Why this is on a football rankings site

Because it is the same problem, pointed at different kids.

This whole index exists on one idea: a recruiting service only carries the players somebody already noticed, and a roster carries everybody. Half the point of building it was that a 1A kid in Kanab putting up numbers deserves to be findable next to a 6A commit.

Girls playing flag football in Utah are further down that same road. They are not under-covered — they are not counted at all. There is no state championship to win, no letter to earn, no official record that any of it happened. You cannot be overlooked by a system you were never entered into.

So this tab stays up until Utah sanctions it. Then it comes down, and the rankings start.

The petition

Petition to Sanction Girls’ High School Varsity Flag Football in Utah
Started by Stacie Carver on August 21, 2026. It had 93 verified signatures as of November 5, 2026, against a goal of 100 (change.org).

It is not our petition and we get nothing from it. It is here because it is the right thing to put in front of people who care about Utah high school football, and because the argument for it is one this site already makes every day about somebody else.

Add your name →

Coaching a girls’ flag program, or playing in one? Send it in. When Utah sanctions this, the rosters and the results belong on this site the same as everybody else’s — and having the history already collected is a good reason to start now rather than at kickoff.

Four kinds of evidence, and they are not equal

Every fact on a player page falls into one of four buckets. The site never blurs them together, because the difference between "a state tournament recorded this" and "a sixteen-year-old typed this" is the entire difference between useful and useless.

What this site will never publish

These are minors. The line is drawn well short of what is technically findable.

• No player phone numbers, email addresses, or home addresses. Not ever, for any reason, including when a player has posted them himself.
• No personal contact information for anyone. Coach contacts are the program’s official published address only — never a personal cell, never a private email, and never an address assembled from a district’s naming pattern.
• No transcripts, grade portals, eligibility records, discipline records or anything from a school information system. A GPA appears only when the player has already made it public himself, and it is labeled as unverified where it appears.
• No X handle or film link that was guessed. A handle goes on a profile only when it came from the player’s own profile. Attaching a stranger’s account to a sixteen-year-old is a mistake you cannot take back.
• No photographs taken from any of the sources above.
• Nothing negative about a player. Teams and coaching decisions get analyzed. Kids get evaluated, never attacked. If a number is unflattering, the page explains what it does and does not mean rather than leaving a teenager holding it.

Corrections and removals

If something is wrong

Say so and it gets fixed. Errors on a page like this are usually name matching — the wrong school, a transfer that did not happen, two players with the same name merged into one. Those are exactly the errors worth reporting because they are invisible from the inside. Corrections are made same day and the fix is noted.

If a player or parent wants a profile down

It comes down. No argument, no form, no explanation required, and no questions about why. A player or his parent or guardian can ask for the whole profile removed or for any single field — GPA, testing numbers, film link, handle — taken off and the rest left up.

If a coach wants to add something

Coach- and player-supplied information is welcome and it is labeled as such: a charted kicking percentage, a camp result, a standout game, an offer nobody published, a transfer, a GPA a player wants shown. Utah publishes no per-game player box scores, no sacks allowed, no kicking attempts and no punting statistics at all — so for those, what a program sends is the only data that will ever exist.

Contact

Use the Send Info form →

One door for all of it — corrections, removals, a stat we missed, a camp result, an offer nobody published, or a coach who wants his program’s numbers fixed. It reaches the same person an email would, and it asks for the few things needed to act on it, so nothing has to go back and forth twice. A real person reads every one.

What this page is honest about

A recruiting site that never admits a limit is telling you something.

Coverage is uneven, and that is a reporting gap, not a talent gap

Statistics exist only where a program entered them. Two schools have entered nothing this season, so no player from either can appear on any leaderboard. That is not a statement about those players. Every list on this site says so where it applies.

Some things Utah simply does not publish

There are no per-game player box scores anywhere in the state. There is no sacks-allowed statistic, which is why the O-line rating infers pressure from quarterback negative rushing and says exactly how confident that read is for each line. There are no kicking attempts, so the specialist boards are makes, not accuracy. There is no punting data at all.

The ratings are young

Early in a season a team rating is still substantially a preseason projection, and the page shows how much. A production grade built on two games is a two-game sample and is shrunk toward the field accordingly.

An unranked player is not a lesser player

This is the reason the site exists. Recruiting services carry players somebody already noticed; a roster carries everybody. Several of the highest-graded players here have no recruiting ranking at all — they are graded on production, honors and years played. If a service has not rated somebody, this page treats that as information about who has been watching, not about how he plays.

How the number is built

One rating, one scale, all six classifications, no ballots and no reputation. A team's rating is roughly "points better or worse than an average Utah 11-man team on a neutral field." Two teams 20 points apart in rating project to a 20-point game between them.

The core model

It's a weighted ridge regression on scoring margin. Every game contributes one equation — home rating − away rating + home field = margin — and the solver finds the set of ratings that best fits all the games played so far with the least total error. That's what makes it opponent-adjusted: beating a good team moves you more than beating a bad one, automatically, without anyone assigning "quality win" points by hand.

Common opponents are already in it

You asked for common-opponent weighting. This is the part people usually hand-build, and the regression does it structurally: if you and another team both played Clearfield, your two results are tied to the same Clearfield rating, so the solver can only satisfy both equations by placing you correctly relative to each other. Every shared opponent is a rung on the ladder between two teams. The Common Opponents tab shows that math explicitly for any two teams so you can see the chain rather than trust it.

The 2026 rule changes

Playoffs are now capped at 16 teams per classification. That is unchanged for 6A, 3A, 2A and 1A, which have 17, 13, 10 and 9 programs — but 5A and 4A previously took 24 seeds with the top eight on byes. This year 12 of 28 in 5A and 13 of 29 in 4A go home. Seeding is by RPI. The Playoffs tab shows the cut line, with a caution about how little RPI means in August.

Two mercy rules now, not one. The long-standing rule keeps a running clock after halftime once a lead passes 34. New for 2026: "if a team leads by 44 points after the third quarter, or at any point in the fourth quarter, the game ends." Deseret's own analysis put that at roughly 19% of last season's games.

This changes what the data can even contain. Final margins are now censored near 44 — a team that would have won by 60 is recorded winning by 45, and several games this season already show an empty fourth quarter because they simply stopped. The blowout cap here tops out at 28, comfortably inside that ceiling, so the rating is untouched by the censoring. Projected margins are capped at 44 for the same reason: the rules do not permit a wider result.

Garbage time, and why the fourth quarter is capped

Every game's quarter-by-quarter box score is loaded — 118 of the 122 games played have one. That matters because a final score lies about decided games. A team that leads by 40 after three quarters, empties the bench, and gives up 21 in garbage time did not play a 19-point game.

So the rating uses an effective margin: if the lead through three quarters is 21 or more, the game counts as that three-quarter margin plus at most seven points of fourth quarter in either direction. Under 21, the final score stands as-is. This runs before the blowout cap below, so the two work together — one strips out the minutes that stopped being a real game, the other stops the rest from compounding.

Quarter data also gives every team card a scoring profile: points scored and allowed in each quarter, and the first-half/second-half split. Some of those splits are stark. Cyprus has been outscored 43–7 in first quarters across two games and then played the second halves within four points — a start problem, not a talent problem.

Blowout cap

Margin earns full credit to 17 points. From 17 to 28 it earns half credit. Past 28 it earns nothing. A win is worth an extra 1.5 points regardless of score. So 42–0 and 63–0 rate identically — running it up buys nothing, and getting run out of the building past four scores doesn't compound against you.

Preseason prior, and how it fades

Every team starts with a projected rating built from program factors. Each game played pulls the rating toward what's actually happened and loosens the prior's grip. Right now, at two games in, the split is about 50/50. By Week 6 the prior is worth roughly 12%. By the end of the regular season it's near noise.

Score and yards

The power rating runs on margin, but margin hides which side of the ball did the work. A second regression, run on points rather than margin, splits every team into an offense rating and a defense rating: points scored in a game equals league average plus the offense, minus the opponent's defense, plus half the home edge. Both columns are opponent-adjusted and read in points above or below an average Utah team — Corner Canyon at +9 on defense means an average offense scores about nine fewer points than usual against them.

Yardage is the other half. Utah publishes offensive yards (passing plus rushing, at the player level, summed up here) but nothing publishes yards allowed — no team defensive totals exist on Deseret, MaxPreps or UHSAA — so the defensive side stays scoring-only. Offensive yards per game get a schedule adjustment: a team that faced two good defenses gets yardage credited back at the league rate of 10 yards per point.

Yards then nudge the rating, capped at three points. Early in a season yardage is a steadier signal than scoring, so a team that moves the ball but stalls in the red zone gets pulled up, and a team scoring on short fields gets pulled down. The Finishing figure on each team card is the gap: points per game minus what their yardage implies. Herriman is at −15 — 453 yards a game and only 32 points — while Syracuse is scoring well above what its yardage says.

One caveat, and it is not small: 18 schools report yardage that cannot be reconciled with their own scores — American Fork's page shows 104 yards against 47 points. Those schools are excluded from the yardage adjustment entirely rather than corrected, and their Yds/g cell is blank. Stat entry at this level is done by whoever the program assigns to it.

Program stature

A handful of Utah programs have standing that runs longer than the three seasons the history term looks at. They win titles across a decade, and they schedule people most programs will not. That is worth something in a preseason projection, but only a little — it carries a weight of 1.2, below transfers, against 5.4 for recent results.

It is built from two things a program actually did, not from anyone's sense of who is a big name:

Who they are willing to play. Booking IMG Academy or Mater Dei is a decision, and only a program that believes it belongs makes it. Credit scales with how good that opponent actually is, so an out-of-state game against a weak team earns close to nothing — that is a scheduling convenience, not a statement. Orem is the clearest case: Kamehameha Kapalama and then Mater Dei, which is most of why they moved up five spots when this term was added.

State titles over a decade, at 0.9 each. The history term already counts these at 1.1, so this is a deliberate small top-up for programs whose reputation predates the current window. It is what lifts Beaver, San Juan and Morgan — small-school dynasties whose last three years undersell them.

What is deliberately not in here: any hand-entered list of which programs are “known.” That would be a thumb on the scale for the schools whose names happen to be familiar, and it would quietly punish a good program in a town nobody writes about. Two caveats worth stating. This term structurally favors large schools, because scheduling a national opponent takes a budget and a travel program; a 1A school that cannot do it is not worse for it. And the schedule data here currently runs only through Week 3, so a program whose out-of-state game comes in October has not been credited yet.

Across the state it moves the average team about one place, with the largest single move being five. That is the size it was meant to be.

Recruiting talent

Four services, blended onto one scale. 247Sports carries the most weight — it is the market marker and the industry prices off it. Rivals/On3 and ESPN sit just behind. Prep Redzone is weighted lowest on purpose: players can create their own profile there, so appearing on a 518-deep board is weak evidence by itself. A Prep Redzone top-100 is treated as real; the tail is not. That single choice is why a program can have a ranked player and still score near zero here.

Each service's numbers are mapped onto one 0–100 quality scale before anything is combined — ESPN's grades run lower than a 247 composite, and Prep Redzone publishes a rank rather than a grade, so a rank is converted through a log curve that is steep at the top and nearly flat past #200. A player's grade is the weighted average of whichever services rate him. 635 players are graded and 64 are rated by two or more services; those are the ones to trust, and they have their own panel on the Talent tab.

A team's score sums its players above a floor of 79, curved so that one 90 is worth more than ten 80s — which is how high school football actually works. Seniors count in full, juniors at 65% and sophomores at 35%: they help this year, just less. Orem, West and Timpview lead the state. Eleven programs have nobody ranked anywhere, and a zero there is a real zero rather than missing data.

It feeds the preseason prior at a weight just under returning production. Deliberately not higher: recruiting rankings and program history measure overlapping things, and stacking both at full weight would double-count the same blue-blood programs.

How a real line gets made, and where this differs

Vegas oddsmakers start exactly where this page does. Chris Andrews, a veteran Nevada bookmaker, describes rating every team on a numerical scale, taking the difference between two teams, and adjusting for home field: "Power ratings give you a place to start." He also notes that home field, long assumed to be three points, has fallen to "1.5 or 2 points" in the NFL and is calibrated per team. This page uses a flat 2.6 for every school because two games is nowhere near enough to calibrate 106 home fields separately — that is a known simplification, not an oversight.

Where the two part company is key numbers. NFL bookmakers hesitate to move a line through 3 or 7 because those are the most common margins. Utah high school football is a different sport in that respect. Across the 122 games played so far, only 30% finished within a touchdown and 48% were decided by 21 or more. Half the schedule is a blowout, so there is no dense cluster of margins to protect — which is also the reason the blowout cap matters more here than it would in the NFL.

Projected scores

The rating gives the margin. The total comes from what the two offenses actually produce: expected points equals league average plus that team's offense rating minus the opponent's defense rating, and then that model figure is blended toward what the team has really been scoring — with the observed number carrying two-thirds of the weight at two games in, and steadily less as the season gives the model more to work with.

This matters more than it sounds. Splitting a league-average total by the margin projected Cyprus for 23 points in a close game when they have scored seven all season. Anchoring to real production puts that game at 14–12, which is what two teams averaging 13.5 and 3.5 points actually look like. Corner Canyon still projects for 51, because they are genuinely scoring 45.5 a game. And no projection ends in a tie — football settles them in overtime.

Strength of schedule

SOS is the average rating of the opponents a team has actually faced. At two games it's a small sample and swings hard — a team that opened against Corner Canyon and IMG will show an inflated number. It becomes meaningful around Week 5. Note this is schedule faced to date, not full-season projected schedule; that arrives once full schedules are loaded.

UHSAA RPI

The state seeds and selects the playoffs on the RPI: 45% your win percentage, 45% your opponents' win percentage, 10% your opponents' opponents'. Score does not count. We show UHSAA's own published number (uhsaa.org/football-rpi), as last published Oct 13, 2026 - UHSAA took the tables down once the playoffs were seeded - in the RPI column and on every team card, so you can see the state's view next to ours.

It also feeds the Ledger Rating, lightly: rating + 6 × (RPI − .500) × 2. That is worth about −4 to +4 points against a rating spread of roughly 23, so it nudges the board without driving it. Added Oct 11, 2026; it moved 14 teams, none more than two spots in its class.

Recency

Each week back multiplies a game's weight by 0.88. Week 2 counts fully, Week 1 counts 88%, and by late October an August result is worth about half a November one. Who you are in Week 9 matters more than who you were in Week 1.

Out-of-state and 8-man opponents

Out-of-state teams and Utah 8-player schools appear as opponents. They're rated too — otherwise the Utah teams that played them would have a hole in their schedule — but their ratings are pinned to a hand-set estimate rather than solved, since one game can't identify a team. They don't appear in the rankings themselves.

Five states are now calibrated rather than guessed. Idaho, Nevada, Hawaii, California and Arizona all publish team ratings. Fitting the eleven games Utah has actually played against rated opponents returned a slope of 0.93 — near enough to 1 that these are clearly one national scale rather than five state scales, with the whole thing sitting about 55 points above this one. So every pin is now the same transparent subtraction: the published rating minus 55. That replaced a page of ad-hoc estimates with one rule.

The residuals are honest about the limits. They run large on single blowouts — Hurricane's two comfortable Nevada wins and Tafuna's 46–0 at East are each one game, and one game cannot separate "the scale is off" from "that team had a night." Tafuna is the clearest case: Hawaii's list rates them 23rd, which the offset would put near −10, but they won by 46 on the road. That pin sits between the two readings and is flagged as unresolved rather than pretending either number settled it.

Those pins are doing real work, so they're set from outside context rather than guessed. IMG Academy opened 2026 ranked No. 1 in the country by Sports Illustrated's Power 25 and No. 1 in Florida by MaxPreps, so it sits at +52 — which is why Lone Peak's 3–59 loss barely dents them. The model expected a beating from a national title contender and Lone Peak still rates third in Utah. Mater Dei is pinned at +47 on the same basis. Get those pins wrong and you punish good Utah teams for scheduling ambitiously, which is the opposite of what the schedule deserves. The Out of State tab shows every one of these games and how Utah has actually fared.

Returning production, not returning bodies

Counting how many players came back tells you very little — a senior who was on the roster for three years but never took a varsity snap is a jersey, not a player. So the model measures production: every player's 2025 passing, rushing, receiving and tackle numbers are scored, summed to a team total, and the share of that total still on the 2026 roster becomes the returning-production figure. A team with 68% back (Box Elder) genuinely returns its team. A team with 17% back (Corner Canyon, after graduating a state-title senior class) is starting over no matter how many names are on the roster.

Alongside it, the model uses the returning starters each program named in its own preseason region-preview questionnaire — the coach's own answer to who actually plays. Every team card lists that school's key returning producers by category with their real 2025 stat lines, plus who is leading the team through two games of 2026.

Transfers

Every school's 2025 and 2026 rosters are cross-indexed by name and graduation year. A player listed at one Utah school last year and a different one this year counts as a transfer. Names appearing at three or more schools are dropped as too common to attribute safely, so the count is a floor — spelling drift between years and out-of-state arrivals are invisible to it. Net transfers (in minus out) feed the preseason prior at a deliberately modest weight, because a transfer is a body, not automatically a starter.

What's not in here yet

• Full-season SOS. Strength of schedule is games played to date, not a remaining-schedule projection.
• Coordinator track records. Names are in for 101 of 106 schools and real backgrounds for 14 of them. Nobody publishes coordinator win-loss history systematically — what exists is scattered across college bios and old newspaper clips.
• Special teams coordinators. Not a single one of the 105 region-preview capsules lists one, and only four programs in the state publish it on their own site: Alta, Westlake, Timpanogos and Copper Hills. Shown where found, blank everywhere else.
• Injuries. No source publishes them at this level, and coaches would not report them honestly if one did.

Known data flags

A note on other rankings

MaxPreps runs the ranking most Utah coaches see. Its own documentation describes "a complex, unbiased computer algorithm focused on strength of schedule" using game results, playoff results and opponent rating — and stops there. It does not publish the formula, whether margin is capped, or how national and state rankings differ. That is their right, and their algorithm has more history behind it than this one. The difference here is that every number on this page can be traced: the weights are listed, the caps are stated, the data gaps are named, and the code is a page of arithmetic you could check by hand.

Sources

Scores, schedules and quarter-by-quarter box scores from Deseret News. Rosters, coach records and tenure from Deseret News school pages. Classification and region alignment cross-checked against UHSAA standings, Deseret News and MaxPreps. Prior-season records and playoff results from Deseret News archives and UHSAA brackets. Recruiting rankings from 247Sports (2027 and 2028 composite), Rivals/On3 (2027, 2028, 2029), ESPN (2027, 2028) and Prep Redzone (2027, 2028). National context for out-of-state opponents from Sports Illustrated's Power 25 and MaxPreps.