Ca Co Cy


If a team loses it is only natural to look for the cause of the defeat. But cause and effect do not really exist. They are abstract concepts that need a justification to be used. In football no such justification can be found. Instead in football we have to work with the probability of a certain sequence repeating in the future. That might sound quite a lot like cause and effect, but it isn’t.

As the following tweet attest, more and more people are becoming aware that cause and effect might not be very useful in football. So let me explain what philosophical issue there are with cause and effect.

Criticism of cause and effect starts in the eighteen century with David Hume. He defines cause and effect as follows: if every time we experience an event B, there has been an event A just before it, the A is the cause of B and B is the effect of A. If Pepe tackles Messi and Messi is injured then Pepe is the cause of Messi’s injury.

At least that is what seems to follow from Hume’s definition. But Hume warns us against such a conclusion for he states that a cause if only a cause if the effect is always preceded by the cause. As not every time Pepe tackles Messi he is injured, we can doubt whether Pepe’s tackle is the real cause of Messi’s injury. Also, Pepe’s tackle is not the only thing that has preceded Messi’s injury. Maybe Messi has been tackled by other players earlier in the game. Or maybe he already started the match injured. The problem with causes is that we can always find more causes than we want and we can always go further back to look at the cause of the cause. Maybe Messi was injured by Pepe’s tackle because Pep Guardiola pushed him to take more risk attacking. But maybe Pep only did so because Barcelona was behind. And maybe Barcelona was behind because Messi missed a big opportunity. So maybe Messi is the cause of Messi’s injury.

In the nineteenth century cause and effect is criticized by Friedrich Nietzsche. Nietzsche shows that there is an inherent mistake in the concept of cause and effect. As the effect has to happen after the cause, cause and effect can’t be simultaneously. For if cause and effect would be simultaneously we would be unable to tell what the cause is and what the effect is. As Louis van Gaal liked to say: “Are we so good or are they so bad?”. Is Messi scoring so easy because Messi is very good or is Messi scoring so easy because the defense is so bad? But if cause and effect can’t be simultaneously, then there has to be time in between cause and effect. What happens in the time between cause and effect? If you don’t know how are you able to connect the cause to the effect. But if you do know what happens in between, then you are coming up with a new cause and effect pair. And then we ask the same question again: what happens in the time between the new pair of cause and effect. In football terms: Messi is able to score because he overtakes Pepe. But what happens in the time between overtaking Pepe and scoring? No matter the answer, we can zoom in to an even smaller time scale. Nowadays we register 2.500 actions during a match, but with more precise measurement we might be able to increase this to 25.000 actions. The smaller the time scale the more irrelevant the cause and effect pairs become. But at a larger time scale we run into Nietzsche’s argument about the time in between.

In the twentieth century Russell expands Nietzsche’s argument. Influenced by the newly discovered quantum mechanics and relativity theory, he states that even if take cause and effect to be connected in time, then both cause and effect as an event on it’s own take time. Messi’s overtaking Pepe take some time. Russell notices that the first half of the cause is ineffective in the same way as the second half of the effect. So we can subtract those ineffective halves of the cause and the effect. But we can keep doing this until there is nothing left of the cause and effect! 

Causation implies absolute truth. If we don’t know the cause absolutely, we cannot speak of cause and effect because the effect might not always be preceded by the cause if things are uncertain. But quantum mechanics show that everything has a measure of uncertainty. The same goes for football: there is so much uncertainty within football that it is overly simple to think in terms of cause and effect. Every time you indicate that you know the cause of something, you are actually proposing a theory. That theory is always uncertain due to the underdetermination of the theory by the data.

The probability of a sequence repeating in the future

It is very hard to live and work without cause and effect. Cause and effect talk is very much ingrained in our daily lives, inside and outside of football. In fact, our brain is hardwired to see cause and effect. Research has shown that whatever is the most salient detail at the same time as we experience the effect, is taken by the brain to be the cause of the effect. We see Pepe tackle Messi and we see Messi falling, so our brain concludes that Pepe’s tackle is the cause of Messi’s fall. But we have already seen that cause and effect can’t be simultaneously. So our brain assigns cause and effect incorrectly. Evolutionary this is a great strategy. To survive you don’t have to be correct, as long as you are safe when you are wrong. Our brain likes finding cause and effect so much that whenever you find a cause you are rewarded with a dopamine rush that is also used to reward you whenever you have achieved one of your goals. So cause and effect talk is ineffective because (a) the concept is incorrect and (b) our brain is biased to interpret the wrong things as if they were causes.

Nevertheless, almost everyone wants to talk in terms of cause and effect. Fortunately, that want can be satisfied by a healthy alternative for cause and effect talk. Instead of talking about how A causes B, one can talk about the probability of B following A. We experience sequences of events multiple times. The more we experience these sequences, and the more other people also talk about the same sequences happening even though you personally did not experience them, the more probable it becomes that these sequences repeat themselves in the future. This is actually how the brain learns game intelligence through associative learning.

So while people feel the need for cause and effect talk, it is much better to stop talking about cause and effect and replace all that talk with talking about the probability of a certain sequence repeating itself. In terms of football: stop talking about the causes of the loss if you lost a match, but instead look for patterns of events that preceded the loss and calculate how probable it is that those patterns repeat themselves in the future. 

Our brains are Bayesian brains. They are best described as biocomputers that use Bayesian statistics to calculate how probable patterns repeat themselves. The more we are able to train players to think in terms of the probability of a specific pattern repeating itself, the easier we make it for the player to increase his game intelligence. For example: the next time he looks at where the opposing players are currently in the match, the easier he can spot different scenarios of how the game will evolve and what probability each scenario has. That is much better than trying to find a single cause. First of all because causes don’t exist. Secondly because football is way too complex to be described in a simple cause and effect. And thirdly because our brain tends to pick up with wrong events as causes.