The model turns power ratings into fair value in three steps: convert rating gaps into game-by-game win probabilities, simulate the entire season and playoff many times over, then count how often each team finishes as champion. That final share is the fair value. Texas tops the national title board at 12.9%, with Georgia at 12.2% and Ohio State and Notre Dame tied at 11.2%.
How does a power rating become a win probability?
A power rating is a single number describing team strength. The model does not trade on that number directly; it treats the gap between two ratings, adjusted for site and rest, as the input to a win-probability curve. A large edge maps toward the high end of the curve, a coin-flip gap toward 50%.
That curve is the hinge of the whole system. It never outputs certainty, because no single college football game is certain. Every matchup on a team's schedule gets its own probability, and those probabilities become the raw material for the season simulation.
From single games to a full-season simulation
One game probability says little about a title. The model chains them together by simulating the complete schedule, conference races, and playoff bracket repeatedly. In each run, favorites usually advance and underdogs sometimes spring through, exactly as the per-game odds imply.
After the full set of simulations, the model counts championships. A team that wins the title in 12.9% of runs, as Texas does, carries a 12.9% fair value. This is why the board is not simply a ranking of power ratings: schedule difficulty, conference path, and bracket variance all reshape the final number.
Where does the fair value board stand now?
The current national title board is tightly packed at the top. Texas leads at 12.9%, Georgia sits at 12.2%, and Ohio State and Notre Dame share 11.2%. Miami rounds out the double-digit tier at 10%, with Indiana the first name in single digits at 7.6%.
The chart below shows the model's fair value for the leading contenders. The compression is the story: five teams inside a 2.9-point range means small shifts in results can reorder the board quickly.
How fair value becomes a tradable edge
Fair value is only half the equation. The model also reads live quotes across Kalshi and Polymarket, strips the vig, and blends them into one consensus so the target is not any single venue's marked-up price. Edge appears where the cheapest tradable price falls below that consensus.
Indiana is the clean example: a 7c best price on Polymarket against a 7.6% fair value means the contract is priced at or under the model's read. The mirror case is Ole Miss, whose 7c best price on Kalshi sits well above its 4.9% fair value, marking the price as rich. A low number in cents does not equal value; only its distance from fair value does.
The chart below places best price against fair value for two names, one cheap and one rich. Traders comparing venues can weigh promo terms such as Polymarket's TGSWC or Kalshi's FADE, but the model's discipline is simpler: buy only where price trails fair value, and fade where price runs ahead of it.
What the model does not claim
The output is a probability, not a promise. A 12.9% fair value on Texas means the model expects the title to slip away nearly seven times in eight. Ratings drift as results arrive, injuries and matchups reprice paths, and the consensus moves with them.
Prices and the model can both be wrong, and this is analysis rather than financial advice. The framework's value is consistency: the same pipeline runs on every team, so the board stays internally comparable even as individual numbers change week to week.
