We’ve gained a lot of readers just in the first week, which is pretty awesome. In a way, I guess there were a lot of people who were interested in analytics just waiting for a site like this to crop up. However, because so many of you are new, you might have missed out on some things we wrote back before this site existed, when we only published in the dark corners of the internet.

I’ve migrated some of my writing on football and stats to StatsBomb, and today I want to recap what I’ve been working on this summer.

I spent most of May and June working on a framework to statistically evaluate players. Most of the writing has been focused on attackers, partly because they are far more interesting to write (and read) about, and partly because midfielders, defenders, and especially goalkeepers are tougher to evaluate just using public stats. The thought behind all of it was that most football teams seem objectively bad at finding young, undervalued players. Is that happening because those players don’t exist? Because building reliable scouting networks is hard?  Or maybe because they just don’t understand how to use stats to do some of their scouting for them? There’s now enough public data that I figured I could help out with at least the first and the last one, so I began donking around with the data.

max_kruseIt really kicked off with two different pieces. The first one was called Walking It Into the Net. In it, I postulate that “assists are the game of football’s walks,” meaning assists are quietly a very useful little stat in finding potentially undervalued footballers. There’s also a bit in there about key passes and Schrodinger’s Goals, which has guided my general thinking on this subject since. This search for value also lead to the next piece, which was about one particular undervalued player. That piece was called The Max Kruse Problem.

From there I started writing transfer dossiers. I didn’t want to discuss the big names everyone knew about. I wanted to find guys most people didn’t know about, and maybe draw some attention to them before their names appeared everywhere in the transfer rumors.

Following the Kruse piece, I stayed in Germany, discussing Farfan, Kiyotake, Gonzalo Castro, and Aaron Hunt. Kiyotake has briefly been linked to Aston Villa, but I haven’t heard a whisper about Castro or Hunt.  I’m linking that one first, because it has a methodology outline there, which at this point is pretty obviously a work in progress.

Anyway, I don’t want to go through these step by step. I just wanted to introduce new readers to them – they mostly highlight what the stats say about all of these players, and provide a number of Youtube videos for your viewing pleasure. (Yay, now we can all shop like Sven Goran Erikson!) For the rest, I’ll just link to them, and if you want to check out what I said about some of your (potentially new) favourite players, you can.

Think you might be interested? Here’s a sample of what I wrote on Manchester City newboy Stevan Jovetic.

Name: Stevan Jovetic
Team: Fiorentina
Age: 23
Position: Wide Forward (Left)
Potential Buyers: Juventus, Arsenal, sundry other interested clubs

Looking for a young player who played a lot smarter at age 23 than he did at age 22? Look no further

[table id=11 /]

Passing percentage up. Turnover numbers (Disp+Trn) waaaaay down. Often you’d think that would come at the cost of dribbling stats and key passes, but dribbling only dropped a touch, while key pass numbers went up. This is exactly the type of improvement you want to see from precocious young talents. Basically, Jovetic got quite a bit better at doing the important good things, and better and not doing the important bad things (like giving the ball away).

The only worrying thing about Jovetic for potential buyers is that he only gets 38% of his shots on target. This is somewhat understandable for players who play on teams with lesser talent, as they are often not in great positions for shooting, but Fiorentina were actually quite good. I find myself wondering what his shot position chart looks like.  That said, this seasons’s non-penalty goal rate is at the edge of elite at age 23, and Jovetic has a ton of stuff going for him. Even clubs at the top levels of the Champions’ League would benefit from having Jovetic around, especially if he keeps getting better.

Update: Thanks to the miracle of the internet, Colin Trainor immediately provided me with Jovetic’s shot chart. BAM!

jovetic_shot_chart

What you see there are waaaaay too many shots outside the good shooting areas. Speculative shots from range have terrible expected value, and Jovetic takes a ton of them. Get him with a good manager who can correct his shooting areas, and teams will have a real gem on their hands.

Likely cost: £25M


(Click the player names to go to their transfer dossier.)

Serie A: Icardi, Erik Lamela, Miralem Pjanic, Lorenzo Isigne, josep Illicic, El Shaarawy, Marek Hamsik

Serie A: Stevan Jovetic, Marco Sau, Adem Ljajic, Gomez, Luca Cigarini

La Liga: Christian Tello, Antoine Griezmann, Diego Costa, Thiago

La Liga: Ivan Rakitic, Pizzi, Koke, Iago Aspas, Patrick Ebert

Ligue 1: Ben Yedder, Alessandrini, Ryad Boudebouz, Saber Khelifa

Ligue 1: Dimitri Payet, Yohan Mollo, Remy Cabella, Julien Feret

When the transfer window closes, I’ll do a review of the players that moved, where, and at what price to see how the projections fared. Assuming time allows, I’m also going to look back at the data and see which players would have been flagged last year, and check how they did/whether they moved anywhere as well.

Welcome aboard. I hope you enjoyed the first week around here, and thanks again for all the support.

Cheers,

–TK

  • Toshack

    Thanks Ted,
    Read most of it earlier, but the “follow up promise” will be a real intersting read!

    • Ryan

      Agreed. The idea of a sports analyst being accountable for their work is still a mindblowing concept, but hopefully one that gains steam!

Improve Performance and Productivity in Your Club:
State-of-the-art Football Analytics