[by Johnny Phillips, January 2016]

Over any medium-to-long term period, the results that a professional football team achieves is almost entirely down to the amount that they spend on player wages.

This is because a pro football team’s level of performance is dictated to a huge degree by the innate level of talent of its players. Better players command higher salaries, so the richest teams get the best players and become the best teams.

The salary market in top level pro football isn’t perfectly efficient, but it’s pretty damn efficient. So the biggest salary spending teams finish at the top of the leagues, the smaller ones down the bottom. There will be the occasional exceptions (like Chelsea doing badly and Leicester doing well this season) but these are always temporary blips, in an otherwise neat correlation between salaries and league points.

So by logical extension then, the degree to which a football coach can impact on the performance of his team is limited. A genuinely incompetent coach can make a big mess of things of course, and coaches can certainly influence playing style. But the way that a team lines up and tries to play has far less influence on its results than the raw talent of its players.

The reality is, contrary to popular footballing wisdom, who a professional team employs as its head coach really doesn’t matter all that much – so long as they are basically competent. It’s like a Formula One racing car – the success any car has is only marginally attributable to the driver, in the sense that you could take any competent pro racing driver from anywhere on the grid, put them in the fastest car, and it would finish towards the front of the pack in qualifying and races. The best coaches/drivers may have some gravitation towards the best teams/cars, but it mostly what’s in the squad/under the hood that matters.

In both F1 and football most people ascribe far too much significance to the person in the driving seat. This is a good example of a common human failing, which is our tendency to fall for ‘narrative fallacy’.

Football managers/coaches are high profile representatives of their clubs, so it is an attractive story to imagine that they have a great influence on the club’s success. But it isn’t true. The coach just happens to be the most visible member of what is often a huge staff which supports a pro football team. So even the relatively small amount which coaching matters is not all down to ‘the coach’ anyway.

Journalists are paid to create interesting narratives, so when they anoint someone as a good/excellent/brilliant coach they are only doing their job. But they don’t really know what a coach actually does every day any more than you do. Unlike football players, the vast majority of a coach’s job is done away from public gaze. So we’re all just guessing what they actually do, and what influence it has.

So the general rule of thumb is that a football coach gets far too much praise when his team plays well, and too much blame when his team plays badly. While there are a few exceptions, and exceptional coaching can have a small benefit, team success is almost all down to how good the players are. And the best measure of a player’s level is the salary that he commands.

This rule works out extremely well over a longish term sample size like, say, 3 league seasons. But a single season of fewer than 40 games is only a short/medium term sample size in a low scoring sport like football, where roughly 65% of games finish in either a draw or a one-goal victory. The margins between success and victory are small, and so luck (or ‘randomness’ as we prefer) plays a big part in a small sample.

The shortish term nature of a single season represents an opportunity for smaller clubs. If they can ride their luck for a while they can enjoy a spell in a higher league position than their salary budget justifies. But a stand-out season represents a danger as well, especially to the coach, because it runs the risk of creating inflated expectations.

Clubs/coaches should be judged on their performance versus salary spend, but rarely are. Fans (‘fan’ is short for ‘fanatic’, remember) and the media are fundamentally indisposed towards rational objective measurements of a team’s performance. It is not in their interests/nature. Instead expectations are often based on bad predictors of performance like where the team happened to finish in the league season the previous season. This is very dangerous for a coach.

Garry Monk is an excellent example. Last season Swansea finished 8th in the league and Monk was being feted as a brilliant young manager. But that final league position was flukey.

At OddsModel we like to look at underlying numbers to get a better idea of teams’ true levels, including creating analytical computer models. But even just looking at goal difference can tell you plenty. Swansea’s goal difference was -3 last season, which was the same as West Ham in 12th. Southampton were a much better team and had +21 GD but finished just one league position ahead in 7th.

Fast forward a few months and the Swans had regressed to just about where they should be, in the bottom third of the Prem Lge table. But the unreasonable inflated expectations generated by the over-performance in the previous season created the illusion they were getting worse, and Monk got sacked. Of course they haven’t improved since he was dismissed. They continue to play at almost exactly the level you would predict given the talent in their squad, and the salaries they pay.

Monk probably didn’t deserve all the excessive praise he got after last season, and he certainly doesn’t deserve to be regarded as a managerial failure now. But that’s the inherent risk in having a one-off good season. It’s good fun at the time, but it can be career suicide for a coach.

Alan Pardew’s time at Newcastle is a good case study. In the 2011/12 season Newcastle finished an eye-catching 5th in the league table, but with a GD of just +5. That was just 6 goals better than Sunderland down in 13th. Chelsea finished 6th that season with a GD of +19. But Pardew was rewarded with an 8(!) year contract and the Geordie fans got a bad dose of inflated expectations – something Newcastle fans are more prone to than most. There can’t be many of them who held up those ‘Pardew Out!’ banners who think their team has improved for his departure, and don’t feel a bit embarrassed about it now.

Brendan Rodgers’ time at Liverpool is an interesting case too. The media and Liverpool fans are prone to inflating their expectations unreasonably for historical reasons, as befits five time European Champions and eighteen time champions of England. But the reality of their recent past is that their salary budget has been about the 5th biggest in England, and 5th is generally about where they have finished in the league.

But in 2013/14 they came within a single extra win of winning the league. It was the inflated expectations that came from this outlier season that meant Rodger’s job came under threat after a 2014/15 season where they finished 7th. So really it was the 13/14 season where they finished second that cost Rodgers his job, rather than the 14/15 effort. If they had finished 5th in  13/14 he would probably still be in the job now.

The Liverpool/Rodgers case throws up some interesting additional points; a big transfer budget is not as important as a big salary budget. A player’s salary is the best gauge of his level of ability. Transfer fees are influenced by a lot more factors such as age, length of contract remaining etc. So while Liverpool have spent lots of money on a lot of transfers that doesn’t mean they ought to be finishing above the likes of England’s ‘big 4’.

‘Moneyball’ doesn’t work. Liverpool are owned by Fenway Sports Group who also own the Boston Red Sox, and are proponents of using sports analytics. But if they, or any optimistic fans think that it’s possible to use Moneyball techniques to bridge the gap between a 5th largest salary budget and regular league championships they are going to be disappointed. The Red Sox have just finished bottom of their American East division for the third time in four seasons, and owner John W Henry has announced that they will now ‘de-emphasise its reliance on analytics’. The intelligent use of analytics could theoretically deliver some efficiency gains to a football club (although there’s not a lot of evidence that it has yet) but these gains are insignificant next to the salary budget rule.

Liverpool’s 13/14 season actually wasn’t a genuine outlier, as in something freaky that was based on luck relative to salary budget. Because, while a little complicated to keep track on, a better way to measure how much innate talent a coach has at his disposal is to measure the salaries the players are getting something like 18 months hence. There can be a natural lag between a player displaying the full extent of his ability, and his salary catching up to reflect it. So when we are measuring Rodgers’ deployment of his 13/14 Liverpool resources it would be fairer to count the salaries Suarez and Sterling are getting now, rather than what they were getting then (recall too, that remarkable though it seems now, Rodgers had the benefit of a fit Daniel Sturridge for almost the whole of that season).

Rodgers was replaced by the charismatic German Jurgen Klopp, a coach who benefits hugely from positive media narrative building. Liverpool haven’t improved one jot since he arrived.

Anybody imagining that veteran coach Claudio Ranieri has suddenly turned into a football genius to produce an improbable Leicester City title challenge should probably wait until he is managing a squad minus the £100k+ a week earners that Vardy, Mahrez and Kante will be in a year’s time.

Being known as a ‘great’ football coach has much more to do with opportunism than it does with ability. Players make teams, not coaches – who have only a small influence over how well their team plays (we are making a distinction here between a ‘coach’, and a ‘manager’ who can be more involved in the genuinely critical task of player recruitment).

So the knack for being regarded as a great coach is to navigate yourself into jobs where your team’s salary budget is greater than that of your rivals. That way you will have better players, get a better team and the media will help to start to construct a narrative of your brilliance. Guardiola and Mourinho are probably no more than average coaches, but they have managed their careers brilliantly – always following the money. Always getting a drive in the fastest car.

And if you are the coach of a team which out-performs its salary budget in a season, the smart thing to do is to move on straight away before you get found out. You almost certainly didn’t do anything that really caused your team to over-perform – you were just in the right place at the right time.

Staying where you are and crossing your fingers that lightning will strike in the same place again is not a smart move. In fact, staying put and copping the blame for a team’s deterioration, which is in fact nothing more than its natural regression to the mean – well, that’s just career suicide.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s