Model Percentile Scorecard v10.1

Fully deterministic JS pipeline computes all projections, edges, and Kelly sizing before the AI layer. Claude validates and writes narrative only — structurally cannot override math, add picks outside the candidate table, or increase units. Zero bypass paths.

Sport-specific cover probability caps prevent overconfidence. Quarter-Kelly sizing with 0.5u step rounding. Hard rules: never worse than -250, minimum edge per sport, EV > 8% floor (raised from 3% in v10.1 to eliminate noise edges). Sport-weighted Kelly multipliers penalize weak sports (NBA 0.85x) and boost proven ones (NHL 1.30x). Every cap is enforced in code, not by prompt instruction.

4u daily exposure cap. 0.5u-3.0u range per pick. Drawdown detection triggers 25% Kelly reduction after 3+ consecutive losses. Parlay capped at 0.5u/day. Institutional bankroll management adapted for retail scale.

Cross-sport edge ranking via normalized z-scores. Sport-specific standard deviations calibrate edges to comparable scales. Edge discount curve compresses outlier edges (log compression above 1 std dev). v10.1 raises NBA minimum edge to 3.0 pts (was 2.0) to filter coin-flip plays, and the 8% EV floor eliminates marginal candidates that were previously diluting the card.

Three-method ensemble: Elo-based power ratings, sport-specific efficiency models (pace-adjusted for basketball, shot-based for hockey), and PPG averaging. Injury impact model adjusts projections for star player absences. Back-to-back rest penalty (3% scoring reduction) applied for NBA/NHL. Home-court adjustment baked in. All computed deterministically in JS.

Automated daily at 10am EDT via Netlify scheduled function. 15-minute background worker. Picks are final once generated — no mid-day overrides. No human in the loop for generation or delivery.

Five independent data sources: real-time odds (13 sportsbooks), ESPN stats + injuries, daily-updated Elo ratings, prediction market cross-reference (Kalshi + Polymarket), and real-time breaking news (X/Grok). Each source fails independently without taking down the pipeline.

11 leagues active: NBA, NHL, MLB, and 8 international soccer leagues. v10.1 disables NCAAB (48.7% accuracy, -7.0% ROI in 97-day backtest) — quality over quantity. Cross-sport mandate evaluates every game with available lines and selects the highest EV regardless of sport. Sport-weighted Kelly (NHL 1.30x, NBA 0.85x) tilts volume toward proven edge sports.

Spreads, totals, moneyline, and puck lines evaluated across all sports. Market-specific edge thresholds per sport. Alternate spreads and totals attempted with automatic fallback to core markets based on API plan. System considers every available market type per game before selecting highest-edge option.

Type-aligned validation against Kalshi and Polymarket. Parses market titles to classify bet type (moneyline, spread, total) and only compares matching types — totals to totals, spreads to spreads. Multi-market aggregation weighted by volume and bid-ask quality. Proportional unit adjustments in 0.5u steps. Prevents false flags like comparing a moneyline market against a totals pick.

CLV tracking infrastructure built and running daily. Calibration blob storage and drawdown detection active. v10.1 ships the first manual self-correction: sport-weighted Kelly multipliers derived from 97-day backtest (NHL boosted, NBA penalized, NCAAB disabled). Automated CLV feedback loop still pending — architecture ready but needs live v10.1 data before activation. This remains the bottleneck to reaching 95th composite.