Image by Vincent Teeuwen
So, Liverpool have appointed a throw-in coach! An out of the ordinary appointment and when first heard seems a tad humorous but is it a shrewd decision?
Marginal gains are proven to work and with 40-50 throw-ins taken per game there could be some big wins overall from throw-in based improvements. The much-maligned Keys and Gray don’t seem to think so but dinosaurs of the game like these are fast becoming extinct.
I agree that focussing more on throw-ins is a sensible thing and from matches I’ve watched and coaching I’ve received, it would appear that not attention has been given to this part of the game in the past.
Is it important however to have coaching from the world record holder in terms of distance thrown? I’m not convinced and most of the benefits to be gained from throw-ins taken would likely be from retaining possession not launching it forward Stoke-stylee.
I find it highly unlikely that Liverpool are going to score many (if any) goals by beaming it forward 30 yards for their speedy frontline of Salah, Mane and Firminho to run onto. If that was the plan and there was an edge to be had, the publicity created from the appointment will surely see the end to that happening anyway.
It is great that marginal gains and data in sports is being taken more seriously as illustrated by the acceptance of expected goals by football fans and pundits but I think we are also seeing a lot of smoke and mirrors where analytics are concerned. If you’re a regular user of Twitter it’s likely you’ll have come across a whole host of charlatan sports and betting accounts using the buzz words to make bogus claims or at least overstating their worth.
There’s no doubt that the quantity of data collected and how it used for comparisons between players has improved and I’m certain that helps with recruitment. There are so many graphs produced though which look pretty but with not enough of an explanation of why it’s significant or how it can be used to improve performance.
I am yet to read a piece that clearly shows how data analysis specifically has been used to improve team/player performance. I’m not saying that it’s not possible (or even has happened) because it certainly is but we need to see more evidence for it to be accepted more widely and used by more of those in the game.
The likes of Ted Knutson and his StatsBomb group show some decent visuals, created by their expertly recruited programmers but again with little information about how it’s used. Yes, I get that data analysis is only part of the answer and that there are client sensitivity concerns too but I think the main reason we don’t see much evidence of this is because we’re still uncertain as to how best use the data to make a difference.