Home > Idle Speculation > Should Emile Heskey be playing for England?

Should Emile Heskey be playing for England?

Warning: This is an attempt to apply back-of-the-envelope calculations to answer one of the more controversial questions in English football – why does a striker who rarely scores goals start for England? It features the use of extremely dodgy guessing and dubious statistical techniques. You have been warned.

My understanding is that the main reason Heskey starts alongside Wayne Rooney is that he helps other players score goals. Indeed, given that Heskey himself scores less than one goal for every eight appearances, I can’t think of any other reason for him to even be in the squad (let alone the first choice team). So, the question is: does Heskey’s ability to hold up the ball, occupy defenders and release other players make up for the fact that he himself is not a goalscorer?

We can at least get a crude idea as to whether this is even likely to be true. Let’s say the players that are most likely to benefit from having Heskey in the team are Rooney, Lampard and Gerrard. We know Rooney likes playing with Heskey, and Lampard and Gerrard (as evidenced by the USA goal) are known to make runs from midfield. Rooney’s scores about .4 goals per appearances, Lampard .25 and Gerrard .21 (this is simply based on goals divided by caps). Let’s assume that Heskey increases their chances of scoring by 50%, which I suspect is extremely generous. Now, if we simply add these three players scoring rates together we get .86 – 50% of which is .43.

Heskey scores approximately .12 goals per cap. So, if we add Heskey’s goal scoring to the extra goals he creates for other players, we get .55. Incidentally, that is almost exactly the number of goals Peter Crouch scores per game (.54). So, even if we assume Heskey increases the chance that Rooney, Lampard and Gerrard score by 50% – which is an extremely large increase – he isn’t obviously increasing the total number of goals scored by the team when we compare putting Crouch in his place.

Now, in order to be a better candidate than Jermaine Defoe, many of whose appearances are as a late substitute, then the Heskey effect needs to be at least 20% – which when you remember is 20% over and above Defoe’s ability to create chances or others is still a massive difference. This kind of back-of-the-envelope thinking suggests to me that any ability Heskey has to create goals for others is unlikely to make up for his low goalscoring.

Obviously, this is no way to go about picking a football team, otherwise Crouch starts looking like one of the best international strikers in the world. Furthermore, I have not separated the ‘ Heskey effect’ from the past stats, which I could have a stab at if I bothered to make a few assumptions, and maybe at some point I will. Everything I have said needs to be taken with a huge grain of salt (or is it many grains of salt? I’m not sure). All I am suggesting is that Heskey needs to be very significantly better than either Defoe or Crouch at helping create chances for other players in order to make up for his own lack of goalscoring. Furthermore, one way that Heskey might create chances for other players is by neglecting to take one that any other striker would, hence not affecting the total number of goalscoring opportunities anyway. I therefore remain as unconvinced as everyone else in the country that Heskey should be starting, but at least I have some questionable maths to back it up…

Categories: Idle Speculation
  1. David Franklin
    July 1, 2010 at 10:28 am

    Just read this – I would’ve thought the easiest way to test Heskey’s effectiveness to the strike force would be to compare the number of goals per game scored by the whole team, with and without him. I did a bit of analysis myself – since his first start in 2000, he has started 40 matches for England, in which we’ve scored 77 goals (an average of 1.93 goals per game). Since that date there have been 56 matches in which Heskey has not played a part (I’ve ignored completely those matches in which he’s come on as a substitute), in which we’ve scored 125 goals (an average of 2.23 goals per game). So it would appear that having Heskey in the starting line-up costs us 0.3 goals per game on average, which (funnily enough) is about the difference between his average (0.12) and what a good striker should be scoring.

    [Just to check, I also did it while only taking into account results against the top 80 in the world – I thought perhaps that Heskey was being unfairly treated because he missed out on lots of games against Andorra, Jamaica, Trinidad & Tobago etc.. In the ‘real’ games, though, we score 1.78 gpg with him and 1.96 gpg without, so there’s still a substantial difference.]

    • David Franklin
      July 1, 2010 at 10:30 am

      (note – at the start of the comment, the ‘read’ in “just read this” is a past participle, not an imperative!)

    • July 1, 2010 at 11:10 am

      I actually prefer my way of doing it, because it is a more direct test of the plausibility of the theory that Heskey’s other qualities make up for his lack of goalscoring – when we realize (very roughly) how large his effect on other’s goalscoring has to be, the fact that it is an extremely crude method becomes less relevant. Kind of like a Fermi problem; you can get the order of magnitude right with extremely sketchy assumptions, and I think that it is highly unlikely that the order of magnitude of the ‘Heskey effect’ is what it would need to be.

      However, the truth is the main reason I chose this way of doing it was that the data could be obtained in two minutes simply by looking at each player’s wikipedia page. But I am always pleased to come across more evidence (insert caveats here) that Heskey should not be in the squad.

      Substituting him on against Germany, was quite simply, insane.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: