This is a post I should have written a long time ago – the purpose is to provide a post (which I’ll link to most of my posts from here on out) which explains the metrics that I have used, and why they’re ‘relevant’
I guess a good place to start would be interesting links/pieces of work which would supplement the stuff I write – My ‘analysis’ is by no means the best on the web. The likes of Dan Altman, the team at StatsBomb (Constantinos Chappas, Colin Trainor, Ben Pugsley, Bobby Gardiner, James Yorke, Ben Torvaney, Paul Riley etc) and other notable ‘fanalysts’ like Tom Worville and Owain Thomas have some really good material.
There are also a myriad of good books out there – my personal favourite is the Numbers Game, but Soccernomics is along the same vein; Paul Tomkins has an excellent list on TTT (another great source of very good articles)
Anyway – The main point of this post is to provide a few key definitions; An easy job for me would be to link to Statsbomb’s glossary page, which is very good and much better at explaining things than I can. And has a few more which I don’t really use as much.
So, I’ll just go through the ones I used here.
My ‘dataset’ has basically been compiled and cleaned using WhoScored.com – It took me a few days but it was worth it to look at a complete form of stats for Championship players. WhoScored is an excellent source for a range of data, as I believe the underlying stats are from Opta. I’m not a huge fan of Squawka, mainly for the lack of actual informative articles (especially given the nature of their website) and it’s similar to accounts like Gerrard8FanPage or whatnot with their click/interaction bait based posts. It’s almost on the level of irksomeness as Adrian Durham’s job at TalkSport to stir a pot of overly intense football fans with an antagonistic devil’s advocate approach which must make TalkSport thousands everytime Drivetime is on.
I digress. As usual.
I try to make the metrics I use as per 90. The ‘per 90’ is a number all on it’s own; this is the total number of minutes played divided by 90 – this then provides a number to weight the actions of a player. For instance, if Harry Kane scored 20 goals in 2500 minutes, it’s good but it might not be as impressive as Papiss Cisse scoring 14 times in 1500 minutes.
|Player name||Non-penalty goals||Minutes Played||Per 90||Non-penalty goals per 90|
|Harry Kane||20||2500||2500/90 = 27.7778||20/(2500/90) = 0.72|
|Papiss Cisse||14||1500||1500/90 = 16.6667||14/(1500/90) = 0.84|
This means that per 90, we could infer that Cisse could get more goals than Kane. I’ll point out a caveat here – The purpose of stats is not to explain, but to provide an alternative idea which isn’t the eye test (what you see as the case). It provides a sense of perspective – Which is why I’ve always been very guarded about players like Cole Skuse and Jay Tabb.
The above table leads me nicely in to non-penalty goals. Penalty goals are successful (~70?) percent of the time, and some players are penalty takers, which would skew their numbers (see, Grant Leadbitter) – Plus are repeatable instances. Non-penalty goals do what it says on the tin.
Assists are assists (remember, p90 means per 90 minutes played (so this is calculated by total minutes divided by 90 to show a more even metric than ‘per game/appearance’)
Key passes are the number of passes which either lead to a shot or a chance – I’m not sure which. But they lead to an attacking action, basically. So this is why I used KP + A per 90 – This shows the attacking instance as well as the number of passes which were actually converted into goals.
Goal Contribution is a metric that’s calculated using Non-penalty goals + Assists per 90 – So a player’s direct contribution to a goal.
Turnovers (Trno p90) is a combination of number of times dispossessed and unsuccessful touches – I refer to this as the ball loss, whereas tackles and interceptions (Tkl+Int p90) are possession regains. This is one of my favourite metrics to map – though this is usually skewed toward defensive minded players, it’s interesting to see how players get back and ‘work’ – when you don’t have the distance covered data.
Another one I use is DngrZ % Conversion & Shots – Shots is merely the number of shots taken from within the penalty area and six yard box (so the percentage of all shots taken from ‘good’ areas – though this would probably include stupid angles too, which is where stuff like Expected Goals models end up being better than the stuff I write about) whilst the DngrZ Conversion % is the number of goals scored from these ‘good’ areas over the amount of shots taken from them. Easy.
The below list is now from WhoScored – Which breaks down some of the more contentious metrics that are used.
Aerial Duel Won
– Winning a header in a direct contest with an opponent
– A pass that directly leads to a chance scored by a teammate
– Average number of passes attempted (short passes, long balls, through balls, crosses)
– Prevention by an outfield player of an opponents shot reaching the goal
– Action by a defending player that temporarily removes the attacking threat on their goal/that effectively alleviates pressure on their goal
– An attempted/accurate pass from a wide position to a central attacking area
– Being tackled by an opponent without attempting to dribble past them
Dribble (Successful Dribble)
– Taking on an opponent and successfully making it past them whilst retaining the ball
– Being dribbled past by an opponent without winning a tackle
– An illegal manoeuvre by a player that results in a free kick for the opposing team (does not include offsides)
– Preventing an opponent’s pass from reaching their teammates
– The final pass leading to a shot at goal from a teammate
– An attempted/accurate pass of 25 yards or more
Man of the Match
– Player with the highest rating in the match
– Being caught in an offside position resulting in a free kick to the opposing team
– The last man to step up to catch an opponent in an offside position
– Percentage of attempted passes that successfully found a teammate
– An attempt to score a goal, made with any (legal) part of the body, either on or off target
Shot on target
– An attempt to score which required intervention to stop it going in or resulted in a goal/shot which would go in without being diverted
– Dispossessing an opponent, whether the tackling player comes away with the ball or not
– An attempted/accurate pass between opposition players in their defensive line to find an onrushing teammate (running through on goal)
– Loss of possession due to a mistake/poor control
– Being illegally impeded by an opponent, resulting in a free kick
Set Piece Goal
– A goal that has been scored via a set piece situation (corner kick, free-kick or throw in)
Open Play Goal
– A goal that has not stemmed directly from a dead ball situation
So this I think covers just about the metrics I use. If there’s anything else, either comment (I think you can comment?) or tweet me @Scribblr_42