Chalk of the Day, 7 August 2026: Shohei Ohtani to homer
Most days this series takes the bet everyone is already on and explains why the price is worse than it looks. Today is the other kind. The chalkiest home run prop on the board came back Positive EV, which has happened twice in nineteen graded selections this slate.
Written by Jesse, NegativeEV. Last updated 7 August 2026.
Watch: https://youtube.com/watch?v=eO-TftOG6yo
- Sample: 5,000 pitch-by-pitch simulations of one game
- Window: 7 August 2026 slate
- Priced at: +276 at DraftKings, +275 across 8 books, read on the day of the game
- Measured: 2026-08-07
The bet
:
| Selection | Shohei Ohtani, over 0.5 home runs |
|---|---|
| Game | Los Angeles at Arizona, 6:40 PM |
| Price | +276 at DraftKings (+275 across 8 books) |
| That price asks for | 26.6 out of 100 |
| Our simulations say | 31.5 out of 100 |
| Gap | +4.9 points |
| Verdict | Positive EV, +19% at +276 |
| Stops being a play at | +217 or shorter |
Is the gap real, or is it noise?
Across 5,000 simulations the simulated number carries a standard error of 0.66 points, so 95% of re-runs land between 30.2 and 32.8 out of 100. The gap to the price is 4.9 points, about seven times that error bar.
Say that narrowly, because it is easy to overclaim: all it means is that re-running the simulator would not move the number this far. It is not a claim that we are right and the book is wrong. A number that is systematically off is off to as many sigma as you like.
What the simulations actually did
Across 5,000 simulations he averages 0.37 home runs and the median is zero. The single most common outcome is zero, 68 times in 100. One homer 27 in 100. Two or more 5 in 100, and the anytime line pays exactly the same for all of them.
He comes to the plate 4.88 times on average. The number moves hard with the trips:
Trips to the plate: share of simulations, and how often he goes deep:
| Trips to the plate | Share of simulations | He goes deep |
|---|---|---|
| 4 | 23 in 100 | 19 in 100 |
| 5 | 61 in 100 | 33 in 100 |
| 6 | 14 in 100 | 45 in 100 |
Nineteen on four trips against forty-five on six. Whether he bats a fifth time is worth more here than any swing-level read.
What we neutralised, one at a time
Every factor is switched off on its own and the game is re-simulated, so each row is that factor's own contribution rather than a story told after the fact.
Working for the over — take it away and the number falls:
Each factor's own contribution, and where the number lands without it:
| Factor | Worth | Neutralised |
|---|---|---|
| His own career profile — who he is as a hitter | ~8.1 points | 31.5 → 23.4 |
| The weather | ~5.7 points | 31.5 → 25.9 |
| His own season form | ~3.1 points | 31.5 → 28.4 |
| On the road, so his side bats all nine innings | ~1.1 points | 31.5 → 30.5 |
The hitter is the biggest single thing in the model, which is the least surprising sentence in this page. Career HR rate .086 puts him 6th of 283 hitters, exit velocity 94.3 puts him 3rd.
The weather is the one worth pausing on. It is 107°F in Arizona tonight with 8 mph of wind blowing left to right — not out to the bleachers, but hot air is thin air, and 5.7 points is a larger contribution than his season form.
Working against the over: nothing cleared its own noise floor. The ballpark came in at +1.7 points — Chase Field is 13th of 30 for home runs, essentially the middle of the pack — but its range crosses zero, so there is nothing to say about it on camera and nothing said about it here.
Positive EV is not a promise
This is the part that matters more than the pick, and it is the reason this series is not a tip sheet.
An edge of this size is small against the variance of a longshot. Betting this exact price and this exact probability, we ran the arithmetic out:
Chance you are ahead, by number of bets placed:
| Bets placed | Chance you are ahead |
|---|---|
| 100 | 86% |
| 200 | 93% |
| 300 | 96% |
| 500 | 99% |
So after 300 bets there is still about a 1 in 26 chance you are behind, and it takes roughly 500 before being ahead is a 99% proposition. That is not a knock on the edge. It is what an edge of 19% on a +276 shot looks like when you stop rounding variance away.
The rest of the slate
Fifteen games had odds today. Nineteen of the chalkiest selections were graded against the simulations: 14 came back Negative EV, 3 roughly fair, and 2 Positive EV. Kyle Schwarber, yesterday's subject, is heavily Negative EV today at +230 against a simulated 21.1 out of 100.
Method
The game is simulated pitch by pitch 5,000 times from the projected lineups, and the prop is graded on how often it lands. Each factor above is a separate re-run of the same game with one input replaced by a league-average stand-in, so those sizes are one-at-a-time swings and do not add up to the total.
Inside the noise floor
One prop in one game is a single draw. A 31.5-in-100 event does not happen most nights, so this being a good price is entirely consistent with Ohtani not homering. We check a price the public is already taking. We do not pick it.
Written up for readers: https://negativeev.com/about/home-run-props All findings: https://negativeev.com/research