Research notes
Model updates and the measurements behind them: how Newton and the Saturview ratings are measured, what changes from one version to the next, and where the numbers still miss.
How Newton v3 Was Built, Step by Step
Every measured gain, every idea we cut, and the checks we set before scoring anything. The gains are modest, and the method behind them is the part we are proudest of.
In a Two-Season Replay, Moving Lines Leaned Toward Newton v3
In a replay of the 2024 and 2025 seasons, when the consensus point spread moved, it moved toward Newton v3's side 54.7% of the time. Here is what that shows, and what it does not.
We Put Our Rankings Up Against an Outside Model. It Was a Tie.
Same games, same information each week, and a result too close to call. What the comparison taught us about early-season rankings, honest ranges, and the cost of a rule we chose.
The Human Layer: What Stats Miss, and How We Will Test It
Stats can see most of a college football game, not all of it. Newton's next layer is built to read the rest, and it has to earn its place on games played after its calls are frozen.
These notes describe model results and opinions, never betting advice.