Skip to content

Conference Races

Analysis

Who wins each of the ten leagues that stage a championship game, from 50,000 simulated 2026 seasons — the same seasons behind the playoff page, so a team’s title odds and its playoff odds are two questions asked of one set of results rather than two models glued together.

A percentage is the share of those seasons ending that way, not a ranking and not a claim about any single game. Rebuilt 2026-09-11.

The leagues do not share a rule

  • Sun Belt keeps divisions (East and West) and sends the winner of each — so its second-best record stays home whenever it shares a division with the best one. Taking the top two league-wide instead would stage an impossible title game in 33% of simulated seasons.
  • ACC, American Athletic, Mountain West play an unbalanced conference schedule — members do not all play the same number of league games — so the standings are a winning percentage, not a win count. Ranking those by wins promotes whoever drew the extra game.
  • Everyone else sends its top two by conference winning percentage, head-to-head breaking a tie at the line where the berth is decided.

Every league

The favorite in each, and how often the last berth in the title game is level on conference percentage — which is to say, how often a tiebreaker decides who plays for the trophy rather than the standings deciding it outright.

LeagueFavoriteTitle oddsTiebreak
ACCMiami56%37%
Big TenOhio State36%50%
Big 12Texas Tech43%45%
SECGeorgia21%48%
American AthleticMemphis21%33%
Conference USALiberty37%42%
Mid-AmericanWestern Michigan40%46%
Mountain WestUNLV35%35%
Pac-12Boise State50%38%
Sun BeltJames Madison29%47%

Pac-12

8 teams, no divisions, 7 conference games. The top two by conference winning percentage play for the title.

Pac-12

top two play for the title8

Each crest carries the team’s chance of winning the league outright. Reaching the title game is a different number and it is in the table below.

Pac-12 standings, projected

Title game and title are different questions, and the gap between them is how often a team is expected to get there and lose. The projected conference record is the mean across all 50,000 seasons, so it carries fractions a real record cannot.

TeamConfTitle gameTitle
Boise State0-073%50%
San Diego State0-038%18%
Washington State0-023%9.1%
Oregon State0-018%6.3%
Fresno State0-014%5.2%
Texas State0-012%4.2%
Colorado State0-011%3.9%
Utah State0-010%3.6%

The title games it stages most often

Every pairing the simulation produced, most common first. A championship game is two teams surviving twelve weeks apiece, so even the likeliest is a small number — the shape of this list says more than its top row.

  • Boise State vs San Diego State23%
  • Boise State vs Washington State12%
  • Boise State vs Oregon State10%
  • Boise State vs Fresno State7.7%
  • Boise State vs Texas State7.0%
  • Boise State vs Colorado State6.0%

What Boise State needs

Boise State wins the Pac-12 in 50% of simulated seasons and reaches the title game in 73%. These are the remaining results that move that number most — including games they are not playing in, which later in the year are the ones nobody works out for themselves.

  • wk12San Diego State at Boise State34.9
  • wk11Oregon State at Boise State31.7
  • wk9Texas State at Boise State30.9
  • wk8Boise State at Washington State30.5
  • wk10Boise State at Colorado State29.5

Points of Boise State’s title probability riding on each result.

Since preseason

The Pac-12’s best title odds, week by week, on a shared 0–100% scale. Earlier weeks are replayed with the current model rather than recorded at the time, so they say what today’s simulation makes of that week.

Who moved

Change in title probability from preseason to week 1. A league is zero-sum — every point one team gains came off the rest of them.

  • Boise State34%51%+17.1
  • Colorado State1.9%3.8%+1.9
  • Washington State7.7%8.7%+1.0
  • Utah State3.8%3.2%-0.6
  • Oregon State6.9%6.1%-0.8
  • San Diego State20%18%-1.4
  • Texas State12%3.9%-8.5
  • Fresno State14%5.0%-8.7

What a conference loss costs

How often each team still wins the Pac-12 having lost exactly that many league games. Affords is the most it can take and stay better than even — blank for a team that is not better than even unbeaten.

Share of simulated seasons a team makes the field, given it finishes with exactly that many losses. ⚠ These are raw conditionals: a team that reaches 9–3 against a brutal schedule is a better team than one that reaches 9–3 against an easy one, and some of the spread below is that rather than the committee rewarding a résumé.
Team0L1L2L3L4L5L6L7LAffords
Boise State78663740001
San Diego State644721200000
Washington State534118100000
Oregon State513717100000
Fresno State533717100000
Texas State513716100000
Colorado State50371820000
Utah State543915100000

What decides it

The remaining league games ranked by how much the title race is reshaped depending on who wins — measured across every team’s odds at once, not just the favorite’s. Early in a season the front-runner’s own schedule dominates this list, which is not a quirk of the method: it is what a one-team league looks like in September.

  • wk12San Diego State at Boise State34.9
  • wk11Oregon State at Boise State31.7
  • wk9Texas State at Boise State30.9
  • wk8Boise State at Washington State30.5
  • wk10Boise State at Colorado State29.5
  • wk5Utah State at Boise State29.5
  • wk6Boise State at Fresno State28.7
  • wk6San Diego State at Oregon State19.6

What this is still blind to

  • Two teams level on conference percentage are separated head-to-head. A three-way tie falls back to point differential, where real leagues run a longer ladder.
  • The simulation is market-blind and its committee is fitted rather than consulted — the same four caveats the playoff page states in full.

How this is built →