Humatrix Research Insight – Coach Dismissals and the Illusion of a New Beginning

When a Bundesliga club is in a sporting crisis, the reaction in Germany is almost reflexively the same: the coach is dismissed. The press conference speaks of a necessary impulse, of fresh energy, of a new voice for the team. The market applauds. The fans breathe a sigh of relief.

And indeed, many clubs observe an upswing in the weeks that follow.

The central question, however, long remained unanswered: Is this improvement really due to the new coach? Or does it follow another, less visible pattern?

Humatrix examined this question in an independent research project — based on ten Bundesliga seasons, 3,060 matches, and 80 manually verified coaching changes. The methodology is based on the study published by Heuer and colleagues in PLOS ONE in 2011, but was transferred to the modern Bundesliga and expanded with a comprehensive robustness check.

The results are revealing — but not in the way many might expect.

The average effect: small, but present

The Heuer methodology compares clubs that dismissed their coach during the season not with themselves, but with comparable clubs in exactly the same sporting situation that kept their coach. Only in this way is it possible to separate what can actually be attributed to the coaching change from what would have happened even without a change.

The results were calculated in parallel using three window sizes — 7, 10, and 13 matches before and after the dismissal. This robustness check ensures that the conclusion does not depend on a single methodological choice.

The pure coaching effect — that is, the points difference between clubs that dismissed their coach and the control group after the change — is consistently positive across all three windows:

• Window of 7 matches: +0.25 PPG (N=55, statistically significant)• Window of 10 matches: +0.18 PPG (N=36, not significant)• Window of 13 matches: +0.32 PPG (N=22, statistically significant)

Projected over half a season of 17 matches, this corresponds to an expected additional gain of around 3 to 5 points. That is a real, but small effect. In an average Bundesliga relegation battle, 3 points can mean the difference between staying up and entering the relegation playoff — in a calm mid-table season, they are barely noticeable.

What is remarkable is that the effect in the modern Bundesliga from 2014 to 2024 tends to be stronger than in Heuer’s original study, which covered 46 seasons from 1963 to 2008. One possible explanation: today’s clubs have more professional structures, faster succession processes, and more resources to support a coaching change organizationally. But the actual finding does not lie in the average.

The real finding: the dispersion

While the average coaching effect is small and manageable, the distribution of individual cases shows a completely different picture. The standard deviation of the coaching effect across all 36 methodologically evaluable cases is 0.59 PPG — around four times the mean.

Individual coaching changes produce up to +1.2 points per match. Others cost up to −0.8 points per match. The range between a successful and an unsuccessful change is around 2 points per match — eight times the expected average effect.

Specifically, the most successful coaching changes of the past ten years — Nagelsmann at Hoffenheim in 2015/16, Stöger at Dortmund in 2017/18, Tuchel at Bayern in 2022/23 — were real turnarounds with enormous sporting impact. Other changes accelerated relegation or led to even deeper crises. Both types of changes occurred under structurally similar starting conditions. The visible difference does not lie in whether a dismissal took place — but in how the decision was made and implemented.

Put differently: In day-to-day Bundesliga operations, a coaching change is not the manageable corrective measure it is often portrayed to be. It is the riskiest single decision a club makes during a season.

Regression to the mean — why part of the upswing is not due to the coach

A coach is usually dismissed after his team has gone through a phase that was significantly worse than expected. From a statistical perspective, this is by definition an outlier in the performance distribution. After an outlier — regardless of any intervention — there tends to be a movement back toward the long-term average. This phenomenon has been known as regression to the mean since the 1880s and is fundamentally measurable in all performance data.

The control-group calculation makes this visible: clubs in a comparable crisis situation that did not dismiss their coach recovered by an average of +0.17 points per match. Clubs that did dismiss their coach recovered by +0.35 points per match. The difference of +0.18 points is the part attributable to the coaching change. The larger share of the observed upswing — around half — would have occurred even without a change.

This explains why the general narrative of the “new impulse” is so persistent. It looks like an effect — although a substantial part of it was purely statistically expected.

The economic dimension — why the decision is not only sporting

So far, coaching changes have been considered exclusively through the lens of sporting performance. In the reality of a Bundesliga club, however, they are first and foremost an economic decision — and one of a magnitude that is regularly underestimated in public debate.

Coaching contracts in the Bundesliga are generally fixed-term agreements and do not include the possibility of ordinary termination. This is a labor-law peculiarity with significant financial consequences for clubs: if a club parts ways with a coach before the end of the contract, the obligation to pay the full salary until the originally agreed contract end generally remains in place. The industry rule of thumb: the severance payment roughly corresponds to the remaining contract duration multiplied by the annual salary.

The sums involved are considerable. Julian Nagelsmann’s dismissal at FC Bayern in March 2023 would have cost the club around €30 million in remaining salary if the contract had run its full course until 2026 — €8 million annual salary multiplied by three years. In addition, Bayern had originally signed the coach from RB Leipzig for a €25 million transfer fee. Even though in Nagelsmann’s case the later move to the German Football Association reduced the severance payment to €1 million, the initial calculation illustrates the economic risk: in top-level cases, the decision to change coach can represent a substantial financial hit in the tens of millions.

For an average Bundesliga coach with an annual salary of around €1.5 million to €3 million and a typical remaining contract duration of 1.5 years, direct severance costs usually fall between €2 million and €5 million — and that is only the legally visible tip of the iceberg.

Additional economic factors also come into play, rarely appearing in discussions about coaching changes but affecting every club directly:

• Double salary payments during the transition phase: the dismissed coach continues to be paid until the end of the remaining contract term, while the new coach receives his full salary from the start of his tenure.• A transfer fee for the new coach if he is under contract with another club. For sought-after profiles, these sums now regularly reach double-digit millions.• The takeover or new appointment of the entire coaching staff — assistant coaches, fitness coaches, analysts, goalkeeping coaches. As a rule, every new head coach brings three to six trusted staff members, which in turn means severance payments for replaced employees and new contracts.• Costs of adapting the playing philosophy: different training materials, different analysis tools, and often also players who do not fit the new system and may need to be replaced in the medium term.• Indirect costs caused by increased personnel turnover afterward: players who see no future under the new coach push for transfers. Transfer decisions are made under pressure.

In total, the financial cost of a coaching change in the Bundesliga typically ranges between €5 million and €15 million even in mid-level scenarios. According to the present analysis, this is offset by an expected sporting gain of around 3 to 5 points during the current season.

Anyone who makes this calculation soberly arrives at a sobering insight: on average, a coaching change is an economically expensive measure with a limited expected sporting return — and at the same time, the dispersion of individual cases ranges from securing survival to accelerating the crisis. Whether this investment pays off therefore depends not primarily on the points, but on the probability of landing the change in the positive part of the distribution.

What explains the difference between success and failure

If the dispersion of individual cases is eight times larger than the mean, the decisive lever does not lie in the question of whether a dismissal takes place — but in the conditions under which the decision is made and how it is implemented.

In the qualitative analysis of successful and failed changes over the past ten years, recurring structural patterns emerge. Changes with a positive trajectory generally occur at clubs where:

• The roles between sporting director, supervisory board, and coaching staff were clarified before the crisis and did not have to be renegotiated under acute pressure.• The personality and leadership fit between the new coach and the existing squad structure was systematically taken into account — not only tactics and availability.• The team leadership group — captain, deputies, informal leaders — was compatible with the new coaching style and able to act as a translator between the old and new leadership culture.• External communication was controlled and aligned with the internal reality, so that the new coach did not also have to manage a wave of media expectations.

Failed changes show the mirror image: blurred responsibilities, successors whose profile does not fit the team, dressing rooms without a stable leadership function, and media communication that immediately disappoints the expectations it has created.

What stands out is this: in none of these areas does the actual problem lie on the tactical or sporting level. It lies in the organization, in personalities, and in the clarity of structures.

The coaching question is usually asked too late

The central challenge in practice is temporal in nature. The question of whether the leadership structure around the coach is truly stable becomes acute at most clubs only once the crisis has already arrived. At that point, every analysis is distorted by emotional and media dynamics. The available information is incomplete, the decision-making horizon is compressed into just a few days, and the options are limited by the current market situation and available successors.

The result is predictable: the decision is made under stress and time pressure. It is communicatively motivated, not strategically grounded. And, as the dispersion in the data shows, it can just as easily succeed as fail.

Anyone who wants to make a truly informed coaching change would need to ask the structural questions months before the crisis. Which personality profiles currently lead the team? What leadership dynamics exist within the coaching staff? Is the division of roles in sporting management really as clear as the organizational chart suggests? Which successor profile would fit the existing squad structure — and which would not?

Humatrix at the intersection of structure and decision

Humatrix works precisely at this intersection. The method analyzes personality profiles in the squad and coaching staff, role and decision-making structures in sporting management, and the fit between leadership style and team reality. Evidence-based, systemic, and situated within the specific context of the club.

This does not make the coaching question obsolete — it makes it informed. Those who know which roles in their sporting management are unclear, which personalities lead in the dressing room, and what type of coach fits this constellation make decisions on a different basis. Not with a guarantee of success — but with a significantly lower risk of falling into the category of catastrophic changes.

The central insight of this study can be summarized in one sentence: the average coaching change brings little sporting benefit and costs a great deal economically — but no coaching change is average. Every individual case lies far above or below this mean. Whether a club ends up with the good or the bad version is a question of structural quality — not chance.

The decisive question is no longer: Do we make a change?

But rather: Do we understand the structure into which the new coach is entering?