One of the main talking points of Chelsea’s season so far has been the lack of playing time for Juan Mata. There was summer speculation as to how much Jose Mourinho liked the player and the possibility that the Spaniard may be on his way out of Stamford Bridge.
It’s fair to say the fans would be disappointed if he’d have left. Due to his elegant passing, Mata has been hailed as the creative talent in the Premier League over the last two seasons. No player has assisted as many goals as Chelsea’s No 10.
Using Differentgame’s Chance Creation Model we can test Mata’s ‘key’ passing from last year against the Premier League benchmark. For clarity, Opta define a ‘key’ pass as “the final pass or pass-cum-shot leading to the recipient having an attempt at goal”. The model distinguishes between 30 different types of chance depending on where the ball was struck from and how the shooter received the ball. Each is weighted depending on ease of conversion.
From just under 100 key passes, Mata assisted 12 Premier League goals. Mapping these key passes to the model, we can calculate that the average number of assists he could have expected was just over 9.
However, take out key passes made directly from set pieces and we see assists fall to 8 with the expected average being just over 7. Considering this is supposed to be one of the Premier League’s finest talents the open play impact of Mata doesn’t look huge.
Of course Mata can do nothing if the shooter makes a hash of passes laid on a plate for him, but Chelsea as a whole were above league average at finishing chances. Also, there was next to no difference in open play conversion rate from balls Mata supplied compared to those supplied by the rest of the team as a whole.
Here’s a couple of situational positions where you’d expect a player of Mata’s calibre to excel:
Mata does seem good from these situations, however…
More than a third of Mata’s key passes were made to team mates who were outside the box, and another third of his key passes were delivered directly from set piece situations. We’re suddenly left with a not so stunning number of good quality chances created from open play such as the ones in the graphic above.
Going forward, the model will allow us to further assess single player contributions in this area. Admittedly, the model is in it’s early stages, but initial impressions are that as usual, we tend to overestimate the quality of our favourite players.
Follow us on Twitter here.