WHAT THE RESEARCH SAYS

A bad run, a sacking,
then the bounce.

Football hands researchers a weekly scoreboard of managerial decisions.

Economists have used it to ask a simple question: when a club sacks its manager mid-season, do results really get better because of the change? The answer from several studies is mostly no.

RESULTS ≠ QUALITY luck plays its part

Check a situation

An empty football dugout beside a floodlit pitch at dusk, with rows of plain seats behind — AI-generated image, not a photograph of a real person or place.

Bad runs end anyway
Results are noisy
Judge the play, not the score
Three lessons from the data

THE SHORT ANSWER

Teams recover,
with or without a change.

Three findings from studies of head coaches
in the Dutch and big European leagues.

THE BOUNCE

Better, but not because of it

A study of 14 Dutch top-flight seasons found teams did better after replacing the coach, but a control group of similar bad runs without a change improved too.

See the numbers

OUTCOME BIAS

Results over performance

Across the big five leagues from 2017/18 to 2024/25, clubs sacked managers after poor results even when expected goals showed the team was playing well.

Check a situation

LUCK

Regression to the mean

If a bad run is partly bad luck, better results tend to follow on their own. Researchers call this regression to the mean, and it explains much of the bounce.

How luck works

A QUICK SITUATION CHECK

How likely
is the sack?

Pick a situation to see what research on head-coach exits found. The bar shows how likely a dismissal is, according to the studies below.

Situation check

Choose a situationTeam doing better than expectedAn experienced coach with past successGood play, bad resultsA run of poor recent results

Team doing better than expectedLess likely

Dismissal probabilities were significantly lower when a team performed above expectations, with the effect strongest for recent games. It also made coaches less likely to quit.

An experienced coach with past successSome protection

Past success and experience reduced the chance of dismissal even after allowing for current team performance, suggesting a protective effect.

Good play, bad resultsOften sacked

Quality of play measured by expected goals had little or no effect on dismissal. Clubs reacted to results alone, which the research calls outcome bias.

A run of poor recent resultsMost likely

Dismissals overwhelmingly follow disappointing results, especially recent ones. Yet teams that had underperformed tended to improve afterwards whether or not the manager went.

Sources: Bryson et al., De Economist 169 (2021); Erasmus University Rotterdam, Irrational managerial changes remain the norm; van Ours and van Tuijl, Economic Inquiry 54 (2016).

Based on Bryson, Buraimo, Farnell and Simmons (2021) and Erasmus University research by Jan van Ours. General findings across many clubs, not predictions for any one manager.

TWO NUMBERS WORTH KNOWING

Short spells,
and most end in a sacking.

From 1,407 coaching spells in the top two tiers
of Germany, France, Spain and Italy, 2002 to 2015.

944 of 1,407

Coaching spells that ended in dismissal

Another 358 spells ended with the coach quitting, and in 105 the coach was still in post at the end of the data.

DISMISSALS · BRYSON ET AL.

453 days

Average coaching spell before an exit

Spells ending in dismissal lasted 380 days on average, against 645 days for spells that ended with the coach quitting.

TENURE · BRYSON ET AL.

The data covered 642 head coaches of 206 teams, from the 2002–03 to 2014–15 seasons.

FOUR WAYS TO JUDGE A MANAGER

Look past
the last result.

The research suggests a few habits for fans, pundits and boards alike. They apply just as well to judging decisions outside football.

Compare with expectations
Studies use bookmaker odds to ask whether results were better or worse than a team's quality predicted.
Watch the underlying play
Measures like expected goals show whether chances were being created, whatever the scoreline.
Expect a bad run to ease
Some recovery is likely anyway, so a bounce after a change is not proof the change worked.
Check the counterfactual
Ask how similar teams did when they kept their manager through a similar run.

Two dice mid-roll above a green baize table, caught in soft light — AI-generated image, not a photograph of a real person or place.

A football resting on the white penalty spot of a green pitch under floodlights — AI-generated image, not a photograph of a real person or place.

HOW LUCK WORKS

Why the bounce
fools us.

Match results mix skill with chance. A team that loses several games it could have won is likely to see some of that luck even out. Several studies of the Dutch top league found a sequence of bad performances is often followed by a recovery, irrespective of a change of coach.

Van Ours suggests boards also face pressure from fans and media who focus on results. Sacking a manager can look like the safe choice, but keeping faith after a run of bad luck is frequently the more rational one.

Questions

Questions people ask

Does sacking the manager improve results?

Studies of the Dutch league and Europe's big five leagues suggest not much. Teams often do better afterwards, but similar teams that kept their manager tended to improve as well.

What is the new-manager bounce?

The rise in results often seen after a mid-season change. Research attributes much of it to regression to the mean: a bad run partly caused by chance tends to end on its own.

What is outcome bias?

Judging a decision or a person by results alone rather than by the quality of what led to them. Erasmus research found clubs sacked managers after bad results even when underlying play was good.

How long do head coaches last?

In a study of 1,407 coaching spells in the top two tiers of four European leagues, the average spell lasted 453 days, and those ending in dismissal 380 days.

Do clubs that keep their manager do better?

In the Erasmus study of the big five leagues, clubs that kept their managers typically ended the season with positive cumulative results, while those that sacked theirs stayed disappointing.

A bad run, a sacking,
then the bounce.