The aim of this article is to offer fellow Evertonians an alternative approach to thinking about football and analysing football. I urge readers to be open minded and tolerant of this article and I hope it gives you something to think about.
In my opinion, statistics are often misused in aiding football arguments. A lot of figures can make teams or players appear better than is the case because standard season by season goal numbers do not take into account that players can have “one of those seasons” where everything just goes in or they just can’t find the net. A well known example of this was the Grant Holt case in 2012. After Wayne Rooney, Grant Holt was the top English goal scorer in the Premier League that season which led many to argue that he should have gone to the Euros that summer. In hindsight, that seems like an absolutely absurd suggestion.
However, a recent breakthrough in football analytics is Micheal Caley’s “Expected Goals Model”. This is a formula that measures the quality of every shot a player takes, and hence the number of goals the player is expected to have scored. If you want to read more about how this model was made, read the article below:
This model has been proven to have a stronger correlation to future goal numbers for players than just general goal numbers which is why it should be taken very seriously as an analytical tool. These numbers help predict whether a player really deserved to score the number of goals that they did or whether they just had a fortunate season.
I will be using this model to analyse the data for Everton’s attacking players this season and will determine Everton’s attacking threat. The absence of players such as Kone, Niasse, Osman and Pienaar is because there wasn’t sufficient data for them.
|
Name |
Open Play Expected Goals (to nearest whole number) |
Efficiency |
|
Lukaku |
14 |
1.18 |
|
Barkley |
5 |
1.24 |
|
Lennon |
3 |
1.96 |
|
Mirallas |
2 |
0.84 |
|
Deulofeu |
2 |
1.1 |
|
Cleverley |
1 |
0.84 |
The table above orders how effective the Everton players above have been at shooting this season. Lukaku and Barkley are unsurprisingly at the top of this table whilst Cleverley is unsurprisingly at the bottom.
The more interesting column however is the efficiency column. This is the ratio between how many goals the player should have scored (the expected goals figure) and how many they actually scored. If a player scores considerably more than their expected goals figure (an efficiency of over 1.3) then it is likely that the player has been lucky with how many goals they have scored and therefore will be be likely to score less the following season.
The data therefore suggests that Lennon has been fortunate and has overperformed this season in terms of goals. If a player scores below their expected goals figure (an efficiency of less than 1) it is likely that the player has been unlucky this season. This implies that Mirallas and Cleverley have had off seasons in terms of goals and that they should score more next season. Meanwhile Lukaku, Barkley and Deulofeu have all performed as expected.
When comparing the expected goals numbers of Everton players to other players in the league, Lukaku really stands out. Only Harry Kane has higher in the league, which shows just how valuable an asset he is to the club and how important it is that Everton keep hold of him this summer. Barkley also ranks impressively, surpassing the likes of Ozil, Eriksen and Payet with only Dele Alli as a defined number 10 ahead of him in terms of goal scoring ability. Everton’s other attacking outlets prove less impressive, all with figures lower than recent departure Steven Naismith, which leaves the question behind to whether it was the correct decision to let him go.
If this data proves correct, there isn’t much to worry about on the attacking front. As a team, Everton rank 7th in the league in terms of expected goals, above Man United, Chelsea and West Ham. Despite the prediction of a Lennon decline, more goals from other sources such as Mirallas, Cleverley and new signing Niasse can aid Lukaku and Barkley in our attacking efforts.
Does this data match your opinion? Has Lennon been lucky? Have Mirallas and Cleverley been unlucky? Comment below! Let me know whether you would like to see more of this kind of analysis, all feedback is appreciated.
By Ed Shorthouse (Twitter – @ed_shorthouse)
