Twelve MLB games sit on the August 21 board. Every single one shows a home spread of ±1.5 with zero dispersion across 4 to 7 books. The market has spoken in unison.
MLB picks today: the full slate
The table below lists every MLB game with its home spread, book count, and dispersion as of the 02:07 UTC snapshot. All numbers come from game_odds rows cited by source ID.
| Game | Home spread | Books | Dispersion | Source |
|---|---|---|---|---|
| Atlanta Braves @ Milwaukee Brewers | -1.5 | 7 | 0 | line-05831e33d31b0aacde409a089a422e6b |
| San Francisco Giants @ Boston Red Sox | -1.5 | 6 | 0 | line-c9714ca149113493d7a398fc85aeff83 |
| Washington Nationals @ Miami Marlins | -1.5 | 4 | 0 | line-4c979daade3eabe626041b803b8ed3dc |
| Tampa Bay Rays @ Baltimore Orioles | -1.5 | 6 | 0 | line-970948ad3e20330dce82c2d540759201 |
| New York Mets @ Chicago White Sox | -1.5 | 6 | 0 | line-e7241087266735b11507872ed82ce40c |
| Athletics @ Houston Astros | -1.5 | 4 | 0 | line-dc562897351ca66ed2c1bca057a99133 |
| Detroit Tigers @ Kansas City Royals | 1.5 | 6 | 0 | line-a3bba280f85f30544dc7f29ab7717ebc |
| Los Angeles Angels @ Texas Rangers | -1.5 | 2 | 0 | line-d084d47c393be04d4f0ce4668194dce4 |
| Cleveland Guardians @ Colorado Rockies | 1.5 | 6 | 0 | line-b9462d69bbe4a0609c8c894fe1ecd438 |
| Cincinnati Reds @ Arizona Diamondbacks | -1.5 | 6 | 0 | line-7bf7382fa8788015ccab3559e8c13234 |
| Chicago Cubs @ Seattle Mariners | 1.5 | 6 | 0 | line-f4568c9b73d981125c0bdb5283479d57 |
| Pittsburgh Pirates @ Los Angeles Dodgers | -1.5 | 6 | 0 | line-42af37af9d3d180bae0e2031201a4a38 |
Nine games list the home team as a -1.5 favorite. Three games — Tigers at Royals, Guardians at Rockies, Cubs at Mariners — list the home team as a +1.5 underdog. Zero games show any spread disagreement between books.
Contrast that with the NFL and college slates in the same snapshot. Jets at Steelers shows a 6-point spread range across 71 books. Packers at Broncos shows a 1.5-point range. Commanders at Lions shows a 3-point range. MLB lines are tight because starting pitchers, weather, and lineups are largely known.
What zero dispersion means for you
When seven books post -1.5 on Braves at Brewers with no gap, the market price is -1.5. There is no shopping edge at the spread level. You either like the number or you do not.
Angels at Rangers only has two books reporting. Thin coverage can mean the line is stale or that sharper books have not posted yet. Two-book consensus is weaker evidence than seven-book consensus.
The model-vs-market gap — the only edge we publish — lives in the difference between our model spread and this market spread. That gap is not in the game_odds table. It is computed in our MLB picks feed where each model's projected run line sits next to the consensus market line.
How to read the trigger price
A trigger price is the market line that makes a model edge bettable. If our model projects Royals -0.5 (Tigers favored by 0.5) and the market shows Royals +1.5, the gap is 1 run. The trigger is Royals +1.0 or better. We do not bet at +1.5. We wait for the market to move to our number or cross it.
This discipline prevents betting a model lean just because it exists. The market number must cooperate. On a board where every game shows zero dispersion, the trigger is a single, clear level per game — no book-shopping required.
Where the model numbers live
The MLB picks page shows every game with three columns: model spread, market spread, and gap. Green gap means the model likes the home side more than the market; red means the opposite. The trigger price is the market spread that would close the gap to zero.
Each pick is tagged with the model name that produced it — tfjs-nfl-spread, xgboost-brick, or a user-published model from the marketplace — so the call can be graded later. No outcome guarantees. Just a numbered forecast next to a market price.
What would change our mind
A late starting-pitcher scratch that moves any of these lines by 0.5 runs or more. A weather delay that pushes a total but leaves the spread untouched. A market move to ±2.0 on any game without a personnel change — that would signal sharp money disagreeing with the consensus. Absent those, the board stays at ±1.5 and the model-vs-market gap is the only signal worth tracking.
For live model spreads, trigger prices, and closing-line-value grades, see /picks/mlb. For the bet-tracking workflow that turns trigger prices into logged tickets, see how to track your bets. For the model builder that lets you train your own MLB artifact, see Build → New model.
Sources
game_odds:05831e33d31b0aacde409a089a422e6b— Braves @ Brewersgame_odds:c9714ca149113493d7a398fc85aeff83— Giants @ Red Soxgame_odds:4c979daade3eabe626041b803b8ed3dc— Nationals @ Marlinsgame_odds:970948ad3e20330dce82c2d540759201— Rays @ Oriolesgame_odds:e7241087266735b11507872ed82ce40c— Mets @ White Soxgame_odds:dc562897351ca66ed2c1bca057a99133— Athletics @ Astrosgame_odds:a3bba280f85f30544dc7f29ab7717ebc— Tigers @ Royalsgame_odds:d084d47c393be04d4f0ce4668194dce4— Angels @ Rangersgame_odds:b9462d69bbe4a0609c8c894fe1ecd438— Guardians @ Rockiesgame_odds:7bf7382fa8788015ccab3559e8c13234— Reds @ Diamondbacksgame_odds:f4568c9b73d981125c0bdb5283479d57— Cubs @ Marinersgame_odds:42af37af9d3d180bae0e2031201a4a38— Pirates @ Dodgers
EV per $100 across win rate × odds grid
Expected value of a $100 stake at each combination of true win rate and market odds. Anywhere the cell is positive you have a long-run profitable bet; the magnitude shows how aggressive Kelly will size it.
Model calibration: predicted vs observed
Predicted win probability bucket vs the empirical win rate inside that bucket on the test set. Points on the y=x reference line are perfectly calibrated; points below mean the model is overconfident in that bucket.
