You have heard this argument at every tailgate. Team A is better on paper, but Team B has been hot, and nobody agrees on which one matters. Everybody gets loud. Nobody gets graded.
Today SPIN is changing that for ourselves. Meet STRANOVA, SPIN's computer rating system. It reads every game result, adjusts for who each team actually played, and turns the gap between two teams into a win chance for each side. Then we publish the number before kickoff, we freeze it, and we post the final next to it afterward. Hit or miss.
STRANOVA is the product of more than two years of work and thousands of hours to get this model right. We are not showing you a weekend project.
And that is not all. Starting next week, STRANOVA runs SPIN's team rankings, period. National, state and regional boards, every one of them, built by the model with no manual edits. Once a full week of games across the country is in, we run it, and the boards update. Weekly, on a schedule, the way the best computer rankings work. This is the debut, so let's do it the honest way: here is exactly what it says about Friday, and exactly how you can grade it.
What STRANOVA is, and what it is not
STRANOVA is a rating built from scores. It does not know who is hurt. It does not know who just changed quarterbacks. It does not watch film, and it does not care how a team looks in warmups. What it does well is the part humans do badly: it treats a 14-point win over a good team and a 14-point win over a bad team as different things, it does not reward anyone for running up the score on a weak opponent, and it does not let one bad night or one hot streak rewrite a whole season.
What it gives you is a win chance, not a pick. A 90% win chance does not mean a lock. It means that when STRANOVA says 90%, the other team wins about one time in ten. Over a 12-game slate, that adds up. More on that below.
The 12 quarterfinal calls
Every game below kicks off Friday, Nov. 6. The home team is the higher seed. Each card shows both teams' records, both win chances, and a short note on what could send the game the other way.
6A
| Game | Kickoff | Win chance |
|---|---|---|
| American Fork (7-4) at Skyridge (9-1) | 6 p.m. | Skyridge 81%, American Fork 19% |
| Davis (8-3) at Herriman (9-2) | 4 p.m. | Herriman 57%, Davis 43% |
| Lone Peak (6-5) at Mountain Ridge (11-0) | 6 p.m. | Mountain Ridge 90%, Lone Peak 10% |
| Syracuse (6-4) at Corner Canyon (9-2) | 5 p.m. | Corner Canyon 99%, Syracuse 1% |
5A
| Game | Kickoff | Win chance |
|---|---|---|
| Fremont (7-4) at West (11-0) | 4 p.m. | West 96%, Fremont 4% |
| Springville (8-2) at Bountiful (8-2) | 5 p.m. | Springville 52%, Bountiful 48% |
| Pleasant Grove (7-4) at Orem (8-2) | 7 p.m. | Orem 94%, Pleasant Grove 6% |
| Woods Cross (5-6) at Alta (8-3) | 6 p.m. | Alta 90%, Woods Cross 10% |
4A
| Game | Kickoff | Win chance |
|---|---|---|
| Highland (8-2) at Ridgeline (11-0) | 6 p.m. | Ridgeline 93%, Highland 7% |
| Crimson Cliffs (8-2) at Snow Canyon (7-3) | 6 p.m. | Crimson Cliffs 83%, Snow Canyon 17% |
| Salem Hills (9-2) at Hurricane (10-1) | 6 p.m. | Hurricane 91%, Salem Hills 9% |
| Provo (7-4) at Skyline (8-3) | 5 p.m. | Skyline 76%, Provo 24% |
Two games stand out. Davis at Herriman is a 57 to 43 call, and on a neutral field STRANOVA calls it a coin flip. Springville at Bountiful is 52 to 48, and STRANOVA has Springville ahead on a neutral field. Home field is what narrows it. If either of those games goes your way, nobody should be surprised, including us.
Two other calls go against the host. STRANOVA likes Springville (at Bountiful) and Crimson Cliffs (at Snow Canyon). Crimson Cliffs already beat Snow Canyon 38-16 on Sept. 25.
The forecast we are committing to in advance
Add up the 12 win chances on the favorite side and you get 10.0. That is STRANOVA's own forecast for Friday: about 10 of 12 favorites win, and about two do not. Two misses is the plan, not a failure. Eight or nine would be a bad night for the model. Eleven or twelve would be a good one. We are saying it now so nobody can say we moved the goalposts on Saturday.
Here is the part we would rather tell you than have you tell us. STRANOVA favors the home team in 10 of the 12 games. So if it goes 10 for 12, the fair reply is that anyone who just backed the higher seed might have done about as well. That is why the win count is not the best test of this thing. The better test is calibration: when STRANOVA says 90% or higher, do those teams win about nine times in ten? When it says 55%, does it look like a coin flip? One slate cannot answer that. A full playoff run can start to.
How to grade STRANOVA
We are setting the scoring rules before any result exists, and we will not change them:
- Favorites that won: how many favorites won, running total each round.
- Calibration: how often teams at 90% or higher, 75 to 89, 60 to 74 and 50 to 59 actually win.
- Projected scores: the cards also show a projected score. In testing, those projections missed by about 9 points a team on average. Treat them as a rough picture of the game, not a prediction of the fourth quarter.
- Baselines beside it: "always go with the home team" and "always go with the higher seed," so you can see what the model adds.
- Brier score and log loss, for the people who will check the math.
- Sample size: 12 games say very little. Judge it at 100.
Every round we will post the pre-game number next to the final, with no defense and no edits. When STRANOVA misses, we will call it a miss. We will not say the model was wrong about a 57% call, because a 57% call that loses was still 43% right. If we find a data error after the freeze, we will publish a correction with the recomputed number and leave the frozen card exactly as it ran.
The running scoreboard has its own page on spin.media: https://www.spin.media/articles/the-stranova-scoreboard-every-call-graded-in-public/FB574781. It is live before the first post, and every card and every social post links back to it.
Why you can trust the freeze
The numbers on these cards were locked on Nov. 5 at 7:51 a.m. MT using model version 0.4. The record of those numbers has this fingerprint:
`d1903400887fd7317e688233db7c8b09b7128ebab131bd9a74334eb7ad5d01fe`
Anyone can check that the file we publish matches it. If a single number changed after kickoff, the fingerprint would not match.
We also ran STRANOVA back through this season before showing it to you. Using only the games played before each week, it called the winner in 191 of 232 Utah games between Utah teams, from late August through October, or 82.3%. That is the regular season, and it is not the playoffs. Quarterfinal matchups are closer than the average game, so expect that number to come down. We would rather tell you that than hide it.
What the model cannot see, and what to do with that
STRANOVA reads scores and schedules. It does not see injuries, lineup changes, weather, travel, or whatever happened in practice Tuesday. Each card carries a short note on something the model cannot see and one way the game could end up differently. Use the numbers as a starting point for the argument, not the end of it.
This is a rating of teams and their results. It does not rate or grade any player, and it is not a betting product. Win chances are for fans, coaches and families who want an honest read on a game. Not for wagering.
What comes next
Results go up Saturday morning, pre-game number beside the final. Semifinal cards follow once the quarterfinals are done. From next week, every SPIN team ranking, national down to each state and region, comes straight from STRANOVA and refreshes each week once the week's games are in. Nobody at SPIN reorders it by hand. We will show the results each time.
So what do you think? Which of the 12 calls will STRANOVA get wrong? Tell us where we are fooling ourselves, and check back Saturday to see how the first slate scored.
