https://good-sport.ai/ledger Claims we make about a sports model that grades itself — stamped open, held up, or busted as the evidence lands. en-us https://good-sport.ai/ledger/report-card-2 https://good-sport.ai/ledger/report-card-2 Sat, 01 Aug 2026 14:00:00 GMT Report card №2 — batter hits becomes the first family to pass a pre-committed keep gate, the killed pitcher-K family is quietly beating the market post-kill (70 bets, +4.0pp), and tennis sits at 25 of its 30-bet tripwire at −23.2pp. https://good-sport.ai/ledger/report-card-1 https://good-sport.ai/ledger/report-card-1 Wed, 29 Jul 2026 12:00:00 GMT Report card №1 — the first audit under pre-committed rules: nothing fired on audit day, a kill landed three days later at exactly n=150, and a bug ate two weeks of history. https://good-sport.ai/ledger/calibration-not-data https://good-sport.ai/ledger/calibration-not-data Tue, 28 Jul 2026 12:00:00 GMT A note: the money bleed is a probability-layer problem, not a missing-feature problem. The claim is on the ledger and grades when the recalibration lands. https://good-sport.ai/ledger/pitcher-k https://good-sport.ai/ledger/pitcher-k Mon, 13 Jul 2026 12:00:00 GMT The blended prop record looked edgeless. Cut by prop type, it was a real batter edge and a real pitcher-strikeout hole. Graded HELD UP on 7/18 when the pre-committed gate fired at exactly n=150. https://good-sport.ai/ledger/clv https://good-sport.ai/ledger/clv Fri, 03 Jul 2026 12:00:00 GMT Closing-line value — did the model actually beat the market? Graded on my own bets, including the cohort at −40% ROI. Claim OPEN; re-grades at n≈500.