There is no disputing that Diego Costa is Chelsea’s – and probably the Premier League’s – best-performing striker right now, but that shouldn’t preclude Antonio Conte’s side from signing another forward this winter.
Chelsea’s interest in World Cup-winner Fernando Llorente has been met with widespread bemusement but clearly Conte does not trust Michy Batshuayi to lead the line and good quality cover will likely be required for Costa through the second part of the season.
While it’s a reasonable bet that Chelsea will win the league - and Costa finish top scorer - there are no guarantees. Therefore the sensible thing to do would be stock up in January.
While Llorente might not necessarily end up being the best choice for Conte and Chelsea it is apparent that someone is going to have to relieve the burden on Costa sooner or later.
We’re into unchartered waters as far as Costa’s Chelsea performances go with the Brazilian having scored 14 times already for the Premier League in only 19 matches.
Costa is enjoying a goal scoring hot streak like never before, netting roughly one in every four chances he gets (24 per cent) according to Opta data. The average conversion rate for a Premier League striker is closer to one in six (17 per cent).
While there are no indications that Costa will slow down, he will still need to outrun expectations in order to help the Blues keep their noses in front.
That is because Costa is converting far more chances than would reasonably be expected. His 14 goals this season have been rich and varied; tap-ins that owe plenty to positional play and long-range efforts from outside the box alike.
Opta’s Expected Goals (xG) model anticipates that Costa’s goal haul should be closer to 6 and not the 14 already netted.
“Expected Goals is a metric that quantifies the quality of a given chance a player has,” says Tom Worville, a data scientist at Opta. “Shots are assigned an Expected Goal value, between 0 and 1. A shot with an Expected Goal value of 0.05, for example, is scored 5% of the time.
“The Expected Goals model takes into account a variety of factors, such as distance from the goal, the angle to which the shot is taken from the goal, the type of shot - headed v foot - and the assist type - through ball, cross - among other factors.
“Expected Goals is also a good indicator of how likely a team or player is to perform in the future. The nature of football as a low scoring sport means that there can be a lot of noise in a player’s conversion of chances, so Expected Goals can give us a better understanding of why they are scoring so many or so few.”
Costa scored 19 goals in his first season for Chelsea with his xG tally at 13.68. He managed 12 in his second season with his xG standing at 12.55.
Furthermore Costa has a patchy injury record at Chelsea, even if he’s managed to stave off any significant knocks so far this season. The most minutes he’s ever played for Chelsea came in his first season – a total of 2379 minutes played.
The 28-year-old has already played 93 per cent of available minutes for Chelsea this season – a total of 1681 of 1800 available. How a leaner, meaner Costa will fare during the title run-in remains to be seen.
Either way it is prudent for Chelsea to mitigate the risks of burnout, injury or suspension by supplementing their attack line with signings and giving Costa a break when possible.
Costa’s over-performance, meanwhile, is mirrored by his club’s own form. Manchester City, Liverpool and Tottenham are all within five points of their simulated total points by expected goals but Chelsea are ten points better off than we would expect based on the quality of chances create and conceded. That in itself suggests that they might not be quite the runaway ‘best team in the league’ despite their significant lead at the top.