Player Props · DraftKings Lines & Edges
August 4, 2026 MLB Total Bases Props
35 lines · Updated Aug 4, 3:34 PM ET
The Almanac's Take
The board has 35 lines but zero projections loaded and no top plays surfaced — there's nothing here to build a note around. With `num_with_projection: 0`, the model has no rate inputs to run against the DraftKings lines, so every edge reads null. Worth noting that the methodology runs a low-bias Poisson, meaning any over edges that do emerge would understate the true advantage slightly — but that caveat is moot when the engine has no projections to work with. The TB board is empty today.
How we model these edges
- Model method:
- poisson_per_game
- Approximation quality:
- Approximate
- Bias direction:
- Model edges on overs are biased low — real overs are slightly better than reported.
- Edge definition:
- model_over_prob - no_vig(over_implied_prob)
- ·Poisson approximation under-estimates the tail of total bases; model edges on overs are biased low — real overs are slightly better than reported.
- ·Integer lines (e.g. line=2.0) are treated as 'over wins on ≥2', which slightly overstates over_prob vs sportsbook push rules. Half-point lines (X.5) — the near-universal case for these markets — are unaffected.
Planned improvement: Replace Poisson tail with ML simulation prop_probs (src/projections/ml/simulation.py) once the sim emits the needed thresholds (TB ≥1/≥4/≥5; full K grid) and the backtest validates calibration improvement.
Total Bases Board
0 of 35 projected| # | Player | Line | Odds O/U | Proj | Model % | Market % | Edge | Pick | Conf |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Alec Bohm | 1.5 | +105/ −140 | — | — | 46% | — | — | — |
| 2 | Alec Burleson | 1.5 | +139/ −186 | — | — | 39% | — | — | — |
| 3 | Angel Martinez | 1.5 | +133/ −178 | — | — | 40% | — | — | — |
| 4 | Ben Rice | 1.5 | +117/ −156 | — | — | 43% | — | — | — |
| 5 | Bo Bichette | 1.5 | +126/ −167 | — | — | 41% | — | — | — |
| 6 | Brandon Marsh | 1.5 | +117/ −155 | — | — | 43% | — | — | — |
| 7 | Bryce Harper | 1.5 | −111/ −119 | — | — | 49% | — | — | — |
| 8 | Bryson Stott | 1.5 | +120/ −160 | — | — | 42% | — | — | — |
| 9 | Chase DeLauter | 1.5 | +140/ −187 | — | — | 39% | — | — | — |
| 10 | Dylan Beavers | 1.5 | +134/ −178 | — | — | 40% | — | — | — |
| 11 | Dylan Crews | 1.5 | +143/ −191 | — | — | 39% | — | — | — |
| 12 | Elly De La Cruz | 1.5 | −107/ −124 | — | — | 48% | — | — | — |
| 13 | Gunnar Henderson | 1.5 | +110/ −145 | — | — | 45% | — | — | — |
| 14 | J.T. Realmuto | 1.5 | +129/ −172 | — | — | 41% | — | — | — |
| 15 | JJ Bleday | 1.5 | +127/ −169 | — | — | 41% | — | — | — |
| 16 | Jacob Wilson | 1.5 | +119/ −159 | — | — | 43% | — | — | — |
| 17 | Jeff McNeil | 1.5 | +145/ −194 | — | — | 38% | — | — | — |
| 18 | Jo Adell | 1.5 | +114/ −151 | — | — | 44% | — | — | — |
| 19 | Jonah Heim | 1.5 | +134/ −179 | — | — | 40% | — | — | — |
| 20 | Jordan Walker | 1.5 | +122/ −162 | — | — | 42% | — | — | — |
| 21 | Jose Ramirez | 1.5 | −117/ −113 | — | — | 50% | — | — | — |
| 22 | Kyle Schwarber | 1.5 | −123/ −108 | — | — | 52% | — | — | — |
| 23 | Luis Arraez | 1.5 | +107/ −142 | — | — | 45% | — | — | — |
| 24 | Mike Trout | 1.5 | +130/ −173 | — | — | 41% | — | — | — |
| 25 | Nolan Schanuel | 1.5 | +157/ −211 | — | — | 36% | — | — | — |
| 26 | Pete Alonso | 1.5 | +102/ −136 | — | — | 46% | — | — | — |
| 27 | Sal Stewart | 1.5 | +111/ −148 | — | — | 44% | — | — | — |
| 28 | Steven Kwan | 1.5 | +152/ −205 | — | — | 37% | — | — | — |
| 29 | Tommy White | 1.5 | +153/ −205 | — | — | 37% | — | — | — |
| 30 | Trea Turner | 1.5 | −103/ −129 | — | — | 47% | — | — | — |
| 31 | Trent Grisham | 1.5 | +124/ −165 | — | — | 42% | — | — | — |
| 32 | Tyler Soderstrom | 1.5 | +111/ −147 | — | — | 44% | — | — | — |
| 33 | Tyler Stephenson | 1.5 | +144/ −193 | — | — | 38% | — | — | — |
| 34 | Vaughn Grissom | 1.5 | +145/ −195 | — | — | 38% | — | — | — |
| 35 | Zach Neto | 1.5 | +123/ −164 | — | — | 42% | — | — | — |
More for this date
Want the why behind an edge?
Ask The Almanac about a player's matchup, recent form, or pitcher arsenal in chat — same data, deeper context.