotions), the interaction between System 1 and System 2 is especially prone to disruption. In the context of emotions and attitudes, System 2 does not act as a self-critic but rather as an advocate for the answer proposed by System 1. It looks for arguments that support its own beliefs.
Let us return to the question posed at the beginning of this section. You most likely found an answer to how successful Relationismus will be in the future. However, instead of systematically thinking through the target question, an answer intuitively came to you that corresponds to your beliefs. The heuristic question you unconsciously answered might have been: How much do I like Guardiola’s Positionismus?
Instead of searching for counterarguments against the intuitive answer, you subsequently searched for arguments that support your beliefs. To make a reliable future prediction, it would actually be necessary to free oneself from prejudices and break down the problem into its components, systematically questioning in which situations relational play is superior to positional play and in which situations the opposite is true. After all, the two philosophies can also be combined.
It can be concluded that substitution, together with WYSIATI, forms two of the most important mechanisms in the interplay of System 1 and System 2 underlying many systematic cognitive errors. The following section will introduce another mechanism: priming.
Priming
Priming is a fundamental mechanism that explains how we process information when thinking. Our memory plays a central role in this—more precisely, the way it connects information. This principle is called association.
Kahneman uses a metaphor to explain association: “Like ripples on the surface of a pond, activation spreads through a small part of the vast network of associated representations.” Even a single stimulus can be enough to unconsciously activate adjacent concepts or memories.
A priming effect occurs when a specific stimulus—a word, an image, a situation, etc.—unconsciously activates certain content in memory. This automatic chain of activation influences how we think, feel, or act without us noticing. The brain essentially lays down mental pathways in the associative network, thereby facilitating the unconscious retrieval of information.
To illustrate, here is a simple example: “If you have recently seen or heard the word eat, you will be temporarily more likely to complete the word fragment so_p as soup rather than soap. The opposite would happen if you had just seen the word wash.”– Thinking, Fast and Slow
Priming is already being used effectively in football contexts. For example, tactical training that includes video analysis and corresponding training formats represents a form of priming. By showing (tactical) videos during team meetings or recreating game-like patterns in training sessions, players can recognize familiar situations during matches, allowing them to react more quickly and effectively. Training decision-making under pressure and developing speed of action can therefore be linked to the priming effect. The specific form of priming in which an action is influenced by a mental representation is known as the ideomotor effect.
Another well-known example of this effect is the “Florida effect.” For those interested in reading more, I’ve linked the original 1996 study by John A. Bargh, Mark Chen, and Lara Burrows.
Priming also plays a role in many other areas of football. It can be deliberately used to create positive outcomes, but it can also have unintended negative effects. Depending on the context, different types of priming are distinguished—such as semantic priming, affective priming, goal priming, and many more. In football, priming is also used, for example, in injury prevention and rehabilitation.
Priming can also be used for the cognitive preparation of a match. In the so-called “Moments of Excellence,” players consciously recall their best past performances immediately before a game. Particularly interesting is the application of “success priming” by the coach, through the use of specific words or phrases in the pre-match talk. Attached is a 2006 study by Stajkovic, Locke, and Blair on this topic—specifically on the unconscious performance enhancement through goal activation.
Analysts, too, are subject to unconscious priming. A glance at the final score before conducting an analysis primes their expectations and subtly shifts the approach toward explaining the outcome, even though the final score, due to its inaccuracy as a performance indicator, says little about a team’s actual dominance.
Given the scope of the subject, an entire article could likely be written solely on the use of priming in football. In this article, however, the foundations of priming are used to understand it as one of the mechanisms underlying cognitive biases that affect football analysis.
“They were primed to find flaws, and that is exactly what they found.” – Thinking, Fast and Slow
Heuristics and Cognitive Biases
At the beginning of the introduction, I explained that distortions in our thinking and decision-making occur in predictable ways under certain conditions. By now, you should have developed a basic understanding of how our thinking is governed by the interaction between System 1 and System 2, how intuitive judgments arise in the process of analysis, and how mechanisms such as WYSIATI, substitution, and priming can specifically contribute to systematic biases in football analysis.
So far, around 200 cognitive biases have been identified. However, unlike many other articles that present a long list of biases with only brief descriptions, I deliberately refrain from doing so here. In my view, a simple enumeration does little to help recognize or avoid these distortions in practice. Instead, I focus on a small number of key biases that are particularly relevant to football analysis. Part 2 of this series will examine additional biases in the context of overconfidence.
Associative Coherence and Confirmation Bias
“Coherence means that you’re going to adopt one interpretation in general. Ambiguity tends to be suppressed. This is part of the mechanism that you have here that ideas activate other ideas and the more coherent they are, the more likely they are to activate to each other. Other things that don’t fit fall by the wayside. We’re enforcing coherent interpretation. We see the world as much more coherent than it is.” – Daniel Kahneman, Edge
What Kahneman describes here is the phenomenon called associative coherence. The processing of subjectively fitting information is facilitated by associative memory, while the processing of mismatching information is hindered. System 1 thus generates a coherent pattern of activated ideas in associative memory.
This phenomenon inevitably leads to errors in analysis. We should be aware that our analysis alone will never capture the entirety of the subject being analyzed. The analysis is always an expression of the analyst’s beliefs and the ideas activated in their associative memory. Even though System 2 can counteract when activated, it is still influenced by System 1. The concept of associative coherence is therefore closely related to WYSIATI and priming.
As part of the outcome of associative coherence, confirmation bias can be seen. It describes our tendency to want to confirm our existing beliefs.
“The coherence of associative activation induces a confirmatory bias when people examine a hypothesis by increasing the accessibility of hypothesis-consistent information. For example, the intention to test the proposition that ‘Sam is friendly‘ preferentially activates evidence of Sam’s friendliness, whereas testing the proposition that ‘Sam is not friendly‘ preferentially evokes instances of hostile behavior.” – Associative Processes in Intuitive Judgement
In analysis, we tend to seek out, select, and interpret information in a way that confirms our expectations. This, of course, applies to all areas of football analysis. Active countermeasures are necessary to avoid overlooking data that contradicts one’s own hypothesis.
Anchor
To simplify the illustration of anchoring, please answer the following two questions one after the other.
Does Cristiano Ronaldo have more or less than 1,200 international matches for the senior national team?
How many international matches for the senior national team has Cristiano Ronaldo played?
You have very likely come to the conclusion that Ronaldo has played fewer than 1,200 international matches. Nevertheless, this value probably influenced your estimate of the second question towards a number that is too high. Ronaldo has (as of March 2025) played 219 international matches.
Judgments are often based on the first piece of information given — the anchor. This biases the estimate in its direction. Daniel Kahneman distinguishes two psychological mechanisms that explain this effect.
- Anchoring as a deliberate adjustment process:
This is a conscious, systematic way of thinking (System 2). A study by Tversky and Kahneman from 1974 showed that people who had previously seen a randomly generated number (e.g., 65) subsequently gave higher estimates for the proportion of African countries in the UN than those who saw a lower number (e.g., 10).
“One starts from an anchor number, assesses whether it is too high or too low, and then gradually adjusts their estimate by mentally ‘moving away’ from the anchor. The adjustment usually ends prematurely because people stop when they are no longer sure whether they should continue moving.” – Kahneman - Anchoring as a priming effect:
Here, the anchor works unconsciously through associations in memory (System 1). In 2001, Mussweiler and Strack published a study in which they doubted that anchoring arises solely from insufficient adjustment effects. The anchor as a priming effect activates a specific mental schema in memory. As a result, thought contents that fit the anchor become more salient, and the selection of a subsequent answer is unconsciously shifted in the direction of that anchor. For example, when imagining a large sum of money and then being asked to name car brands, luxury brands are mentioned more frequently.
The anchoring effect is measured and demonstrated in most psychological studies using numerical estimates, for example by estimating prices or probabilities. However, such quantitative judgments are not directly present in video analysis in football. To nevertheless apply the anchoring effect to this context, a change of perspective is useful: Richards J. Heuer Jr., among others a former CIA analyst, investigated the influence of cognitive biases, including the anchoring effect, on qualitative judgment processes within the intelligence community. In his book Psychology of Intelligence Analysis, Heuer deliberately does not refer only to mere numerical estimates, but to all kinds of ongoing judgments or assessments that an analyst inherits from predecessors or gradually revises themselves. Empirical evidence for the effectiveness of anchoring effects in sports video analysis is still lacking, but an analogy via the intelligence community is plausible.
As already described above, the match result provides the analyst with the first available piece of information, acting as a kind of anchor. A 4–0 or 2–1 score leads us to attribute tactical superiority to the winning team and to base our tactical assessments too strongly on the result.
“We should, in any case, assume that every number presented to us has an anchoring effect on us (…).” — Thinking, Fast and Slow
Another example of an anchoring effect in video analysis can be found when analyzing multiple matches of a team. The first game analyzed sets an initial assessment of the team’s tactical approach. As the analysis progresses, the analyst may need to revise their judgments but often fails to adjust sufficiently and remains primarily anchored to the initial assessment. This results in a biased analysis skewed toward the anchor game.
„Even when analysts make their own initial judgment, and then attempt to revise this judgment on the basis of new information or further analysis, there is much evidence to suggest that they usually do not change the judgment enough.” – Psychology of Intelligence Analysis
It is not surprising that people look for reference points when faced with complex issues—the anchor provides exactly this orientation. Unfortunately, in certain cases, it distorts our decision-making. Mussweiler and Strack investigated in their study “Overcoming the Inevitable Anchoring Effect: Considering the Opposite Compensates for Selective Accessibility” how the anchoring effect can be mitigated. Simply being aware of the anchor and the effect is unfortunately not enough to avoid the bias. Instead, a strategy of deliberately thinking about the opposite proves effective in preventing it. Imagine, instead of a 4–0 victory, a 0–4 defeat, and consider how your analysis would look in that case.
Outcome Bias
Outcome Bias describes the tendency to judge performance based on the outcome. It is closely related to the principle of anchoring. The result serves as an anchor that shifts the judgment in its direction.
Unlike the example mentioned above regarding bias from the match result, Outcome Bias can be applied to all types of results. This means that individual actions and sequences by players are evaluated based on their success or failure — regardless of the actual quality of execution.
An example of this is a successful pressing sequence that leads to winning the ball. Because of the positive outcome, the pressing is overall rated as good. However, other relevant factors — such as intensity, timing, the opponent’s reaction, etc. — are either not considered at all or only insufficiently taken into account.
If the same pressing sequence is executed again in a different game situation but fails to win the ball, the pressing is suddenly rated as poor — even though the execution was identical. This bias demonstrates how strongly the evaluation is influenced by the outcome rather than by the quality of the action itself. All actions in the game can be subject to Outcome Bias in analysis.
To prevent Outcome Bias, the evaluation criteria should be adjusted. Instead of judging a performance by its outcome, the process quality should be assessed. This increases the likelihood of future successful actions.
Availability
The availability heuristic describes a cognitive shortcut in which people assess the frequency or likelihood of events based on how easily examples come to mind. In doing so, the original question—How often does an event occur?—is replaced by a simpler one: How easily can I recall relevant examples? This is a typical example of the substitution mechanism.
More important for the availability of an action than the number of examples one can recall is how easily they come to mind—what is known as ease of retrieval. A paradoxical example illustrating the effect of the availability heuristic was demonstrated by Schwarz et al. in 1991: people are less convinced by a decision when asked to list as many arguments in favor of it as possible. This approach plays the two factors—number of arguments and ease of retrieval—against each other: the more arguments that must be generated, the harder they are to retrieve.
In the context of football analysis, this heuristic can lead to an overestimation of certain in-game actions or patterns, particularly during live matches. Striking, prominent, or emotional moments are more easily remembered and therefore disproportionately influence our overall evaluation of the game—regardless of how frequent or relevant they actually were. The availability heuristic has a particularly strong impact on our perception of risk (explored in Part 2).
The game scenes most readily accessible to the analyst are also heavily influenced by their subjective experiences. A wide range of impressions—such as those from previous roles, past matches, or preferred playing philosophies—shape which factors and thus which moments are emphasized in the analysis. A strong bias toward particular strategies, such as overloads in possession, can distort the analysis by neglecting other decisive elements like timing. This simply stems from the complexity and diversity inherent to football. In this way, personal experience encourages subjective conclusions through the combination of the availability heuristic and WYSIATI.
Representativeness
The representativeness heuristic describes judgments about probability based on how representative something appears to be. Representativeness here refers to stereotypes and perceived similarity. When asked about probability, people often engage in substitution, evaluating plausibility instead of actual likelihood. Kahneman illustrates this with the following example:
“You see a person reading The New York Times on the New York subway. Which of the following is a better bet about the reading stranger?
She has a PhD.
She does not have a college degree.”
While the more plausible story might lead you to bet on the PhD, you should consider the base rates. Given that far more people without a college degree ride the New York subway than people with doctorates, the statistically sounder bet would be on the person not having a college degree. This example highlights the tension between base rate information and representativeness.
“The representativeness heuristic is misleading when it causes people to ignore base rate information that points in a different direction.” – Thinking, Fast and Slow
Kahneman accordingly advises questioning the informativeness of data and anchoring probability judgments to plausible base rates. To further illustrate the relevance of this concept in a football context, consider the following recent example:
At the beginning of the 2024/25 season, Bayern Munich employed an aggressive, man-marking high press. This tactical approach faced heavy criticism, primarily due to the potential for conceding counterattack goals. Despite the fact that several clubs—such as Atalanta Bergamo—had already been successfully using man-marking in high pressing for years on the international stage, the approach was viewed as atypical in Germany. Since the 1990s, man-marking had largely been replaced in the Bundesliga by zonal marking or hybrid forms. The evaluation of Bayern’s pressing was guided by historical prototypes of man-marking, particularly those involving a libero and, thus, a structural numerical inferiority. Although the man-marking was implemented in a completely different tactical fashion, the comparison seemed plausible due to the similarity and influenced judgments about its success.
In reality, the base rates painted a different picture: Bayern dominated their early Bundesliga matches, prevented opponents from gaining a foothold, and significantly limited their scoring opportunities. Statistically, the approach was highly successful: in their first 11 matches, 7 opponents failed to exceed an xG value of 0.5. The previous season’s champions, Leverkusen, only managed an xG of 0.17 and just 27% possession.
Nonetheless, the assessment of Bayern’s man-marking pressing was distorted by the representativeness heuristic: instead of evaluating the actual likelihood of success based on objective data, the plausibility of an outdated playing style was used as the basis of judgment. This illustrates how stereotypical mental models can lead to misjudgments and how crucial it is to critically examine information in football analysis.
For the sake of completeness, it should be mentioned that individual statistics (base rates in football) are not necessarily meaningful on their own. However, a combination of multiple statistics and data points can produce a more reliable picture to guide analysis. While xG values (as a combination of several data inputs) are still prone to error and need to be contextualized, they can nonetheless provide valuable insights into a team’s actual performance over time.
The Conjunction Fallacy and “Less is More”
The conjunction fallacy is one of the cognitive errors that arise from the representativeness heuristic. It describes the tendency of people to judge the combination of two events as more likely than one of the events alone—even when directly compared. The prerequisite is that the combination creates a more coherent and plausible story than the single event. One of the most controversial and well-known examples is the “Linda problem” by Kahneman and Tversky:
“Linda is 31 years old, single, outspoken and very bright.
She majored in philosophy. As a student, she was deeply
concerned with issues of discrimination and social justice,
and also participated in anti-nuclear demonstrations.
Which of the following is more probable?
-
-
Linda is a bank teller.
-
Linda is a bank teller and is active in the feminist movement.”
-
Despite the intuitive plausibility that Linda’s characteristics align more closely with the feminist movement, the first option is statistically more probable. After all, the conjunction of two events (“Linda is a bank teller and is active in the feminist movement”) is always less likely than one of the events alone (“Linda is a bank teller”).
„The most representative outcomes combine with the personality description to yield the most coherent stories. The most coherent stories are not necessarily the most probable, but they are plausible, and the concepts of coherence, plausibility, and probability are easily confused by the unwary.“ – Thinking, Fast and Slow
The “less-is-more pattern” by Christopher Hsee ties into this problem. For those who want to read more, I’ve linked the relevant study here. It states that the probability of the broader event (e.g., Linda is a bank teller) must be higher than that of the event that is included within it (e.g., Linda is a bank teller and active in the feminist movement). However, the more plausible event (less comprehensive) is judged to be more probable than the less plausible, more comprehensive one. This creates the “less-is-more” pattern. As a result, the probability estimate for the more detailed scenario is higher, even though this contradicts logical reasoning.
“This is the trap for forecasters and their clients: the more detailed the scenarios, the more convincing they are, but the less probable they become.” – Thinking, Fast and Slow
In the context of opponent team analysis for an upcoming match, the analyst must be aware that they are the forecaster, making probability estimates about the opponent’s future style of play. Thus, they should assess the likelihood of events—not merely their plausibility.
Applying Hsee’s “less-is-more principle” to this context means that detailed, plausible match scenarios may sound impressive, but are statistically less reliable for prediction than broader descriptions. Each additional assumption in a forecast multiplies individual probabilities and thereby reduces the overall likelihood of an accurate prediction.
However, this is not a call for a lack of detail in opponent analysis. On the contrary, the more detailed, precise, and comprehensive the analysis, the better the ability to derive general playing principles, patterns, and behaviors of the opponent. Therefore, the analyst should aim to analyze the opponent as holistically as possible before making generalized statements that increase the predictive probability.
Individual Conditions of Different Analyses
While in Part 2 I will address structural conditions in football that affect the intuition of analyst and coach experts, in this section I will focus on the individual conditions of different types of analysis. Even though I speak here about football analysis in general, it is important to clarify that there are various forms of analysis in football (see table). Different types of analysis also have different individual susceptibilities to cognitive biases.
| Category | Variations |
| Objective | Prediction (of opponent’s sequences), Evaluation (strengths/weaknesses analysis), Optimization (own processes) |
| Subject | Player, Team, Game situation, Matches |
| Condition | Live, Pre-match, Post-match, Long-term |
| Level of analysis | Micro/Atomic tactics, Individual tactics, Group tactics, Macro/Team tactics, Strategy/Game philosophy/Game principles |
| Data basis | Video, Tracking data, Statistics |
| Target audience | Coaches, Players, Media |
Certain biases are additionally triggered by specific types of analysis. Prediction, for example, is particularly affected by the representativeness heuristic, since it involves assessing probabilities. Team analysis over multiple games increases susceptibility to the anchoring heuristic, and post-match analysis is prone to hindsight bias—the tendency to overestimate the predictability of an event after it has occurred.
Similarly, focusing on a single level of analysis increases vulnerability to WYSIATI. You should now be familiar with the cognitive errors mentioned here, so they should not come as a surprise. One factor not yet addressed in this article, which demonstrably increases susceptibility to all biases, is time pressure. Time pressure plays a significant role especially during live analysis, though not exclusively.
Time pressure forces analysts to make quicker decisions. Research shows that under such conditions, people rely more heavily on heuristics, increasing the likelihood of errors. Among other studies, Finucane et al. (2000) examined the influence of the affect heuristic on the evaluation of benefits and risks, demonstrating that time pressure amplifies this effect.
Analysts therefore have to rely more on their intuition during live analysis. This brings me to Part 2, where I will identify conditions that increase experts’ susceptibility to intuitive errors.
Conclusion
By now, it should be clear why we delve so deeply into the mechanisms of heuristics. Without a fundamental understanding of how they work, heuristics and cognitive biases are difficult to recognize.
Many behavioral psychology insights can be profitably applied in sports analysis. This includes not only heuristics and cognitive biases but can also be extended to other areas of football analysis. However, to cover these application areas sufficiently, there will be a Part 2, Part 3, and, if necessary, a Part 4.
In Part 2, I will address the topic of overconfidence. Among other things, it will examine the susceptibility of expert intuition as well as risk strategies and risk assessments in football. Strategies for avoiding biases in football analysis will also be presented.
Part 3 (and possibly Part 4) will focus on “noise” in football analysis. Unlike bias, noise does not describe the distortion of judgments in a particular direction but the unconscious variability of multiple judgments under identical decision conditions.
To be continued…
Author: “MH” is a young analyst who regularly produces content for Spielverlagerung NextGen and TotalFootballAnalysis. On X, he can be found under Mh_sv5.
Keine Kommentare vorhanden Alle anzeigen