MLB Home Run Props: Modelling 'To Hit a Home Run' Bets

Home-run props are MLB’s most volatile single bet

An MLB home-run prop is the highest-variance single ticket on the slip. The to-hit-a-home-run market, where a UK bettor backs a specific batter to clear the wall once or more in a single game, sits at typical odds of +400 to +900 – and that range alone tells you the implied hit rate is a coin flip with most of the coin underwater. This is not the same animal as a full-game total or a moneyline. It is the most expensive single prop on a UK MLB coupon, and you need to know exactly why before you press a unit on it.

I want to set the frame here clearly. I bet HR props selectively, perhaps two or three a week during the regular season, and only when the model and the conditions align hard. The market is volatile enough that anyone telling you it is a “reliable income” angle is either lying or has not run the numbers. What we are doing in this article is identifying the narrow set of conditions where the price the book is offering is meaningfully softer than the true probability – and how to size into that without bankrupting a season.

How ‘to hit a home run’ is priced

A typical UK book lists the to-hit-a-home-run prop at odds reflecting a roughly 12% to 22% implied probability for a power hitter in a normal park, scaled up for elite sluggers in good spots and down for contact bats in pitcher-friendly matchups. The hold built into these markets is heavy – often 7% to 10% – because the volatility makes books defensive on the price.

For context on the broader hold environment, US legal sportsbooks ran at a 7.01% hold in 2019 and pushed it to 9.13% by 2023, with US-licensed operators returning $13.71 billion on $149.8 billion of handle in 2024 at an aggregate 9.3% hold. UK books on MLB props sit in a similar band. On HR props specifically, the hold runs higher than that aggregate because of the variance – books need a thicker margin against a single low-probability outcome moving sharply.

The implication is mathematical. If the true probability of your hitter going deep is 18% and the book is implying 14%, you are getting genuine value. If the book is implying 16% on a true 18%, your edge is real but tiny – and HR props at a tiny edge will not survive variance over a season. The bar for a HR prop in my book is at least three points of edge between true and implied probability. Below that threshold I pass, no matter how clean the match-up looks.

Hitter-side stats: ISO, barrel %, pull rate

The first stat I look at is ISO – Isolated Power, calculated as slugging percentage minus batting average. ISO strips out singles and gives you a raw measure of extra-base power per plate appearance. An ISO above .220 is elite power. Above .260 is generational. The current league average sits around .160 to .170. ISO over the last 60 days, not season-long, is the input that matters – power streaks are real, and so are slumps, and the prop market often prices off a longer rolling window than the bettor should.

Barrel rate is the modern statcast measure that has changed the game for HR-prop modelling. A barrel is contact at the optimal exit velocity and launch angle to produce extra-base hits at high rates. A hitter running a 12% barrel rate is more likely to homer than one running 7%, even if their counting stats look similar. I weight barrel rate heavily in my projections because it is descriptive of real contact quality, not just outcome.

Pull rate matters when you stack it with park dimensions. A right-handed pull hitter in Yankee Stadium sees a 314-foot right-field porch. The same hitter in Comerica Park sees 330 feet down the right-field line and a deep right-centre power alley. A 38% pull rate against an opposite-handed pitcher in a park with a friendly pull side is a stack-the-edge spot. A 22% pull rate in a deep-park-pull setup is a fade.

The other stat I refuse to skip is plate appearances. A hitter only homers when he is hitting. A leadoff bat in a high-scoring lineup sees five PAs in most games; a number-eight bat sees three or fewer. Holding everything else equal, a leadoff hitter has roughly 60% more chances to clear the wall than the eight-hole. The market is bad at pricing that across the board.

Pitcher side: HR/9, fly-ball rate, gopher games

The pitcher you are betting against is half the bet. Three numbers matter – HR per nine innings, fly-ball rate, and the related stat of HR per fly ball. A starter with a 1.6 HR/9 over the last 30 days is a different opponent than one running 0.7. The book usually knows this, but the magnitude of adjustment in the HR-prop number is often smaller than my model suggests it should be.

Fly-ball rate is the precursor. A starter who induces ground balls 55% of the time gives up roughly half as many homers as a 38%-ground-ball pitcher with otherwise similar stuff. So when I am modelling a HR prop, I want a hitter against a high-fly-ball starter, in a park where fly balls carry, in conditions that further amplify carry. That is the stack.

I use the term “gopher game” for a starter on a clear HR-allowing trajectory – high fly-ball rate, declining velocity, recent allowed-HR cluster. These are the games where multiple HR props from the same lineup can have value simultaneously. They are also where the bookmaker is fastest to adjust between pre-game and live, so prices on the slate are often softest at the morning open and tightest 30 minutes before first pitch.

Park and weather stack on HR props

Park and weather are not edge multipliers; they are edge prerequisites. Most of my HR-prop bets in a season are in fewer than half the league’s parks, on days where the wind helps, and against fly-ball pitchers. The data is unambiguous on the magnitude – at 90°F a fly ball travels 4 to 6 feet farther than at 50°F, which converts a meaningful percentage of warning-track outs into home runs. Multiplied across a hitter’s three or four PAs in a game, that 4-to-6-foot effect is a real shift in HR probability.

Wind compounds the temperature effect. A wind blowing out of the park at 8mph or more produces a measurable increase in run scoring and an outsized increase in home runs – that wind direction has historically run a 1174-1045-125 record on overs at 52.9%, returning +84.59 units and a 3.6% ROI for the simple “bet the over with strong out-blowing wind” rule. HR props benefit at a higher rate than full-game totals, because the wind effect is concentrated on the airborne fly balls that determine HR outcomes.

The stack: target hitters in pull-friendly parks against fly-ball pitchers in 80°F-plus games with a tailwind. That setup will not appear every day. Some weekends it will not appear at all. The discipline of waiting for the stack – rather than betting HR props daily – is what separates the model that works from the one that bleeds vig.

How to size a high-vig prop within a bankroll

HR props are where I size smaller than my baseline unit. If my standard MLB unit is 1% of bankroll on a moneyline or run line, my HR-prop unit is 0.5%, sometimes 0.25%. This is fractional Kelly applied to a high-variance market – the model edge is real but the variance is enormous, and full-Kelly sizing on a +500 prop risks the kind of drawdown that triggers a UKGC affordability check at the £150 net-loss threshold without my consent.

I also cap the slate. No more than two HR props per day, no more than five per week, no parlays involving multiple HR props on the same slate. Same-game HR-prop parlays look like the dream ticket. They are the worst bets on the slip – the bookmaker prices them at compounded juice and the variance multiplies. If you want park-and-weather exposure across a slate, a moneyline on the highest-total game is structurally better than a parlay of HR props from that game. For framing the same logic on broader temperature plays, see my work on MLB temperature betting.

Last sizing note. Track every HR-prop bet separately from your main bankroll log. The variance is so different from your run-line and total bets that mixing them in a single ROI calculation will mislead you. I keep a separate column for HR props specifically so I can see if the segment is paying.

The honest payoff on a volatile prop

HR props will not be your bread-and-butter market. They cannot be – the variance forbids it. But run with discipline, in the right park-pitcher-weather stack, with sizing scaled to the variance, they produce a small positive ROI over hundreds of bets. The question to ask before every HR-prop ticket is the same one – am I betting because the conditions stack, or because the +700 number looks fun? If the answer is the second, the bet is a leak. If the answer is the first, the bet is part of the model. The discipline is in being honest about which one you are pressing.