by Brain Traysers »
13 Sep 2018 22:58
John Smith Yeah yeah you can look at stats until the cows come home. There's so many variables in football that these ultimately mean nothing. We are only 6 games into a 46 game season. We will pick up points along the way where we "shouldn't". So what if we play those teams in the final run in too? We have to play them sometime - Stoke have looked poor so far and the pressure could well get to a Leeds or Boro should the be in with a chance of promotion
Certainly a common held view, but one I disagree with. As unique as football is, I don't think there are any mainstream sports where performance couldn't be improved by effectively utilising the right statistics. Most US sports are years ahead of us for this, with numerous success stories documented across sports. Closer to home, the Brentford xG approach has also been well documented, and allowed them to consistently punch above their weight (relative budget/wages). In addition their stats based player recruitment has given them a stellar record in the transfer markets, and this season they look a good bet to finally crack the play-offs. This has been led by their chairman, Matthew Benham, who made his fortune by using stats to identifying and exploit edges in betting markets. He also owns FC Midtjylland in Denmark and led them to their first ever Danish title in 2014-15, where they scored 25 goals from set pieces alone (2nd best was 11, league average was 9).
England also suddenly being good as set pieces wasn't a coincidence - that's a set piece coach who almost certainly has designed routines ultimately based on stats. Indeed, Liverpool have hired a coach who is the world expert in throw ins - that is another 'edge' gained through analytics. All these things add up over the course of a season.
More examples of forward thinking clubs include Man City who arrange an annual 'hackathon' (inviting analysts in to 'play' with their player data, presumably to hire the best), and PSG who have just been publicly announced they have partnered with a leading sports analytics company (which possibly explains the seemingly bizarre signing of Chupo-Moting). These are two huge clubs who have made their involvement public, I can almost guarantee others are quietly in on the act so as not to reveal anything and lose whatever edge they are targeting. (Check out Bournemouth joint topping the PL charts for goals from corners last season for a likely candidate - 0.24 per game)
Put another way - we demonstrated that we could't afford to buy a 15-20 goal striker, but if we hired a (data driven) set-piece expert to consult it wouldn't be unreasonable to get those 15 extra goals for a fraction of the cost (and if we hired an in house coach to learn from the expert it would cost even less going forward...) Those 15 goals would be worth a lot of very valuable points.
Unfortunately I think it's pretty likely our club is one of the many that still share your mindset - one of guys who worked for Benham at Brentford/FC Midtjylland started an analytics company who offer set piece services publicly said
zero clubs had been in touch with them after the World Cup, despite England clearly showcasing the benefits on the largest stage.