The number
Every February, recruiting services grade the high-school class each college program just signed and boil it down to a single score per team. That score then follows the program around for years. It builds the preseason top 25, and it's the argument somebody reaches for whenever a team ought to be better than its record.
We measured how long it stays true.
Through the first four weeks of a season, the recruiting gap between two teams explains 32% of the final margin. From week 5 on, it explains 5%.
That's 2,501 completed games between top-division teams, from 2023 through week 3 of this season — every one where we hold a recruiting score for both sides. The gap accounts for 32.4% of the scoreboard in weeks 1 through 4 (811 games), 5.4% in weeks 5 through 9 (820 games), and 7.0% from week 10 on (870 games).
The usual objection is that conference play matches similar rosters against each other, so a narrower spread of talent gaps would weaken the connection on its own without anything really changing. That's fair, and it isn't the whole story, because the size of the effect moves too. A 100-point edge in recruiting score is worth 5.33 points of margin in September and 2.81 in October. The relationship doesn't just get harder to see. It gets smaller.
Ask the November question instead
The margin of one game is the wrong thing to ask about anyway. The question that actually comes up on a broadcast in November is about the rest of the season: this team out-recruited that one, so back them the rest of the way.
So we measured that directly. After each week, take everything a team has done on the field so far, take its recruiting score separately, and ask which one better predicts the share of its remaining games it goes on to win.
Recruiting never wins. Not once, at any week of the season. After week 7, recruiting explains 1.7% of a team's remaining win rate across 414 seasons' worth of teams; the record it has already put on the field explains 20.2%. After week 10 it's 0.9% against 16.6%, across 405. There is twelve to eighteen times more signal in the games than in the stars, and that ratio is roughly stable from the moment a team has three results to its name.
0.9% is the number worth carrying into a playoff argument
By week 10 the selection committee is meeting and the rankings talk starts. This is where the recruiting argument gets made most often, and where it holds up worst.
After week 10, a team's recruiting score accounts for 0.9% of the difference in how teams finish. Not zero — we're not claiming zero, and 405 teams over three and a half seasons can't support a claim that clean. But 0.9% is indistinguishable from noise in any way that matters. If two teams are being compared in November and the tiebreaker on offer is that one of them signed a better class, that tiebreaker is carrying less than one percent of the available information while the other ninety-nine sit right there in the results.
The honest version of the recruiting argument is a September argument. It's a real one then. Thirty-two percent of a game's margin is a lot of explanatory power for one number. It has simply expired by the time anyone reaches for it.
Why it decays: the games answer the same question, better
Recruiting is a projection of how good a roster will be. Results are a measurement of how good it turned out to be. Early in the year the projection is all you have, so it does real work. Every week after that, the measurement gets sharper and the projection gets more redundant.
That isn't a theory we happen to like. It's what happened when we tried to put recruiting into our own model and couldn't. We carry a recruiting score for all 643 team-seasons in our college database and deliberately keep it out of the ratings, because every weight we tried made the projections worse, and worse in order — the more say we gave it, the further off we got. Testing across 2024 and 2025 week by week, only ever letting the model see games that had already happened, our ratings tracked actual margins best with no recruiting input at all. Adding a little hurt. Adding a moderate amount hurt more. Giving recruiting real say was substantially worse than ignoring it.
What makes that worth spelling out is that the simpler check which started the whole exercise looked great. On its own, recruiting did add a little predictive power to game margins in weeks 1 through 4 — exactly the window where the measurement above says it should.
Two things flattered that check. The first is that the simple test hands a new ingredient credit without making it compete; our model already knew most of what recruiting was telling it, and nothing in that test asked recruiting to prove it knew anything more. The second is that adding recruiting quietly changed which games were in the sample, because a score only exists for teams the service rates. Every game against a lower-division opponent dropped out — and those are the easiest blowouts on the schedule, the ones a talent gap predicts best. The edge didn't survive being made to compete for its place.
So what do you do with it in November
Use the classes to set expectations in August, then drop them the moment real games start arriving. Three results is enough for a team's own record to carry more information about its next month than its recruiting ever did.
In practice that means the November arguments worth having are about what a team has actually done. Who it beat, by how much, on the road, against whom. That's how our college ratings are built too: fit on results, blended across two seasons, with the recruiting score sitting on the team pages as context and nowhere near the projection. A team that signed the best class in the country and then went 5-4 is a 5-4 team, and we price it as one.
The exception is genuinely new information. A roster reshaped through the transfer portal, a quarterback change, an injury that changes what an offense can do — those move the measurement, because they change what the games are telling you. A signing-day ranking from twenty months ago doesn't.
What this doesn't say
It doesn't say recruiting is worthless. Programs that sign top-ten classes year after year are the programs whose in-season results are good, which is precisely why the results carry the signal — recruiting sits upstream of the thing we're measuring, not off to the side of it. Whether recruiting causes winning and whether it tells you anything you don't already know are different questions, and this only answers the second.
It doesn't say a week-1 preview should ignore the classes. In September it's the best single number available, and our own ratings lean on last season's results the same way, for the same reason.
And it's three and a half seasons: 2023 through week 3 of 2026, 2,501 rated games between top-division teams, one recruiting provider. The week-12 row of our season-ahead test has 34 teams in it and we've ignored it here, because a sample that size will tell you whatever you want to hear. The transfer portal is also actively rewriting what a recruiting class even means, and if that relationship moves, this measurement moves with it. We re-run it every year and publish whatever it says.
Sources
Every number here comes from the graded ledger the boards publish. Research from a statistical model, not betting advice; no outcome is guaranteed. Method: how MatchWiz works.