If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion. – David Hume

Jonathan Wilson wonders whether 4-2-3-1 has “lost its gloss” in his latest The Question article for the Guardian.

Tactical hermeneutics is a very interesting subject both to read and to discuss, but to my mind it is less reliable when applied to an entire formation irrespective of individual game, players or club.

Wilson makes a sensible argument for his view on the primary weakness of the 4-2-3-1: the space behind wingers, particularly when they are advanced on their full-back counterparts. He offers an example:

Robinho, whether by design or instinct, prospered there in the first half of the World Cup quarter-final between Brazil and Holland in 2010, never playing close enough to Gregory van der Wiel for the full-back to get tight to him but equally left largely untroubled by Arjen Robben. His goal stemmed from a run made from space into further space that opened in front of him, with Robben trailing hopeless in his wake.

This seems like a sound argument, but the 4-2-3-1 is used in such broad application by so many clubs in so many contexts, and with so many subtle (to the point of imperceptibility) mid-game changes that it becomes difficult to accept definitive statements of strength and weakness beyond single-game examples.

I wrote about this before with regard to Wilson’s similarly broad end of year review:

I think my problem…is one of form. Football tactics are so diffuse and context-specific that to point to any one overarching trend or movement is to risk gross oversimplification or omission. The best kind of tactical writing in my opinion is that which is anchored in specific games or moments; measuring their effect however on future trends presents a major challenge.

That’s why I think too that for too long, tactical analyses have given way to far too much subjectivity. If one amazing team like Barcelona favours a particular approach, it becomes very easy to read it into the approach of teams with ostensibly similar tactics. I also think that many tactical assertions are made without recourse to empirical evidence.

This is not to say that this kind of subjective interpretation of tactical trends, strengths and weaknesses is without value, but I do think it is subject to abuse. For example, it’s too often the case that some writers will imply a strong causal link between a certain, game-specific formation and a final outcome or set of outcomes.

In the case of Wilson’s argument for the weakness of 4-2-3-1 for example, there isn’t much in the way of non-anecdotal evidence that wingers are conceding goals by failing to track back in the empty space behind them. One could theoretically attempt some sort of analysis of the formation to see whether teams playing a 4-2-3-1 are more prone to attacks that begin from the exploitation of the space middle flanks, but this would be prone to all kinds of control problems. Formations shift in-game, wingers cut-inside, defensive midfielders move out wide, the game is in constant flux, and then managers make mid-game substitutions, change formations entirely.

One could argue that a certain tactical system would afford a team more chances than another, but again, this does not take into account the difference in opposition, or the ability of individual players.

Therefore, tactical analysis on the macro level resembles a kind of football metaphysics, a set of assertions that can neither be confirmed nor denied. You might think it merely dickish to point out, but I think the more we are aware of the shortcomings of this kind of tactical writing, the more honest we can be about what exactly we know about the relationship between one formation and a set of tendencies and outcome, and whether they in fact exist.

That said, I think it is also extremely wrong-headed to attempt to make the leap from broad statistical evidence like TSR or PDO to prescriptive tactical formations (this is essentially what Charles Reep attempted with his recommendation that teams try to rack up the shots from as few passes as possible).

But analyzing football tactical trends is still far too vague and subjective to be regarded as definitive. Even single game analysis is prone to different causal interpretations (the movement of a central defender allowed a forward to exploit space, etc.). This does not make it bad (or, as Hume said, worthy of the flames) necessarily, but it also doesn’t necessarily true in the empirical sense.

Comments (6)

  1. A great read, and a very interesting and objective look at football tactics and their (mis)interpretations.

  2. First off let me say I really like Jonathan Wilson and tactical writing in general, but I think you raise an interesting point.

    Drawing assumptions like the decline of the 4-2-3-1 or any such idea is the classic case of statistical “over-fitting”. This is a problem we will never really be able to remedy in football, because the sample will always be by definition skewed. For example to really see how a 4-2-3-1 matches up against say a 4-4-2 we’d need a much more equal sample. In the United-Liverpool match over the weekend United’s 4-2-3-1 had difficulty in the second half against Liverpool’s 4-4-2 for many of the same reasons that Wilson mentions, although the result didn’t reflect it. If United were to stick to the 4-2-3-1 for the rest of the season and Liverpool to the 4-4-2 we could draw different conclusions about the merits of the two formations, however we would be over-fitting our model to a sample bias sample. The sample is bias by the simple fact United and Liverpool are very different teams with different abilities (and it would be hard to make any argument in which Liverpool is a better team than United). So to effectively test our model about the relative merits of the 4-4-2 vs. the 4-2-3-1 we’d need to see the opposite and have United replay the entire season playing a 4-4-2 and Liverpool replay the entire season with a 4-2-3-1. This of course is the timeless problem that social sciences face in non-controlled experiments.

    Soccer is in a sense a social science, and because it is very rarely viewed as a science of any sort I think these classic problems we see in statistical analysis of a social science are exacerbated with soccer. Over-fitting happens all of the time when we create a model that describes what we are seeing instead of the state of the world, and macro-level tactical analysis is a perfect example of this.

    • Very good summary. Kind of a reverse engineering…happens a lot in evolutionary biology too.

  3. A nice read, as always. However, I would be careful not to conflate “objective” with “empirical” (or rather, juxtaposing “subjective” and “empirical”). The interpretation and reporting of results (never mind the actual measuring of phenomena) seems to be subject to the particularities of the researcher/author. As I think you suggest in this piece, even the use of empirical results requires argumentation as to the soundness of a particular interpretation or position, one way or another (and really, that’s one of the big reasons it makes arguing about sports so much fun).

Leave a Reply

Your email address will not be published. Required fields are marked *