Debate has been raging this season between FPL managers about which Manchester City midfield player is the best addition to their teams. The Citizens’ £55m signing of Kevin de Bruyne on August 30 has refreshed discussion as managers assess the likely impact on their teams.
Clearly it will be a few weeks before we have sufficient data to judge de Bruyne and how his team mates work with him. What we do have is four weeks worth of data from the current starting midfielders: Yaya Touré, David Silva, Raheem Sterling, Jesús Navas and Fernandinho. This is too small a sample to make a high quality assessment, but it may throw up some pointers and give us a very rough baseline from which to re-assess later this year.
I’m not going to spend too much time on Fernandinho and Jesús Navas, who is widely predicted to lose his place to the new signing from Wolfsburg. Fernandinho has had two shots all season – and scored from both of them. If he keeps this up this 100 per cent conversion rate throughout the season, we will have witnessed the birth of a new FPL phenomenon unlike anything that has gone before… but I doubt that is going to happen. He has also created five chances for colleagues so occasional points returns are not out of the question.
Navas has hit five of his six shots inside the penalty box and two of those six were on target. He has created seven chances, which is one more than Sterling, but I’m going to leave him out of further discussion because he is thought the most likely to suffer a loss of game time following de Bruynes’s arrival.
So how have the other three got on?
The eagle-eyed will spot that I have followed the Premier League’s lead and credited Silva with the disputed first goal against West Brom in Gameweek 1, which in FPL has been credited to Touré. I dislike getting into “if only that had been different” scenarios because they could go on without end, but I’ve made an exception here because the Premier League did the same. Please delete one assist from Touré and give him Silva’s goal if you disagree.
What the radar chart shows is that while Silva is creating a few more chances for his colleagues, Touré is taking a few more shots and getting more on target.
The interesting one to watch going forward will be Sterling. He has not been shooting as much as the other two, but his 62.5 per cent accuracy is twice that of Touré (30.8 per cent) and thrice that of Silva (20 per cent).
The next chart shows how their performances have translated into points per 90 minutes played (P90) and P90 per million pound each player costs (p90/£), which is an important consideration when Silva currently costs 1.3m more than Touré and Sterling. I have run the calculations with the first goal against West Brom credited to Silva (Adj.) and to Touré (FPL) and altered the bonus points accordingly.* I also display P90 results based on the players’ start-of-season prices and current prices to aid consideration over whether to hold or buy.
Looking at the FPL data, Touré looks the best performer so far particularly when value for money is considered and there isn’t much separating Silva and Sterling. However, with the adjusted data, Silva looks the better performer and there is little difference between him and Touré in value for money. If you buy Touré, the question is whether you can squeeze more than one point per 90 minutes out of that spare 1.3m – but that’s a question for another day.
As a final thought, the difference the attribution of that one goal against West Brom makes to the results of these charts shows how just one event can have quite a big effect on a small data sample. Therefore it would be prudent not to leap to conclusions based on the evidence we have so far.
- *By my calculation, Touré would have still received three bonus points from the West Brom game if the first goal was credited to Silva, but Silva would have come away with two bonus points (FPL gives him none).
- I have been struggling to do this post through a bad cold. I think I have the calculations correct, but please let me know kindly if you spot a mistake.