Every number on this site comes from one place: a database of Alabama high school football results gathered from the Alabama High School Athletic Association (AHSAA) and its member schools’ published schedules, and a rating computed from those results.
The rating
A team’s rating is how many points better than an average Alabama team it has shown itself to be. It is solved from every result in the state at once: beating a good side by a little counts for more than beating a poor one by a lot, and a narrow loss to a strong team can still raise a rating.
Margins are capped at 28 points. Beyond that the scoreboard stops describing the difference between two teams and starts describing whether the starters stayed on the field.
A team is not ranked until it has played two games. Opponents are measured with games against the rated team removed, so when everyone has played once there is nothing left to measure them by. After week 1 that leaves only the handful of teams who played twice; the table fills out from week 2.
Classification enters once, as a starting assumption worth about one game of evidence, and fades as real results arrive. It matters most in the AISA and the small classes, where schedules rarely cross.
Strength of schedule
Schedule strength is the average rating of the opponents a team has played — not their win-loss records. A record cannot tell a 1-1 6A side from a 1-1 1A side. Tested against the 2025 season, rating opponents this way picked 83.4% of the following week’s winners, against 82.1% for the record-based version.
Projections
A projected margin is the gap between two ratings, adjusted for home field and scaled to match how results actually fall. It is a projected margin, not a spread: nothing here is a betting line, and no money should be anywhere near it.
Certainty is the favourite’s chance of winning. High school results scatter around a projection by about 14 points, so a game projected at 8.5 points is roughly a 73% proposition, and one projected at 3 points is close to a coin toss. Certainty stops short of 100%: the curve reaches it past about 60 points, but teams predicted to win by that much still lost 1.8% of the time. A number saying a football match is certain is wrong however good the model is.
The accuracy record
This is the part that matters. Projections are written to a permanent log before the games are played, and graded afterwards against what happened. A pick counts only if it was saved before 5pm on the day of the game. Nothing is re-run after a result is known, and no past projection is ever rewritten — the log is added to, never edited.
The record is also compared with the ASWA sports writers’ poll on the same games: where the poll ranks both teams, the better-ranked side is taken as its pick. That is the fair comparison, and it is published whether it flatters this model or not.
Where the data comes from
Schedules, scores and classifications come from the Alabama High School Athletic Association (AHSAA), the governing body for Alabama high school sports, and from the schedules and results its member schools publish.
What this model does not know
It sees scores and nothing else. It does not know who is injured, who has transferred, who is suspended, whether a starting quarterback is out, or that a team emptied its bench in the fourth quarter. It has no opinion on character, coaching or effort. When it disagrees with someone who watches these teams every week, that person may well be right.
Cookies and privacy
This site has no accounts and doesn’t ask for your name or email. cabalsports.com itself sets no cookies.
We use Google Analytics to count visits. It sets cookies in your browser (named _ga and ga followed by letters and numbers) and records standard information: which pages were viewed, roughly where in the world the visit came from, and the type of browser and device. We use it only to see how the site is used, not for advertising.
Page lettering is served by Google Fonts, so Google’s servers see your device’s address when a page loads.
You can block or delete cookies in your browser settings, or use Google’s opt-out add-on at tools.google.com/dlpage/gaoptout. The site works the same either way.