Moneyball AI — Baseballiana.com

LOADING MONEYBALL AI...

🔬 Skill Translation Index

Select a player to measure how efficiently batting processes convert to production.

Select Player

💥 Pull-HardHit Value Calculator

V = (Pull% × HardHit%) / K% — measures efficient hard contact.

🎯 HR/FB Composite Analysis

Composite score from HardHit%, Pull%, and Barrel% weighted against HR/FB rate.

⚔️ Team Matchup Predictor

Away Team

@

Home Team

Game Date

📅 Today's MLB Games

Select Date

🏟️ Minor League Hitters

Browse hitting stats from DSL/Rookie through AAA with derived metrics.

League Level

🌟 Future Star Identification

Age-adjusted prospect scoring combining OPS, ISO, K%, and BB% with level-specific baselines.

Filter by Level

⚾ Pitcher Profiles

Power/finesse classification, vL/vR splits analysis, and composite scoring.

League Level

🔀 Hitter Splits Lab

Platoon splits (R v L, R v R, L v R, L v L) and vs Power/Finesse analysis with HR/FB, ISO, K%, BB%.

Analysis View

🎯 Pitcher Splits Lab

Pitcher platoon splits and projected performance vs hitter types (GB/FB, Contact/Power).

Analysis View

🎯 Daily Hitter Targets

Scans today's probable starters for their worst platoon split, then ranks every hitter in the opposing lineup who exploits that weakness — adjusted for park factor and wind direction.

How it works: For each game, the starting pitcher's L/R OPS-allowed split is checked. Hitters whose bat side exploits the pitcher's worst split are scored by their platoon OPS × split magnitude × park factor. Wind out at Coors (or Wrigley) will push certain hitters higher in the overall ranking. Requires pitcher profiles + hitter splits data to be loaded.

🗂️ A Priori Matchup Board

Ranks every hitter on today's 2026 active rosters against each probable starter — no lineup card required. Works the moment probable pitchers are announced.

How it works: Fetches each team's 2026 active roster from MLB, matches hitters to our splits DB, then scores them against the opposing probable pitcher's worst platoon split. Park factor and live weather are applied per venue. The result is a pre-game ranked board of every viable platoon exploit on the slate — hours before lineups post.

⚡ CONVERGENCE BACKTEST STUDY

Retrospective validation: did CONVERGENCE-flagged game appearances (high model tier + strong career BvP) actually produce better batting results in 2025? Pulls game logs, identifies opposing starters via boxscore, fetches career BvP, runs Fisher's exact test.

HITTERS TO STUDY

📊 Predictive Validity Study

AI ML Bro Workbench & Nerd Lab

Model validation, machine learning metrics, and advanced analytics workbench.

📈 Model Metrics & Validation

🌬 Daily Weather Edge — Winarick V1 Dynamic

Live wind + temp from Open-Meteo • Altitude air density • Park dimension sensitivity • Outfield orientation alignment

Fetches live weather for all 30 ballparks — ~8s

⚙ Matchup Intelligence Filter

Empirical cross-filter: actual split records × live park conditions × batting order. How does this pitcher/hitter type perform in this environment against this opponent?

Analyze

Park Type

Wind Condition (today)

Opponent Hitter Type

Pitcher Classification

Batting Order Zone

Fetches live weather + today's lineups — ~8s

📊 Weather × Outcome History Collector

Pulls completed MLB games from Stats API + historical weather from Open-Meteo Archive. Builds an empirical record: how many runs/HR actually scored in each wind condition? 100% free, no API key needed.

Days Back to Collect

Game logs stored: — • Each collection run: ~5–25s depending on days. Re-runs skip already-stored games.

📐 Predictive Power Index

Quantifies how much each analytical signal actually contributes to outcome prediction. All effects converted to Pearson r with 95% confidence intervals — Statcast metrics, pitching profile, platoon splits, and weather/park all ranked on a single scale so you know what's signal vs noise.

Uses DB data already on hand — instant. Add game logs for weather signals.

💰 Betting Lines + Implied Runs

Live DraftKings/FanDuel lines via The-Odds-API. Shows game totals, per-team implied run totals, win probabilities, and compares market O/U against our Dynamic Park Factor model. Requires free ODDS_API_KEY secret.

Requires free ODDS_API_KEY → the-odds-api.com (500 req/month free)

🔬 Lab Notebook

Tamper-evident research audit trail. Every logged analysis is SHA-256 hashed and chained to the previous entry — history cannot be quietly altered. Auto-generates APA methods paragraphs and validates 9 completeness checks.

LOG NEW ENTRY

Analysis Type

Season (data)

Roster Season

N Hitters

BvP OPS Threshold

Min PA (BvP)

Model Tier Min

CHAIN INTEGRITY

Re-walks every entry oldest-to-newest, recomputes each SHA-256 hash from its raw fields, and compares to the stored hash. Any silent edit to a logged entry is immediately flagged.

● Entry hash = SHA-256(timestamp ‖ type ‖ params ‖ results ‖ datafp ‖ prev_hash)

● Data fingerprint = SHA-256(current table row counts)[:16]

● Chain broken if any stored hash ≠ recomputed hash

Loading notebook...

Loading...