The Sports PageHe Inherited the Roster. He Wins One Game Fewer Than It Should.
When does a team stop being the last guy's team? Everybody has an opinion and nobody has a number. So we built one.
A college roster is roughly the last four recruiting classes. A class belongs to the new coach only if he signed it. That gives a share with no judgment in it. In his first season none of the roster is his. In his second, a quarter. Then half, then three quarters, and by his fifth season all of it.
Then the question becomes answerable. Do coaches beat the talent they were handed, and does the answer change as the talent becomes theirs?
The answer, across 2,685 seasons
First, the exchange rate. One standard deviation of recruiting — roughly the gap between the twentieth-best class in the country and the fiftieth — is worth 0.086 of winning percentage. About one win a season. That is what talent buys.
Now watch what coaches do against it.
A first-year coach finishes 8.3 points of winning percentage below what his inherited roster should have produced. Over twelve games that is a full win. He is not just failing to improve the team; he is getting less out of it than the talent on hand says he should.
Then it climbs, steadily, and it stops being negative at almost exactly the point where half the roster is his. By the fifth season he is 4.5 points above par. The swing from first season to fifth is about a win and a half.
The first year is not a small sample of the coach. It is a clean measurement of somebody else's players learning a new system.
— The Sports PageThe obvious objection, and why it does not hold
You might think this is survivorship; bad coaches get fired before their fourth season, so of course the men still standing in year five look good. The curve could be selection and nothing else.
So we tested it inside careers. Take only the 265 coaching spells that actually reached a fifth season, and look back at what those same men did earlier:
| season | all coaches | only those who lasted five years |
|---|---|---|
| first | −0.083 | −0.063 |
| second | −0.023 | +0.001 |
| third | −0.004 | +0.024 |
| fourth | +0.031 | +0.058 |
The men who survived five years were also underwater in year one. The climb happens inside their own careers, so survival is not creating it. Even the coaches who turned out to be good looked bad with somebody else's players.
So what does a first season tell you?
Less than the shouting suggests, and something different from what people think they are watching. A first-year record is mostly a measurement of the roster, run through a staff that has not had a signing day yet. The number to compare against is not .500. It is whatever that roster should have won, minus about a win.
So the honest version of "give him three years" is not patience; it is arithmetic. Three years is roughly when the roster crosses from mostly his predecessor's to mostly his. That is the first season whose record describes the man you hired.
What to take home
Before you judge anybody's first year at anything, ask what they inherited and how much of it they chose. Not as generosity; as measurement. A new manager, a new principal, a new head of a department — the outcomes you see in year one were mostly set in motion before they arrived.
The useful question is never "did it get better." It is "did it get better than what they were handed should have produced". Then, separately: how much of that did they pick?
Where this is generous to coaches, and it is generous in three places. Recruiting rankings are a market forecast, not a measurement, and they price the same hype a big-name hire generates. They ignore transfers completely, which in the portal era is a large and growing hole. And attrition is not random: struggling teams lose their best players. Every one of those pushes toward crediting a coach for talent he did not develop, so read the whole curve as flattering.
And one result we are not publishing as a finding. Beating your roster persists better on an inherited one (r = 0.45) than on your own (0.31), which is backwards from what you would expect. The likely explanation is uncomfortable. On an inherited roster, part of what we measure is a stable gap between a school’s recruiting rankings and its real quality. That is a property of the school, not the coach. We are working on it.
The script is public. More at thesportspage.net.
Argued about something you could not settle? Send it to us · what makes a good question
The Almanac · A Milestone Observance
Every fiftieth issue of this newsletter is, by our own coinage, an “edition” in the same fake-Latin manner that the federal government has decreed 2026 the semiquincentennial of the Republic. This is The Bicentennial Edition. In observance, the Almanac looks back at what was actually happening at the five prior fifty-year marks of our national life.
- 1776(250 yrs ago) — The Continental Congress sat in Philadelphia, debating whether thirteen colonies could become anything at all. The vote that mattered came on July 2.
- 1826(200 yrs ago) — John Adams and Thomas Jefferson, the last surviving signers of the Declaration, both died on July 4 — fifty years to the day after independence. Adams's last words were reported to be “Jefferson still survives.” Jefferson had died five hours earlier.
- 1876(150 yrs ago) — The presidential election that fall was decided not at the ballot box but by a fifteen-member electoral commission. Samuel Tilden (D) won the popular vote. Rutherford B. Hayes (R) was awarded the White House in exchange for ending Reconstruction.
- 1926(100 yrs ago) — Sinclair Lewis published Elmer Gantry the following year, but the manuscript — an attack on American religious hypocrisy — was already circulating in 1926.
- 1976(50 yrs ago) — Cincinnati's Big Red Machine won its second straight World Series, sweeping the Yankees in four. The roster included Bench, Rose, Morgan, and Perez.
