The New Coach Recruited None of Them. We Start Counting Anyway.

A new head coach holds up the jersey at the press conference. He talks about culture, discipline, development, a new direction. Then something strange happens: before he has recruited a single freshman or coached a single practice, we start keeping score.

So ask the question plainly. Whose team is he actually coaching?

That sounds like a philosophical question. It is really a question about counting, and you can answer it. A college roster is roughly the last four recruiting classes. In a coach's first season, every one of those four was signed by the man he replaced. None of it is his. Not some of it — none.

By his third season about half the roster is his. By his fifth, all of it. That curve is the whole argument, and notice what it does not do: it never jumps. There is no moment when the team stops being inherited and starts being his. It bleeds over, one signing class at a time.

The Coach Is New. The Observations Are Not.

Here is the trap. The coach changes overnight; the team does not. The returning players were recruited by someone else and developed in someone else's weight room. The schedule was set. Some assistants stayed. The budget, the facilities, the conference — none of that moved because a man walked through a door.

Yet the record gets his name on it immediately.

And there is not one clock here, there are four, running at different speeds. He can change the lineup on Monday — that clock starts at once. Changing an athlete takes months. Recruiting takes years. Changing what a program is — the relationships, the expectations, who returns a phone call — may take longer than he gets.

Year One Will Fool You

A team goes 4–8. New coach arrives. Next year it goes 8–4. He doubled the wins, obviously.

Maybe. Or the previous staff's good class finally matured. Or the quarterback came back healthy. Or the three one-score games that went the wrong way went the right way. Or the schedule softened. Or — and this is the one almost nobody accounts for — 4–8 was simply an unusually bad year, and unusual years are followed by ordinary ones.

A school fires its coach at the moment performance looks worst. That is close to a definition of the moment improvement was coming anyway. You cannot tell those two stories apart from a before-and-after comparison, because both predict exactly the same thing.

Somebody did the work properly. Adler, Berry and Doherty matched programs that replaced their coach against comparable programs that kept theirs, across 1997 to 2010. For the genuinely bad teams, replacing the coach bought little improvement over the ones who stood pat. And for teams with middling records — the ones where a new man's conditions look most favorable — the ones who fired their coach did worse in the following years than the ones who did not.

Read that twice. The fire-him-and-move-on move, in the exact case where it feels most justified, was associated with getting worse.

The Treatment Changes the Subject

Now the modern wrinkle, and it is a strange one.

Under current NCAA rules, a player whose head coach leaves gets an additional 30-day window to enter the transfer portal, opening the day after the departure. The Division I Council approved that framework on 9 October 2024.

So firing the coach does not hold the roster still while you swap the man at the top. It opens a door in the roster. The new coach is then judged, in part, on a disruption caused by the event that hired him.

In an experiment you would call that contamination. The treatment changed the subject it was being tested on. Statisticians dislike that sort of thing for good reason: once it happens, you can no longer say what you were measuring.

It cuts both ways, which is what makes it interesting rather than merely unfair. The portal lets a coach rebuild faster than any coach could in 2005, so he arguably earns attributioni sooner than this chart suggests. It also means he inherits more volatility. Faster credit, faster blame.

Ask a Better Question

“Give him three years” is not a statistical claim. It is a number somebody picked because it sounds patient.

The better question is not how long he deserves. It is what evidence should exist by now. After ten games you can judge whether his fourth-down decisions make sense. After one season you can judge organization and whether anybody got visibly better. After two signing cycles you can start judging talent acquisition. After four years the roster is his and the excuses are gone.

The evidence accumulates gradually. Your confidence should too.

What to take home

This is not really about football, and you will use it within the week. New leaders inherit old staff. New principals inherit teachers and students. New doctors inherit patients whose treatment started somewhere else. Organizations have memory, and so do datasets.

So when something improves or collapses after a change at the top, do not start with “did the new person cause this?” Start with the question that has to come first: how much opportunity have they actually had to cause anything?

In sport the inherited team does eventually become the coach's team. The interesting part — the part worth counting — is working out when.