SCHEDULE DIFFICULTY · EXPERIMENTAL

How hard was the road?

Independent opponent ratings built from completed results, with diminishing returns for blowout margins. Compare the full-view retrospective with a chronological reconstruction.

Higher mean opponent ratings indicate a tougher observed schedule. Rating units are not points or win probabilities.

Read the timing labels. Historical completion and publication times are unverified. Reconstructed as-played ratings use only results from earlier UTC dates. Strict mode withholds games without verified result timestamps. These results do not replace the preserved historical SOS board.

Evaluation snapshot Schedule view Find a team Maximum results

Rank	Team & opponent detail	Mean opponent rating	Games	Observed inputs

Formula, source coverage and limitations

The result model fits regularized additive opponent ratings to tanh(score margin / 21). Schedule difficulty is the mean rating of opponents actually faced. The experimental ridge is 4. No poll or conference label enters the fit. Venue is shown, but no venue, travel, injury or weather coefficient is applied. Prior-only inputs and incomplete schedules are visible; sample counts are not calibrated uncertainty intervals.

View preserved historical SOS → · Download calculation evidence