So it’s always fun to talk about things like Key Passes, and Chance Creation, right? Oh my gosh, look at how many chances Chelsea creates! Our problem is simply finishing! Those dang forwards! Do your jobs!
Yet the reality of the situation isn’t nearly so clear, or obvious. Sure, Chelsea create a lot of chances. In fact, Chelsea are tenth in Europe (Top 5) and second in the PL in chances created, with 12.8 per game. Yet somehow, we only have 38 goals (coincidentally, also tenth in Europe, but 6th in the PL). Should we blame the scorers, for simply not scoring? Probably, to a certain extent, but should we also blame the chance creators, for not creating better? How do we figure that out? You go to whoscored, and you see Key Passes, but how does one rate a specific Key Pass as being better than another? You can try to approximate by looking at the Assists, but that’s a crapshoot at best.
One way is to use xG and xA metrics. Now, there are plenty of websites out there explaining what these are in detail, for the inundated, but the short version is they are a rating based on historical shots with a certain number of key factors comparable to the individual shot in question. Everyone seems to have a different set of criteria, based on data available, math applied, etc. but things like location, type of shot, situation, are some of the common ones. And usually the historical data involves thousands upon thousands of shots, at least for those with access. A lot of people create their own table, but some of us, that don’t have access to a lot of data, use what’s available. In my case, I use the data at understat.com, because it’s about the only public source available that I’m aware of. Unfortunately, they’ve black boxed their data criteria, but being what one assumes must be a gambling related site, and a rather popular one, and touting the accuracy of their data compared to the real results over several years, I’m comfortable enough to use their data.
Now, technically we could go and hunt down each and every single key pass made, and rate them all individually, but I’m going to simply go for the average quality of the set as a whole. I’ve taken the Top 25 Key Passers in terms of KP volume (25+ this season), and put them in the table below, including their overall xA value, and thus showing the xA/KP, or what I’ll call Key Pass Quality:
As can be seen, with heat map shading, certain players really stand out. Fraser’s number shows quite possibly why Bournemouth might be getting 50 million pounds for Callum Wilson soon. Salah shows that he’s not just a scoring machine this season. The Manchester City players (Sterling, the Silvas, Sane) show why they score so many goals.
There’s an x-factor to this, too, though. When we talk about Key Passes, this actually includes Open Play, Set Piece, Direct Free Kick, and Corner sub-categories. What I normally care about, when I’m talking about chance creation, are the ones from Open Play. And I’ve got that table, same players, below:
Remove Holebas as an incredible outlier here (only 6 of his 28 KP are from Open Play) and the table really starts to show who creates well, and who simply creates a lot.
Now, let’s take this back full circle, and see how Chelsea specifically do with these more specific metrics, shall we?
Oh dear, oh dear. These are not good numbers. I don’t have them sorted any specific way, but it honestly wouldn’t matter anyway, it’s quite dreadful. Your best creator in terms of xA/KP in open play is one of your center backs. Your third best creator is a player everyone pretty much agrees has no football brain. Your fourth best creator is your backup center forward. And for those of you constantly parroting about how many KP Willian has… at this level of quality he’s better off passing to someone more capable.
Now, this isn’t to take away from our attacking players’ scoring woes, but it does paint a rather dreary picture on the reality of our chance creation quality. It’s simply not there. Let’s hope we bring in some talent soon to turn this around.
