Eleven games on the board, one of them a split doubleheader. The screen is the usual one — best price no worse than −160 where we could verify it, at least 80% over one hit-rate window, across bases, hits, RBIs, runs, strikeouts and the hits+runs+RBIs combo — deduplicated to one leg per player and spread one per game.
Today the matchup layer removed more than it kept. The two largest Edge Gaps we generated, plus six of the next ten, are not on this card. That is not a hedge or a soft sell — it is the layer doing the only job it has. The reasoning for every removal is below, with the numbers, so you can disagree with us specifically rather than generally.
The Card
Lines and hit-rate windows captured from our props screen Mon Aug 17 · 7:55–8:15 AM ET. Odds shown are the best price we could read directly; several books were collapsed behind a "+N more" control we could not expand, so better numbers may exist — shop your own. Implied probability and Edge Gap are computed in code from the odds in the table above, never typed by hand. Implied probability includes the vig, so Edge Gap is a ranking tool, not a profit projection.
First pitch check. The earliest game today is St. Louis at Cincinnati at 1:40 PM ET, and no leg on this card is in it. Every leg above is in a game starting 6:05 PM ET or later (Arizona–Boston 7:10, Athletics–Kansas City 7:40, Atlanta–Minnesota 7:40, White Sox–Cubs 8:05, San Diego–Mets 7:10). Start times confirmed against the league schedule this morning. Confirm your own before placing.
The reasoning
Adley Rutschman — Under 0.5 Runs (−145). The largest Edge Gap that survives the matchup gate. He clears this in 100% of his last five, 90% of his last ten and 80% of his last twenty, on a 69% season rate, against a price implying just under 60%. A runs-scored Under is one of the cleanest props in baseball because it needs two independent things to fail — he has to reach, and someone behind him has to drive him in. The honest weakness is his head-to-head number (33%), which we are reporting and not weighting; it is a handful of games and we treat H2H as a tiebreaker only.
Isaac Collins — Under 1.5 Hits+Runs+RBIs (−119). A near-identical edge to Rutschman's at a much better price. 100% last five, 90% last ten, 75% last twenty, 60% on the season. That season number is the soft spot: it is the heaviest-weighted input at 40%, and 60% is only modestly better than the 54% the price implies. The recent windows are doing most of the work here, which is a thinner argument than it looks on the table.
Ryan Jeffers — Over 0.5 Hits (−210). The highest blended rate on the entire board at 87.0 — 80% on the season, 90% over the last twenty, 90% over the last ten, 100% over the last five, an ascending gradient rather than a decaying one. He is a right-handed bat facing Martin Perez, a left-hander, which is the direction you want. The price is the problem: −210 is heavy, and it eats most of a genuinely excellent history. This is the clearest example on today's card of why we rank on the gap rather than the hit rate — the best history on the board finishes third.
Randal Grichuk — Over 0.5 Hits (−150). 100% last five, 100% last ten, 75% last twenty, 66% season, and a right-handed bat against Shota Imanaga. Same structural shape as Jeffers at a friendlier number and with a weaker season baseline.
Jorge Polanco — Under 1.5 Hits+Runs+RBIs (−150). The smallest edge we are publishing and the one we would drop first. 74% on the season and 80% across each of the last five, ten and twenty — unusually flat, which we read as a stable role rather than a hot streak. He is also the player we were wrong about on Friday, when we screened him out by inventing a data-integrity problem that did not exist. He is a Met, he starts, and the correction stands.
Why the two biggest edges are not here
1. Brandon Young Over 3.5 Strikeouts — the largest edge on the board, dropped. Young screened at a blended 84.0 against a −119 price implying 54.3%, an Edge Gap near +30 and comfortably the biggest number we generated today. He is confirmed as Baltimore's starter against Tampa Bay at 6:05 PM ET. We are not playing it, because Tampa Bay has struck out fewer times than any other team in Major League Baseball this season — 875 strikeouts, a 19.1% rate, dead last in the league at making the mistake this prop requires. A strikeout Over is a bet on the opposing lineup, not on the pitcher alone, and this is the single worst lineup in baseball to be making it against. That is the strongest form of conflict our rules recognise, so it is a drop rather than a demotion.
2. An entire ballpark, dropped. Six of the ten largest remaining edges were Unders in Los Angeles at Colorado — Teoscar Hernández on four separate markets, Freddie Freeman on three, plus Connor Norby, Tommy Edman and Cole Carrigg. The biggest was Freeman Under 1.5 Bases at plus money, roughly a +25 gap. They are all dropped for the same one-line reason: Coors Field. It is the most run-inflating park in baseball by a wide margin, and a hit-rate window built mostly on road games at sea level does not transfer to it. Stacking a card with Unders in the one stadium designed to break them would be a systematic error, not a diversified position.
3. Brandon Lowe Over 0.5 Bases — dropped on handedness. A +15 gap, and he is a left-handed hitter for Pittsburgh drawing Framber Valdez, a left-handed starter. Lowe's career platoon split against left-handed pitching is pronounced and long-established. We did not pull a 300-plate-appearance split to put a number on it, so we are not going to quote one — but the direction is clear enough to drop the leg rather than publish it with a caveat.
4. Three legs in St. Louis at Cincinnati — dropped on ambiguity, not on merit. JJ Bleday (+20 gap), Dane Myers and Noelvi Marte all screened well. That series is a split doubleheader today, 1:40 PM and 6:40 PM ET, with different starters in each game. Our data source lists one row per player without saying which game it prices. We could not resolve it, so we are not publishing it. A leg we cannot tell you how to place is not a leg.
The case against our own #1 pick
Rutschman's edge rests on the weakest ingredient we have, and we did not check the thing most likely to break it. Two specific holes. First, Under 0.5 Runs is a lineup-position bet in disguise. If he bats sixth or seventh tonight instead of second, the number of plate appearances drops and the Under gets easier; if he is hitting near the top with the middle of the order behind him, it gets much harder. Lineups for a 7:10 PM game are not posted at 8:20 AM, so we published this without knowing where he hits. That is the single largest uncontrolled variable on the leg and it is not small.
Second, we did not pull his platoon split against Arizona's starter, Mitch Bratt, and we do not have a reliable read on Bratt at all — he is not a pitcher with a long track record to check. So the matchup verdict on our top pick is Neutral by default rather than Neutral by verification, and those are different things. If you want one reason to pass on this leg, that is it.
Second, we did not pull his platoon split against Arizona's starter, Mitch Bratt, and we do not have a reliable read on Bratt at all — he is not a pitcher with a long track record to check. So the matchup verdict on our top pick is Neutral by default rather than Neutral by verification, and those are different things. If you want one reason to pass on this leg, that is it.
What we did not run today
No projection comparison, so no three-signal test. Today's card is Edge Gap plus the matchup layer only. Without a projected value for each player we cannot compute a Projection Gap, which means we cannot tell you whether an independent model lands on the same side of these lines. Treat all five legs as two-thirds checked.
No game-lines post and no Crowd vs Cash column. The public betting splits we build both on — percentage of bets against percentage of money — sit behind a page on our data provider that would not load this morning after repeated attempts. We are not going to reconstruct those numbers from memory or estimate them, so both are skipped rather than faked. If the page comes back we will publish the column late rather than not at all.
No full injury sweep. We checked availability where it bore directly on a leg. We did not run a slate-wide out/doubtful/questionable pass, which means we cannot promise that a ruled-out teammate has not quietly changed the usage picture behind one of these five players.
No game-lines post and no Crowd vs Cash column. The public betting splits we build both on — percentage of bets against percentage of money — sit behind a page on our data provider that would not load this morning after repeated attempts. We are not going to reconstruct those numbers from memory or estimate them, so both are skipped rather than faked. If the page comes back we will publish the column late rather than not at all.
No full injury sweep. We checked availability where it bore directly on a leg. We did not run a slate-wide out/doubtful/questionable pass, which means we cannot promise that a ruled-out teammate has not quietly changed the usage picture behind one of these five players.
Odds and lines move; everything above was recorded pregame and may differ by the time you read it. Always confirm your own price and that the player is in the lineup.
Sources: hit-rate windows, lines and prices from our props screen (our premium data platform), captured Mon Aug 17 7:55–8:15 AM ET. Game times, confirmed probable starters and roster affiliations verified against ESPN's MLB schedule and Aug 16 box scores. Tampa Bay's league-low strikeout total via StatMuse.