To Sack or not to Sack that is the question

Again, we are pleased to have a blog post kindly scribed by guest and colleague at FC Sports Marketing, the great Steven Seggie. He titles it: “Don’t Change Manager During the Season”.
“We are only a few weeks into the new English Premier League season and already we hear rumblings about managers whose jobs are on the line if the results don’t go right soon etc. During the 2013-2014 season, the number of sackings was in double figures (if you include the departure of Ian Holloway from Crystal Palace by mutual consent) and I am sure this year will be no different. Yet these mid-season sackings make little or no sense to me.
I have always wondered whether or not replacing your manager mid-season has any impact on the performance of the team. I would hazard the guess that the primary driver for most of these sackings is the hope that performances will improve yet we rarely see any analysis or discussion in the media or elsewhere as to whether or not performances improved. This may well be a function of the fact that most sports journalists do not have the ability to conduct such analysis, nevertheless that’s not really an excuse. Luckily some academics have analysed these mid-season changes and the results make for interesting reading. There are 3 major studies that I will discuss briefly here and they all demonstrate that changing manager mid-season is an exercise in futility.
The first study that I looked at was conducted by Allard Bruinshoofd and Bas ter Weel . These authors examined all the mid-season managerial sackings/changes in the Dutch league between 1988 and 2000. A little bit of analysis led them to the conclusion that you shouldn’t sack your manager mid-season. In fact, they find that if the manager had stayed his team would have performed better than the team with the new manager.
Similar results are found in a study of Italy’s Serie A between the 2003-2004 season and the 2007-2008 season. The authors of this study, Maria de Paola and Vincenzo Scoppa found that changing the manager during the season was essentially a wasted exercise as it made no difference to subsequent performance.
Finally, Andreas Heuer, Christian Müller, Oliver Rubner, Norbert Hagemann and Bernd Strauss , in a study of the Bundesliga from the 1963-64 season to the 2008-2009 season (a total of 14,018 matches) found no impact on performance of changing the manager mid-season.
So there we have it. The results are fairly clear cut. Changing manager mid-season has no positive effect on performance and may even lead to worse results than if the incumbent had continued. While performance may not be the only reason clubs change manager, I would guess it is the most often cited reason. So for all the supporters baying for the blood of the manager after some bad results and all the club directors thinking that changing the manager mid-season will improve performance, I have some bad news. It won’t. So you may want to think twice!”
Bruinshoofd, A. and B. ter Weel. 2004. Manager to go? Performance dips reconsidered with evidence from Dutch football. European Journal of Operations Research 148:233-246.
De Paola, M. and Scoppa,V. (2008) The effects of managerial turnover: evidence from coach dismissals in Italian soccer teams, Journal of Sports Economics, 13(2), 152–168.
Heuer A, Müller C, Rubner O, Hagemann N, Strauss B (2011) Usefulness of dismissing and changing the coach in professional soccer. PLoS ONE 6/3: e17664. doi: 10.1371/journal.pone.0017664
Ah…..this is a thorny issue Steven and I am not sure it is ever what it appears to be.
Here’s how I’d look at it as someone with a foot in both camps. I work as a scout and also for betting syndicates and I also work with academics with my other hat on.
Let’s start with the central thesis here: Does replacing your manager mid-season have any impact on the performance of the team?
And as you rightly say, the primary driver for most of these sackings is the hope that performances will improve.
Actually though, sometimes declining no further (stabilising results) is a laudable achievement. It can be the best most ailing clubs can hope to achieve, for reasons I will discuss in a second.
Often simply putting the brakes on a decline in results can represent a success in real terms. And that is something that the academic researchers probably neither realise nor credit as a positive outcome.
In my experience, I’d would say the there are always a number of drivers at play when managers are sacked but while we are consistently told that football is a results business, pure and simple, it is amazing the extent to which results are used to justify decisions that emanate from another place entirely.
Results seen in isolation are often a subsidiary factor in deciding a manager’s fate. A sacking is usually a manifestation of complex issues that the human custodians of a club (board of directors, football staff, players, fans) and the media are constantly evaluating in good and bad times.
Share price value, a general level of boredom with the status quo, a fear that the status quo will have a negative effect on cash revenues and crowd levels in the short to medium term, the fluctuating personal rapport between the manager and his board, players and fans are all brought to bear on the decision process that leads to sacking the manager. Also, you can never underestimate the chairmen’s impulse to tinker when the best course of action is often to do nothing and let the short term noise blow over.
Clubs will always prioritise subjective judgements based on emotional or commercial reasoning to justify their sackings but they will invariably express them with reference to the pressure on results.
In this sense results are used as a public justification for a sacking but in themselves they have far less impact in isolation than most outsiders (and some self-deceiving insiders) probably assume they do.
Tempting as it is, a management change that makes no qualitative improvement in results short term cannot automatically be written off as a waste of time and money. Other things are always going on…..
In fact maintaining the status quo results-wise while improving or stabilising other prospects in the club (commercial revenues, personal relationships within the club, relationship with fans and the jaded media (the feel-good factor), strategic planning in terms of infrastructure, recruitment and club philosophy ) can all be laudable objectives that might necessitate a managerial change.
All these things can all be improved both publicly and behind the scenes without any obvious improvement in results short-term. Running a club is always a balancing act: managing expectations for today and planning for a better tomorrow.
With that in mind here are a few knee-jerk thoughts:
1. Managers can really only ever work with what they have at their disposal. If a sacking happens mid to late season then the opportunities for a results transformation are minimal. Most managers estimate they need a minimum three transfer windows to enact their own imprint on a team and of course they may not last that long in the job.
In that context, maintaining the status quo (even if it means relegation in the short term) can still be a more positive than negative outcome if the team can regroup and come back stronger.
The impression is always that the outgoing manager has been sacked because results have hit rock bottom. The fact is that while the potential for improvement is always limited based on all known factors, the potential for decline is probably infinite.
This means that for a new man, even stabilising performance can, in reality, represent an improvement when the manager who has just been sacked was likely to perform even worse had they remained in their job. Often no results improvement is probably an improvement in real terms when compared to the real life potential for further decline.
2. Everyone in a football club (with rare exceptions of lifers like kit men, tea ladies etc) operates on an accelerated cycle of turnover – short term goals, short term tenures. This is driven by football’s ‘grass is greener, industrial/cultural need to consistently renew itself to maintain its financial value, its press profile, its glamour, its prestige and its illusion of progression.
Radical, regular managerial regime change (it is always regime change rather than solely manager change) is simply part of the reality of the constant cultural revolution that exists within football clubs.
For example, the manager at my former club is working to an 18 month plan, the current manager is on a six month plan (I am guessing) and the managers in elite clubs tend to refer to their life cycle as three years max. After this point their words grow stale and players, staff, fans and press grow bored, tune out and it is time to move on.
3. Generally speaking, managers are sometimes merely overseers of results rather than shapers of them. They literally are managers rather than magicians. The expectation expressed by fans and academic researchers (that results should qualitatively improve to a significant degree) is largely an unrealistic one.
For example my scouting mentor Tony Collins (ex-Leeds and England assistant under Don Revie) can list countless examples of poor managers who owe their careers to an initial positive splash on the back of medium term groundwork in youth development, player recruitment, coaching and scouting enacted by a previous boss. Their predecessor did not last long enough to see the good things he’d put in place bear fruit. Then the new guy inherits a tailormade open goal opportunity to over-perform. The new kids on the block then in turn, get their chance to move on quickly up the ladder. But they are unable to repeat the success they first had on the back of another manager’s labour, in more adverse circumstances.
For example, such a process may explain André Villas-Boas’ early stellar success at a stable Porto side before subsequent high profile flops at Chelsea and Spurs.
The problem for managers and football in general, is that results are the dominant metric but often in the short term results are really an output rather than an outcome and a function of the unreliable testimony that regularly occurs when analysing a small sample of games.
The fact is that results, sooner or later, are the product of clubs’ wages budgets: and this is the elephant in the room for every manager.
That’s why Wycombe will ‘always’ finish lower than Chelsea in the overall league standings and will also likely be subordinate to the better resourced likes of Bury, Burton and Southend come the end of the season in their own league.
The reality is that managers should only realistically be judged in the narrow terms of how well (or not) they match the expectations implicit in the buying power of their wage budget.
At most clubs and at all levels, the room to manoeuvre is smaller than most fans and academics assume.
The extent to which meaningful improvement is possible is probably negligible without a major alteration to a clubs buying power in terms of wages and/or transfers.
Nobody, as far as I know, has ever systematically investigated which managers overachieve relative to their clubs’ wage bills. But, in my opinion, the definition of a good manager is a manager that performs merely to expectation when the potential to underperform is very high indeed. Anyone improving on that is very, very good indeed. And greats, genuine greats, are few and far between.
The leagues I work in are littered with examples of managers who fail with the biggest budgets in the league – these people are the bad managers.
Sometimes the best manager in the league is the one that secures safety season after season when he has the lowest budget, or the manager who over-performs only to fail in the play-offs from a position of being set a season’s target of avoiding relegation by the board.
It’s also hard for outsiders to judge managers on their day-to-day work. Most of the biggest managerial decisions are made in private, and the outcome of those decisions – signing a certain player, say, or hiring a fitness expert or new chief scout – might only become apparent much later and are also heavily influenced by blind luck good and bad.
Players generally take six months to a year to settle in a full-time club. So a panic buy bought in the winter transfer window will more likely than not, make a less effective contribution prior to their first preseason at their new club. Often the next manager in will be able to bask in the glory of ‘getting more from an underperforming signing’. The reality is often more prosaic as the player has merely settled and got used to his new teammates’ play.
Injuries and suspensions at key points – especially where lower league clubs are running with squads of 20 players augmented by youngsters – will also undermine a manager’s plans. This is why a key player back fully recovered from long-term injury can be ‘just like a new signing’ for a replacement boss.
To evaluate how well a manager does or doesn’t do (in any walk of life) we need a realistic understanding of the limits on what is possible. For example, current thinking in terms of team dynamics seems to suggest that any team should be appraised in terms of the impact of its worst rather than best player. The manager who signs and or selects the best worst players will give himself the greatest chance of success.
4. No two jobs are ever directly comparable. For example, it makes sense that it may be possible to consistently overachieve relative to wage budgets when a club produces its own players internally rather than relies on transfers. Arsene Wenger, George Graham (both Arsenal), Bobby Robson (at Ipswich) are examples of managers who were all able to over-perform relative to their budgets because they had the benefit of dynamic youth systems with talent on tap. Managers at serial underachievers with less good youngsters at their disposal may not show up so well. The improbable Job can become an Impossible Job with a requirement that a manager’s only short term levers in terms of improving the starting XI come from his current pool plus limited loans and transfers.
Results-wise it may make little sense to change a boss midseason but are only ever part of the equation….what we really need is a fairer metric for evaluating success and failure.