2016

Analytics in Football - Have the Geeks Inherited the Turf?

By admin | March 31, 2016
2016

Analytics in Football - Have the Geeks Inherited the Turf?

We don't often get a chance to run video on the site, but I was lucky enough to be invited on a panel this month at Charles Russell Speechlys on Analytics in Football. It was the first panel I have ever been on and - I rarely say this - but I am genuinely proud of what came out there. Ian's a great moderator, and Chris and Blake are both big experts in the field with plenty of practical experience. My comments are fairly wide ranging, and bring in a lot of diverse topics I have had a chance to learn about the last two years. This is nearly as good as any job interviews I will be giving in the coming months. 😀 I know it sounds heavy, but some of the material covered here will crop back up again and again over the next five years, both inside and outside of football. Though maybe not the part about the bear developing IP... Analytics Cliches That Are NOT True

  • Stats guys don't watch games. In fact, they generally watch a ton of matches to verify what the stats are suggesting is correct, both at the player and the team level.
  • Stats guys can only add value to the numbers side, like technical scouting. In fact, if your stats guys are at all like me, they are voracious readers and can have a major impact in fitness, medical, style of play, team and opposition analysis, cutting edge research like sleep/diet/fatigue AND in player recruitment.Essentially, if you are only getting basic stats stuff from your stats guys, you are either employing their skills badly or you need to find new stats guys.
  • Technical scouts/stats scouts hate traditional scouts.They only generally dislike BAD traditional scouts with terrible biases. You could create a whole book chapter of hilarious reasons why traditional scouts have explained why they don't like a player. Just when you think it can't get more ridiculous, it often does. HOWEVER, you need good traditional scouts - they are the lifeblood of recruitment. If you can educate the biases out of your scouts, everyone can probably get along swimmingly.
  • Every team uses stats and data now so first mover advantage is gone. Wrong. Wrong. Wrong. Few teams have built an organization that employs data-based decision making throughout the club. As I say in the video, it does not matter how amazing or incisive the information at your disposal is if decision makers are not using it, or continue to make poor decisions despite it.Having been inside football now and talked to tons of people there as well, I am more convinced than ever about the general level of backwardness in England. Clever teams can generate huge, sustainable edges because of it.

Analytics Cliches That ARE True

  • Some "stats guys" are definitely charlatans. They speak buzzspeak, but develop little or no IP of their own, but are good at repackaging ideas as if they did the work. They probably also happen to be good at marketing themselves. I say this again and again, but these same people are found in every discipline, inside and outside football. It's not a stats problem - it's a people problem.
  • Teams are terrified of making change. What if change doesn't work out? As Chris says in the video below, relegation from the Premier League can be existential for some clubs.
  • Plenty of traditional teams fail. Some stats teams will fail as well. What you don't know is WHY they failed. Without process information, it's damned hard to judge a team from the outside.

I'm sure there are more, but these were the ones that quickly came to mind. Anyway, enjoy the video. Hopefully in the next year, someone will do a few more of these. Note: for those who want us to do more podcasts, ignore the video and it sounds JUST LIKE A FOUR-MAN POD. You're welcome!