The stats behind Arteta’s influence
The closing of the transfer window tends to throw up a collection of statistics aiming to quantify the influence of certain players on the teams they have left.
These stats are often misleading. Take Phil Jones:
34% – Blackburn’s PL win percentage with Phil Jones playing since he made his PL debut, compared to just 25% without him in the team. Hole.
In absolute terms, Blackburn won 12 of the 35 matches Jones played for the club, and won 3 of the 12 in his absence since debut. Statistically speaking, however, this difference is insignificant because the win frequency and sample size are both small; there’s nothing to suggest Phil Jones was the reason for the fall in win percentage.
In fact, the fall is largely attributable to Jones missing harder matches. Whilst injured during the winter last season, his team played a number of solid mid-table and top half teams, matches they would normally expect to struggle in anyway.
So naturally, when Opta tweeted:
10 – Since January 2005, Everton have averaged 61 points per season with Arteta playing, compared to 51 points without him. Lynchpin.
I was sceptical as to how significant this difference was. Can one man really have that big an effect on the team? The stats will never give the whole answer, but they can give a few pointers.
At first glance, it seems as though Arteta’s presence is all-important. As alluded to the in Opta statistic, a win percentage of 43 equates to about 16 wins in a season, level with fifth-placed Spurs in 2010/11. The Arteta-less win percentage equates more to a mid-table finish with about 13 victories.
But like Phil Jones above, it’s important to know which games Arteta missed:
Everton not only tended to play the majority of their games without Arteta away from home, but they also tended to play marginally more difficult opponents. As such, games with Arteta were largely games Everton might’ve expected to win, and games without him were matches they might’ve expected to be tougher.
In fact, taking into account the location of matches and the strength of the opposition (measured by their season points total), Arteta’s playing presence had statistically no significant impact on Everton’s win percentage.
As always though, statistics don’t provide the full picture. It’s obvious Arteta was a key player for Everton, if perhaps less so since his cruciate knee injury in February 2009. But what the stats do suggest is that Everton perhaps shouldn’t get too carried away with his absence.
More likely, defeats and draws can be attributed to a series of injuries, suspensions and sales, as opposed to the loss of a single individual.
In the end though, this is as much an exercise in understanding the deceptive power of statistics as analysing the influence of Mikel Arteta on Everton’s results. Time will very much tell us the true effect of Arteta’s departure.