Do statistics drive modern football or are modern fans overly obsessed with them?

[sullydnl]

I noticed someone suggest "statistics" as a daft obsession of modern football fans in the so-named thread.

Yet as we've just watched the moneyball-driven Brentford beat Arsenal having made it to the PL via their analytics team, it seems to me that it's less that statistics is a daft obsession for modern fans and more that football is an increasingly stats-based game, with the discussion around it reflecting that. After all, Europe's top clubs seem to be spending fortunes on analytics teams that drive recruitment, management and tactical decisions and we've seen top-level managers like Tuchel openly speak about games in terms of the stats like xG (a type of stats that can draw particular ire from more old-school fans).

So are statistics in fact inessential to understanding how football works, just an annoying affectation of modern fans and a crutch used by people with a lack of real football knowledge? Or is modern football itself now so stats-driven that people who dismiss statistics simply aren't going to have a full understanding of what is happening in the game?

 

[abundance]

Managers talking xGees... well managers rely a lot on "lucky / unlucky" and "we created a lot of chances / they didn't create much chances" to positively spin the performance of their team in post match talks, so xG numbers fit perfectly to reinforce that narrative with a sciencey, objective aura.

(and it is mostly just an aura, because the methods to weight shots for xG are so crude still, those numbers aren't that much better than eyeballing "lucky/unlucky" from memory and take a look at total shots/on target in match reports)


Anyway, in general, statistics and data analysis are super useful for football professionals, who collect troves of data this big, and continously develop new models to answer questions, quantify impressions, isolate features, identify trends, measure progress, assist scouting...

Fans on the other hand shouldn't read too much into the few data points and sintetic scores they can find online, those can start a conversation but are too limited and crude to prove, reveal or settle anything.

 

[Iker Quesadillas]

Yet as we've just watched the moneyball-driven Brentford beat Arsenal having made it to the PL via their analytics team

It's probably best to have a bit of healthy skepticism about these types of claims.

 

[sullydnl]

It's probably best to have a bit of healthy skepticism about these types of claims.

How so? Their approach is pretty well documented.

 

[Iker Quesadillas]

When you say "well documented", do you mean "their models are publicly available and/or they're published data and methodology on economics and business journals" or do you mean "news articles have been published in which they're the only source"?

 

[LoneStar]

I don't doubt that data would help teams in a sport at the very top level. Where every bit matters. Of course the whole thing depends on how the data is used, but its value is undeniable.

As a data analyst for a media company, it's unbelievable how many times data has suggested something which seems very counter intuitive.