A treasure trove of data has been released at FB Ref via StatsBomb. It’s hard to understate just how big a leap this is for public analysis and the understanding of the game. It’s by far the biggest moment there has been since I’ve been involved in the rough and tumble world for soccer analytics in 2015. I encourage everyone to go, play around, click a ton, reference it, spread it on twitter, reddit, wherever soccer nerds are found. It’s amazing.
The stat the editorial team here at Saturdays on Couch has been clamoring for open acceptance of for years in Yards Gained is now widely available. I do think it will work better as a proportion of field gained but that still needs a catchy name, Progression Rating? Eh…I digress.
Today I want to look at a fascinating stat just now accessible to regular plebs, which is targets. How often your team tries to get you the ball, including incomplete passes ist very interesting to me because we can see where teams are trying to get the ball and who is getting open.
So the metrics I used in these charts
Target Share: simply how often a player is targeted per minute relative to his teams passes per minute
Actual/Expected Yards Gained Per Ball Loss: Total progressive yardage divided by ball losses per minute is how we get yards gained than a rough
npXG + XA: Hopefully you know this, basically expected goal production with pens stripped
Field Tilt: Where on the pitch a players offensive touches generally come
We want to see which players are receiving the ball relatively often and then get a glimpse at what they are doing with it.
So for Arsenal a few things jump out:
- Dani Ceballos. He gets targeted 40% more than the normal player despite playing in the center of the pitch. 94% of passes intended for him are complete, better than Torreira by a few clicks and similar to Guendouzi playing further back the pitch.
- Guendouzi vs Torreira
|Player||Yards Gained Per Minute||Yards Gained per ball loss|
- Aubameyang and Lacazette are playing relatively similar distances up the field in this rough metric and receiving a similar share of passes along with contributing similar goal production but Auba is contributing 17 yards gained per ball loss vs Laca’s 10. Auba passes a bit more but really the carrying is where you see the difference, over a typical 90 Aubameyang will carry the ball 50 yards more towards goal than Lacazette.
- This Grealish fella seems to be important? He is targeted at nearly 150% the rate of a normal Villa player despite playing very far forward. I had to bump up the y axis to see him. On a team where basically everyone is below average at ball progression, Grealish is 63% above what you’d expect at yards gained per ball loss for someone who plays on his part of the field. He’s one of the most versatile players in the Premier League and one of the best.
- Harry Wilson simply doesn’t get the ball passed to him that often. It’s odd in a way because 81% of passes toward him are complete, compared with 59% to Solanke, 43% to Wilson, 62% to King and a similar 79% to Fraser.
- Ake progresses the ball twice as far per ball loss (66 yards) compared to Francis (33).
- The color scale gets a bit wonky on the very ends for pure penalty box strikers without comparison on their team so I wouldn’t put too much stock into the Callum Wilson coloring.
- You knew who was going to be the player getting the ball a lot before you scrolled didn’t you. If not, you haven’t been reading me since my Big Bundesliga Preview days at StatsBomb when I used to write articles titled “Ingolstadt Vol 2: Großer than before” and posted this image showing Ingolstadt’s touches their first Hinrunde in the Bundesliga
- Dunk progresses the ball in sheer volume terms almost twice as far as Duffy, and does it much more efficiently as well.
- Trossard is at 18 yards gained per ball loss, Murray at 5…Trossard seems a player who should start every match to me, just does so much.
- We have to change the scale here. The strikers are getting targeted at 160% the average Burnley player. Dyche-Ball certainly is an ethos.
- The color scale is destroyed because they don’t progress the ball at all, but McNeil’s light blue should stand out more than it does.
- I’m guessing Christensen plays in in a back 3 maybe or some sort of different formation where he simply isn’t a passing outlet.
- I honestly thought Reece James was going to be much more of a standout here, his numbers are indistinguishable from Palmeiri and Alonso.
- Kovacic and Ceballos occupy similar spots on these charts as ball receiving hubs who progress excellently, Jorginho is playing so deep that his raw volume numbers are misleading. They look good at first glance but are below par for that spot on the field.
- Passes intended for Kante are only completed 85% of the time, compared to 95% for Kovacic, 94% for James, 90% for Barkley and even 80% for Hudson-Odoi. Kante’s numbers are a bit concerning I’d say. Mount is down at 73%…
- Hudson-Odoi and Pulisic’s are certainly not. Hudson-Odoi has seemed to have his reputation suffer a bit on twitter (or at least in the 4 tweets I’ve read in replies to a Martial tweet from a stan account) but I don’t see many reasons to doubt him yet.
- I don’t think it’s mentioned enough just how amazing a club name Crystal Palace is.
- Two defenders in PvA and Sakho gain more total yards per minute than Zaha, but just barely. It’s kind of Grealish-esque.
- Mamadou Sakho has always been a progression king, as in literally one of the best of the world at it.
- I used to think Max Meyer was going to be a star, I remember favorably comparing him to Julian Brandt at one point several years ago. That aged the worst.
- Again the very edge penalty box striker scale is a bit skewed so don’t overreact to the Kean number
- If I started again I would probably make the max circle size bigger and at a bigger cutoff, so DCL’s .62 number would pop a bit more. He’s targeted so much to get goals. It’s interesting to see he gets targeted 30% more than average Everton player while say Abraham was 20% less. DCL rec% of 52.4% is higher than Abraham’s as well.
- As expected, Sigurdsson and Schneiderlin show up as kind of nothings here. I remember Schneiderlin being hyped as a legit PL superstar by some data guys back when he left Southampton…that did not age well. I remember selecting Gylfi and Ryan Fraser high in my fantasy drafts….that did not age well.
- I think Harvey Barnes is a megastar.
- I don’t know if James Maddison is. Barnes and Maddison gain similar yards per ball loss (~22) but Barnes is getting forward way more often
- I’ve noticed these types of charts aren’t good at identifying fullbacks really.
- Here we finally see some fullbacks. TAA and Robertson are targeted nearly 20% as often as the average Liverpool player, helped by midfielders being below average like Fabinho, Ox, and especially Wijnaldum
- It’s the old efficiency paradox. I originally was going to just look at yards gained per minute as the “color” of the dot, in which of course Trent is third in the league.
|Player||Yards/Min||Yards/Ball Loss||Live Shots Created/90|
- He and Robertson are just so different from everyone on this list (except Kovacic) that it’s hard to get across. I mean losing the ball a lot is generally bad. I think TAA is in a system that maximizes many counting stats, he could certainly be more efficient if he needed to be or was asked to be. If a Real Madrid or Barca buy him, they will probably expect him to be, but they don’t know how far he could swing. He’s a fascinating proposition as far as how valuable is he? If Reece James was given the same role how much would Liverpool suffer?
- The difference in how Liverpool and City target their strikers is obvious here.
- Sometimes I think Mahrez is the best player in the Premier League. More than sometimes actually.
- United basically didn’t have a ball-dominant player at all, I’m not sure Fred is naturally that type, until Fernandes came in.
- His efficiency is terrible as we all know, Rashford and James gain similar amounts of yards per ball loss further up the pitch, but there was a gaping hole in this attack for someone to center on. 30% more touches than average.
- I don’t think I realized how rarely Mctominay touched the ball in advanced positions.
- Axis change!
- Poor Jetro. Newcastle basically ignore all players listed as defenders except for Matt Ritchie.
- This is just a brutal looking chart.
- Why is Moritz Leitner getting the ball so much?
- Ben Godfrey looks to be a real-deal center back on the ball, Ake-esque.
- Lundstram and Norwood are who they want to have the ball from deep.
- Most of the forwards are at about 50% receiving, so half the passes played to them are complete
- Such an odd team really, there are just no pieces “between” forwards and midfield.
- A clear plan of our passes should go forward
- Is Sofiane Boufal having one of the best per minute seasons in the entire league?
- Maybe I should have had rec% as the color….78% of passes targeting Lamela are complete while just 70% are to Alli and 65% to Moura, despite them getting a similar target share at a similar part of the field. Lamela is clearly the best progressor of the three and has similar goal production.
- Eriksen has a higher goal production circle but actually has fewer live shots created than Lo Celso, while Lo Celso comes clearly ahead in raw progression and progression efficiency.
- They do not want Cathcart to have the ball.
- Sarr and Deulofeu are both similar but very good players.
- 86% of passes attempted toward Felipe Anderson are complete, just 72% are of those toward Fornals further back the pitch.
- Anderson and Lanzini still stand out as great players with the ball, receiving and progressing. Something isn’t working though.
- Traore and Jiminez have a lot of the hype but Jota is my favorite Wolves forward and I think clearly the best.
|Player||npXG + XA||Live Shot Creations||Total Yards Gained Per Min||Yards Gained Per Ball Loss||Reception%|
- 23 years old.
Maybe I jammed too many various stats in here and strayed to mentioning different metrics, I’m rusty ok? But it’s fun to see the target profile of teams and how it differs: Liverpool’s FB’s so high and forwards above average with nonexistent midfield, Burnley’s bomb toward forwards, Sheffield’s lack of “attacking midfield” targets, Norwich’s only attacking midfield targets, United’s reliance on a very few ball dominant players, and on and on. It’s unbelievable data and free and easy to access for everyone at football-reference. It’s the biggest leap forward for public analysts by far. Please support the site and give them clicks, references, links, etc. I don’t think most realize just how big a deal this is.