Prediction-market prices move for three reasons: liquidity (how much size the order book can absorb), news (anything that changes an outcome's expected probability), and the model's fair value, which acts as the gravitational anchor prices drift back toward. On the current CFB national title board, Notre Dame sits on top at 12 percent fair value and a 12c best price on Polymarket, with Oregon and Ohio State a step behind at 11c. Those three forces explain both why the number changes minute to minute and why it does not drift far.
How does liquidity control the size of price moves?
Liquidity is the depth of resting orders around the current price. A contract with deep two-sided books absorbs a large order with little movement; a thin one reprices on a single fill. This is why favorites behave differently from longshots.
Notre Dame at 12c, Oregon at 11c and Ohio State at 11c draw the most volume, so their prices grind in one-cent increments and rarely gap. Move down the board to Oklahoma at 2c or Texas Tech at 3c and the same dollar order shifts the price further, because there is less standing size to soak it up. The takeaway: a move on a liquid favorite carries information, while a move on a thin longshot may be nothing more than one trader clearing the book.
How does news reprice CFB title contracts?
News is the input that changes an outcome's true probability: a depth-chart change, a transfer, a result once the season starts. When it lands, informed traders adjust and price discovery does the rest, pulling the contract to a new level.
The magnitude of the reaction scales with how much the news shifts expected wins and how concentrated the field is at the top. The market is tightly bunched right now, with Notre Dame at 12 percent, Oregon at 11.1 percent, Ohio State at 10.9 percent and Texas at 9.7 percent all inside three points. In a cluster that tight, a modest piece of news can reorder the top of the board, which is exactly why prices at the top move more than their small spreads suggest.
Where does the model's fair value fit in?
Fair value is the de-vigged consensus the model builds by stripping the margin out of quoted prices across Kalshi and Polymarket. It is the anchor: when the traded price drifts away from it on noise rather than news, the gap is the signal.
Value sits where best price trades below fair value. Where best price meets or exceeds fair value, the vig has already priced out the edge. Alabama is the clean example: a 3c best price against a 2.6 percent fair value means the quoted contract sits above the model's estimate, so there is no discount to capture. The model does not predict the next move; it defines the level that the next move is measured against.
Reading the three forces together
The three forces work as a system. Liquidity sets how far a given order can push the price, news supplies the reason to push it, and fair value marks whether the new level is rich or cheap. A move on a deep book after real news is price discovery; a move on a thin book with no news is a liquidity event that often reverts.
Practically, that means treating a one-cent tick on Notre Dame or Oregon as more meaningful than a larger swing on a 2c longshot. The prices, and the model, can both be wrong, and none of this is financial advice. But the framework holds: separate the liquidity noise from the news signal, then check both against fair value. Best prices across the board currently sit on Polymarket (code TGSWC), with the de-vigged consensus drawn from Kalshi (code FADE) and Polymarket together.
