It’s the stage of the season where nearly everyone is asking questions of the manager of their football club. What people should be doing is asking questions of the board of their football club.
Back in September, Aston Villa extended Paul Lambert’s contract until 2018. Early in the season, Villa had been flying high and second in the league. The underlying numbers told a different story: Villa were actually the worst team in the league. Five months later, Lambert was sacked.
After a record breaking first season at Everton, Roberto Martinez was handed a new and improved 5 year contract. Everton over-performed by around 9 pts according to my numbers. A lot of it was based on defensive luck.
Despite what the league table said, Everton had a good solid season, not a great one and this year the club has just about avoided getting involved in a relegation scrap.
Put simply, (most) football clubs do not have a hold on the true level of their performance. The football world is slowly changing, though. Dutch writer, Michiel de Hoog wrote recently about Danish club FC Midtjylland and its Chairman, Rasmus Ankerson.
Ankerson was installed as Chairman by Brentford owner, Matthew Benham, who also owns FC Midtjylland. Benham owns Smartodds too (‘we provide statistical research and sports modelling services to customers, including professional gamblers’). Ankerson is co-founder of 21st Club Limited (‘our expertise lies in talent management, data analytics, high performance’).
21st Club’s other co-founder is Blake Wooster – previously Prozone’s Buiness Development Director. Head of Football Intelligence there is Omar Chaudhuri, again formerly of Prozone, and an economics graduate. Back in Brentford, they recently hired Ted Knutson, who built the live sports department at sportsbook Pinnacle before becoming their lead trader for the Premier League.
What you have there is some serious experience both in and out of football. Guys that have analytical smarts and experience of building their business worlds from the ground up. These are people that wouldn’t have said: “Here’s a huge new contract for winning a few games during a hot streak.” In fact, here’s what Ankerson did have to say:
‘We have a manifesto now, which says that the model rating always overrules the league position in evaluating our performance. The table is lying. We have been the best team in the league this season, but nine points ahead is probably not a fair reflection of the difference between us and Copenhagen. The difference is that we got lucky in some games and they got unlucky. This is what I keep telling the coaches: don’t think we’re good just because we lead the league. We’re good when our model says we’re good.’
18 months ago, Brentford’s Knutson wrote about a framework for assessing managers. If I was running a football club, coming up with such a framework would be the first thing on my agenda. So I had a go at it this week…
I used the same expected goals concept I linked to above (raises glass to @bertinbertin). It basically measures a side’s attack and defence against the league benchmark. I included data for all managers who’ve been in charge of a Premier League side for at least 10 games over the last 5 seasons.
However, for today, I’ve just concentrated on attack:
The horizontal axis is just expected goals for per game (summed not simmed). The vertical axis is expected goal efficiency – basically how many goals a side did score against how many they were expected to score. If it’s above 1.00 the team’s attack has been more ‘efficient’ than the average side and has scored more goals than the average side would have with the same chances.
Immediately we see Louis van Gaal at the top of the graph there. His Manchester United side create less goal opportunity per game than the average side has over the last 5 years. Yet there he is, the most ‘efficient’ attacking manager there’s been in that time.
Van Gaal has a reputation to match the cluster of big name managers in the top right quadrant. He just doesn’t have the underlying numbers to match them. This should immediately ring alarm bells. He must find a system to create more goal opportunity because that efficiency figure will not last. The Premier League is a ruthless eco-system, teams will figure his side out and luck doesn’t last forever.
You’ll see that van Gaal’s circle is small. This is because he’s been in charge for a relatively small number of games. Tim Sherwood’s circle is small too as he embarks on his managerial career. Both Sherwood’s goal creation and goal efficiency numbers are impressive so far.
Perhaps in part, that’s been down to inheriting a decent team at Spurs. Goal creation is at a similar level to Redknapp and Moyes. Is this likely to continue at a team like Aston Villa? Probably not. You’d also have to consider him on par with the elite managers to sustain that kind of efficiency over a decent period of time. It’s likely that Sherwood will move down and to the left on the chart by the end of next season.
Of course all of this data is raw. It doesn’t take into account spending power in the transfer market or squad wage bills. It doesn’t tell you the style of football played or whether the teams were effected by busy Europa League schedules or injuries. The point is you could factor all these things in to get a more realistic picture of how your manager is doing. You could do it for every league to keep your eye on the market too.
There’s plenty of talk this week about Manuel Pellegrini’s job being on the line. This comes less than 12 months after he won the Premier League. Looking at the chart, it doesn’t look like any of the ‘great’managers who’ve also had access to big budgets could do a better job – attack wise at any rate.
Give the chart a good study. There’s plenty of manager comparisons and stories to be made. Hopefully soon, the board of your club will keep analytical tabs on its manager (and other potential candidates) instead of rewarding or firing them them based on the back of a few results.
Follow me on Twitter @footballfactman