Obsession with stats doesn’t add up for Brentford
Mathematical modelling is the philosophy at Brentford this season. Yet here’s an anomaly. When, after eight games, the team were in 19th place with just eight points and a single win at home, Brentford didn’t sack the analytics experts. they fired the manager.
marinus Dijkhuizen is gone, along with his assistant Roy hendriksen, amid reports of disorganised training sessions and serious doubts about their stewardship, expressed by first-team players. all sounds a bit subjective, doesn’t it, for a regime run by numbers?
almost as if there is a little bit more to the game than data; maybe an organic feel, a bent for improvisation and intuition that cannot be gleaned from a computer print-out.
Brentford owner matthew Benham does not believe this, of course. he made his fortune from statistical evaluation and gambling — and has ploughed upwards of £90million of it into Brentford, so credit for that — and believes football can be approached the same way.
having won the Danish Superliga as owner of midtjylland, he wanted to pursue the same blueprint in the championship. it was this that led him to part company with mark Warburton, who last season took Brentford improbably into the championship play-offs. except the analytics didn’t see it like that. midway through the season when Brentford were fifth, Benham’s data suggested they had been lucky and should really be 11th.
Now here’s where statistics go wonky. having agreed that Warburton should leave in may, Brentford maintained their position and came fifth anyway. So they were either lucky for 46 games straight — and ask any successful gambler the odds on that, by the way — or they really were the fifth-best team in the championship.
this, then, gives the analysts who pegged them as 11th a 25 per cent margin of error, six places wrong in a 24-team league. and wouldn’t we all like to be cut that slack?
imagine if George Osborne were 25 per cent out on all his calculations. it would sink the country. if chelsea had 25 per cent more points now they would be in the top half of the table; if manchester United had earned 25 per cent more points last year they would have won the league.
No doubt Benham has got a set of figures confirming that Dijkhuizen was underachieving, yet so much of what has gone wrong at Brentford this season cannot be found in the realm of analysis.
the pitch at Griffin Park was left in an abysmal state following the installation of undersoil heating. Record signing andreas Bjelland, bought for £2.1m from Dutch side Fc twente, was ruled out for the season after injuring a knee on his debut. Josh mceachran, another costly signing from chelsea, fractured a foot preseason and is not expected back until November. Jota will be returning at roughly the same time, having damaged ankle ligaments in the opening championship game against ipswich.
Brentford were buying by numbers, too, Moneyball style. Yet did anyone research the probability of their new centre half lasting 45 minutes of a capital One cup tie against Oxford United, and this being his only contribution to the season? No. Such information does not, and cannot, exist.
Brentford will have known how many crosses Bjelland can repel, his effectiveness in the opposing box, they will no doubt have studied his Prozone performances and may even have a vague calculation of how many points he could be worth. But not on one leg. lame, none of it matters. manchester city have 10 analysts going over matches but that hasn’t stopped them losing three games on the spin in the two major competitions, the last by a 4-1 margin at tottenham.
Now the data may say differently but to most eyes the big problem is that city are not the same without Vincent Kompany at the back and the other centre halves are not as effective.
Yet as we live in times when it can be statistically argued that Barcelona are better off without lionel messi — a win rate of 89 per cent in the games he has missed, against 71 per cent when he is in the side — who knows what information manuel Pellegrini is being fed to explain this slump?
lee carsley is the new man in charge at Brentford and will be expected to comply with the same statistical formula as his predecessors. if he cannot turn the club around, however, he should know where the buck stops. Not in the computer room, he can be sure of that.