I think football has come a long way in a short time when it comes to looking at a player’s statistical influence on a game. Simply looking at goals and assists are a thing of the past thanks to the drive in more detailed statistical information available. Companies like OPTA & Statsbomb have opened many eyes to the very advanced but very applicable ways to measure a players’ impact. It’s now widely understood football is such a low scoring game that looking at goals and assists to evaluate players may not be the most efficient method!
What concerns me is the old version of me. Once upon a time i wasn’t the data nerd i am today, i was what i would call a “regular fan” a young lad who went to football with his pals, had a few pints, a burger and was only really concerned with the end result (p.s. i still do the beer, burger and mates thing). What i’m trying to come around to is that i think being in this analytical bubble, sometimes we tweet, write or talk about things that might be alien to many out there, or they have heard of it but have no real understanding of why it’s important. So to begin a short weekly series i thought i’d “go back to basics” and look at some very basic stats, why they matter and what can we use them for.
One of the easiest avenues to explore in this area is assists. Everyone knows the guys at the top of the assist chart are generally very creative players. I think also 90% of fans out there are now aware of the term “Key Pass”. To be honest it’s poorly named. A Key Pass to most of the data providers out there is in essence a shot assist, a pass that directly leads to a shot. So naturally for many years to look beyond assists for creative players, the first batch of data guys used Key passes. Let’s take a quick look at the EFL Championship for the top 5 players for “Key Passes”:
- John Swift – Reading – 54 Key passes
- Alex Mowatt – Barnsley – 44 Key passes
- Izzy Brown – Luton – 43 Key Passes
- Stewart Downing – Blackburn – 42 Key Passes
- Barry Bannan – Sheffield Wednesday – 41 Key Passes
Some of those names i’d argue make a lot of sense. Izzy Brown is a very talented no 10, Barry Bannan one of the better players in 2nd tier and John Swift is a huge talent known for his creative powers but unfortunately had too many injuries in the past. However, these are total key passes. If we are looking for a real consistent creative player, someone who could perhaps continue that skill set in other sides, we want to look at those who are creating chances from Open play only. Interestingly in the EFL Championship this season, the top 5 creators overall are also the top 5 from open play (not always the case!).
Now we’ve quickly ran through Key Passes i feel it’s important, as with any metric, to discuss the weaknesses of that metric. Key Passes are shot assists and that’s all they are, a key pass laying the ball across the 6 yard box for a striker to miss an open goal gets credit 1 key pass. That’s the exact same as a midfielder laying the ball off for his partner to hit a shot from 25 yards. There is nothing in the key pass metric that accounts for how “good” the key pass was, or better, how good the chance was that comes from the key pass. It’s why alone, Key passes have some value but need a deeper look to see how those key passes are in terms of quality.
This brings us to expected assists (xA). I wrote about how xA is calculated and why it is an important metric here:
What we can do with Expected Assists and Key passes is look at who is producing the most quality per key pass they make. This CAN give an indication as to who is the creating the chances with best chance of leading to a goal. To compare this fairly regardless of mins played i looked at my database of players, and the average value from the list for Key passes per 90 mins played. Anyone with a value below the average was left out of the next part of the analysis.
Here’s the top 10 players in the EFL Championship for xA per Key Pass
As you can see this is a much different list to that original “top 5 key passes” players. In my mind this list of players makes a lot of sense. Grady Diangana has been very good for West Brom, Kamil Grosicki is constantly laying on big chances for his team mates at Hull and Pablo Hernandez is one of the best creators in the league to the eye test. There’s also some quite well known but very interesting players from a recruitment perspective here such as Sergi Canos, Joe Williams and Fran Villalba.
By simply looking deeper on a single metric and providing it with some context we are able to show a much different view to what many people are using as analysis to look at a player’s creative output, we’ve not only gone beyond the assists, we’ve gone beyond Key Passes too!
Thanks for reading!