The model turns power ratings into fair value in two moves: it simulates each team's schedule from those ratings to produce a national title probability, then de-vigs the live prices across Kalshi and Polymarket into a single consensus number. The result is a clean probability per team, benchmarked against the cheapest contract on the board. Georgia sits on top at 16.8% fair value, with a best price of 19c on Polymarket.
Everything downstream, the best-venue read and the fade signals, flows from that pipeline. The sections below walk through each step and ground it in the current national title board.
What does the model actually compute?
Power ratings are the starting input, a single strength number per team. The model feeds those ratings into a schedule simulation: it plays out the remaining games many times, tracks who reaches the playoff and who survives each round, and counts how often each team wins the national title. The share of simulations a team wins becomes its raw title probability.
That step matters because a title price is not just about team quality. It also encodes path difficulty, conference routing, and how many rivals sit in the same bracket. A strong rating with a brutal draw can still produce a modest title number, which is why the simulation, not the raw rating, drives the probability.
How do market prices become fair value?
Market prices do not translate to probability one for one. Each venue builds in an overround, so the implied probabilities across all teams sum to more than 100%. De-vigging removes that margin, rescaling the field so the numbers behave like true probabilities.
The model de-vigs within each venue, then blends Kalshi and Polymarket into a consensus fair value. Simulation output and the de-vigged market anchor each other: the ratings-based probability sets the model's view, while the cleaned market prices keep that view honest against live liquidity. The published fair value is the reconciled result.
The best-price field is a separate read. For each team the model scans both venues and surfaces the lowest available contract, which is where a given outcome trades cheapest right now.
Where does the gap between fair value and price come from?
On the current board every best price sits above its fair value. Georgia is 16.8% fair against a 19c best, Ohio State 15.6% against 17c, Notre Dame 13.1% against 15c. That consistent gap is the overround showing through: even the cheapest venue still prices each outcome a little rich relative to the de-vigged consensus.
The spread is not uniform. Favorites near the top of the board tend to carry a wider cents gap, while deep longshots compress toward a one-cent floor. Oregon and Oklahoma both show 1.1% and 0.9% fair value against a 1c best, where the price grid simply cannot round any lower.
What the current title board shows
The model's fair value ranks Georgia (16.8%), Ohio State (15.6%), Notre Dame (13.1%) and Miami (11.3%) as the clear top tier, with Texas (7.5%), Alabama (6.9%) and Indiana (6.8%) forming the next group. Polymarket carries the cheapest price on most of that board, including Georgia, Notre Dame, Miami, Texas and Alabama, while Kalshi owns the best Ohio State price at 17c.
The best-price chart below reads as the trade-entry cost for each outcome. Pairing it with the fair value chart is the whole point of the model: it exposes which venue is cheapest and how far that price still stands above the de-vigged consensus.
For readers mapping where to trade each contract, Polymarket (code TGSWC) and Kalshi (code FADE) hold the best prices across this board. Prices and the model can both be wrong, and none of this is financial advice; the fair value is a probability estimate, not a guarantee.
