market gives it 30%+, model is 15+ points better — unders only
What counts as a value spot — both rules have to be true
1The market gives it at least a 30% chance. The price you'd pay has to be 30¢ or more. This is the rule that keeps penny longshots off the list — a 6¢ ask against an 11% model chance looks like a huge edge in percentage terms and is still a bad bet.
2Our model puts it at least 15 points higher.30¢ ask → needs 45% · 50¢ → 65% · 70¢ → 85%. The gap between the two, in percentage points, is the edge shown in the Polymarket column.
Both rules are tested against the cheaper of the Polymarket and Kalshi prices for the side the model takes that has at least $25 offered at it (or within 2¢) — the price you'd actually pay; a thinner price would move when you bet — and a prop also needs 6+ effective games of history behind it. On this board only unders qualify. Everything clearing both rules is listed, ranked by that edge. Prices come from Polymarket's MLB player-prop markets, so this list stays empty until those are posted. Props waiting on news stay off this list and Top 25 Surest: when the hitter, or a regular in his team's lineup, is day-to-day on ESPN's injury report — until the lineup is posted (a hitter not in it is off too) or he's ruled out or cleared.
🔒 Top 25 Surest
Distribution · last 20
Matchup
Game log
Recent games
🔥 Hit streak leaderboard
longest active over streaks — consecutive games clearing the site's realistic line · tap one to open it
📈 My bets
props you tracked — graded from the box score, saved on this device
⭐ Watchlist
players you've starred — saved on this device
⚖️ Compare players
Add players with ⚖ Compare on the Player tab, then set a stat, line and side there — everyone is scored for that exact prop so you can see who's the better bet.
🎲 Build me a parlay
pick risk & legs — we assemble it from the model board
🧾 Bet-slip scanner beta
upload or paste a screenshot — we read the legs and grade them
Drop a bet-slip screenshot here, , or paste it (Ctrl/⌘-V).
Best-effort OCR of a DraftKings / FanDuel / Polymarket-style MLB slip. Each leg's player, prop, side and line are matched to the model and graded, and same-game legs are flagged as correlated. The image is read in your browser and never uploaded.
💡 Suggest a feature
Got an idea for something we should add to the MLB side? Send it our way — it goes straight to the team.
📒 Past picks — how they held up
Every pick is recorded live by the auto-updater before first pitch (every prop on the Polymarket board with a model chance, plus the site's own board, tagged Top 25 Surest / Value where it qualified) and graded from the official box score. Filter by anything below.
Calibration
When the model said X%, how often did it hit? Bars should match the dotted diagonal.
By week
Hit rate of the filtered picks, week by week (Monday to Sunday; bar label = picks graded).
No graded picks match these filters yet.
Profit — if you'd bet 1 unit on every pick
Running units for the filtered picks that had a recorded market price, staking 1 unit on each at the price you'd have paid. A win at 60¢ returns +0.67u; a loss is −1u. Picks with no market price aren't counted here.
No priced, graded picks match these filters yet — profit tracking fills in once picks with recorded prices are graded.
Breakdown
Record by recommendation list and by prop type for the current filters (the list filter itself is ignored here so every list shows).
Picks
Lists key:T = Top 25 Surest — up to 25 props with a 90%+ model chance and a live Polymarket or Kalshi priceV = Value — the cheaper of Polymarket and Kalshi priced it at 30¢+ and the model was 15+ points higher, over or under
We use only essential local storage for your preferences and cookieless, privacy-friendly analytics — no ads, no cross-site tracking. Cookie Policy · Privacy