Sharpest Sportsbooks for MLB Player Props: 2026 Data Study

Kalshi and ProphetX post the sharpest MLB player prop lines in 2026, and Bookmaker and Pinnacle give away the most value. That is the short version of what we found after studying over 600 million MLB player prop line movements and 1.3 million closing market lines across six prop types and every major book we could track.

Figuring out which sportsbooks are actually sharp on MLB player props is harder than it sounds. Everyone in the betting community has an opinion, but very few of those opinions are backed by data. We set out to change that. The goal was simple: determine which books post the sharpest lines, which ones give away value, and what that means for anyone betting MLB props this season.

The MLB player prop sharpness hierarchy

Here is where the books landed, sharpest (the reference lines you want to price off of) to softest (the lines you want to bet against):

Tier	Books	What it means for you
Sharpest (reference)	Kalshi, ProphetX	Price off these. Their side of a disagreement is almost always the right one.
Sharp	Novig	A solid reference, just below the two exchanges.
Middle	DraftKings, FanDuel	Tighter than their recreational reputation, softer than the exchanges.
Softest (target)	Pinnacle, Bookmaker	Consistently the value-giving side. If you have access, pay attention.

The rest of this article shows how we got there, and why closing-line accuracy alone would have pointed you in the wrong direction.

How we measured sharpness

We looked at sharpness two ways, because the obvious method and the useful method disagree.

The first is closing-line accuracy. For every MLB player prop across six markets (total bases, hits, RBIs, home runs, strikeouts, and batting walks), we stripped the vig out of each book's over/under pricing to recover its true implied probability, then scored how close that probability was to the actual result. Removing the vig is the same devigging step you run before comparing any two prices, so the number reflects each book's genuine opinion rather than its margin.

The score we used is the Brier score: a lower number means a book's prices were, on average, closer to reality. We ran it on each book's closing line, the final number right before first pitch.

The second method is crossed markets, which we get to below. It measures who is driving prices during the day, not just who is accurate at the buzzer. As you will see, that distinction is the whole story.

Finding 1: by closing time, every book posts nearly the same line

One correction before the numbers. Pooling every line together the way a naive Brier score does quietly rewards the books that post the deepest alt-line ladders. An "over 3.5 home runs" is a near-certain no, and pricing it correctly earns an almost-free low score, so books that flood those extreme lines look sharp for a reason that has nothing to do with skill. To strip that out, we compare each book on the same market, using its main (roughly 50/50) line, with each game counted once. Lower is sharper, and you should only compare down a column, because markets are not comparable to one another.

Book	Bases	Hits	RBIs	Home Runs	Strikeouts	Walks
Hard Rock	0.224	0.235	0.207	0.100	0.245	0.205
Novig	0.231	0.236	0.207	0.099	0.247	n/a
DraftKings	0.240	0.236	0.207	n/a	0.246	0.208
Bodog	0.241	0.236	n/a	n/a	0.244	n/a
Bookmaker	0.242	0.233	n/a	0.110	0.245	n/a
BetMGM	0.244	0.236	0.206	0.091	0.247	n/a
Pinnacle	0.244	n/a	n/a	0.140	0.245	n/a
Fanatics	0.244	0.236	0.208	n/a	0.246	0.214
Bet99	0.244	0.236	0.207	0.096	0.245	0.212
Caesars	0.244	0.235	n/a	n/a	0.246	0.215
PointsBet CA	0.250	0.236	0.208	n/a	0.247	n/a
FanDuel	n/a	n/a	n/a	n/a	0.245	n/a

What jumps out is how little separates the books. On the highest-volume markets the gap between the sharpest and softest book is a rounding error: RBIs span 0.002 of Brier, strikeouts 0.003, and hits 0.005. Bases shows a little more (about 0.025, with Hard Rock and Novig nominally ahead), but the leader changes in every column, which is what you would expect from noise rather than a real edge. Read down any column and, for all practical purposes, the books are posting the same closing number.

The one market with real separation is home runs, and it runs opposite to the usual story: the mainline books price it best while Pinnacle sits well behind the field. But home runs is a low-probability market where the Brier score is driven by how a book handles longshots, not by pricing skill, and when we checked whether Pinnacle's weaker number was actually beatable, it was not. So even the single market that separates the books does not hand you an edge.

The honest takeaway is that closing-line accuracy tells you very little about who to bet. By the time first pitch arrives, the market has consolidated, every book has watched every other book and landed on essentially the same price, so measuring who is closest at that point mostly measures noise. That is not a knock on the books, it is the nature of a closing line, and it is closely related to why closing line value is the metric sharp bettors actually track.

It does, though, point straight at the more useful question. If every book ends up on roughly the same number, the edge was never in where they land. It was in who got there first, and which books were slow enough along the way to be handing value to everyone else. That is exactly what crossed markets show you.

Finding 2: crossed markets expose who is soft

A crossed market is when two books price opposite sides of the same prop so generously that betting both sides covers each other. For example, if one book has a player's hits over at +105 and another has the under at +105, staking both sides returns more than you risked no matter how the game ends, at least on paper. That is arbitrage.

In practice, nobody wants to grind 1% arbs on $100 limits, and a crossed market is not truly free money once you account for line moves between placing the two bets, lowered limits, and account restrictions. The more useful signal is figuring out which side has the real value. When a sharp book and a soft book disagree enough to cross, one of those prices is wrong. The soft book is offering a better price than the outcome justifies, and that is the side you want.

You can watch this happen in the minutes around a news event. When the lineups drop, ProphetX (the sharpest book in our study) reprices in seconds while Fanatics (the softest) is slow to follow. For the couple of minutes Fanatics lags, the two books are crossed: its stale over sits on the wrong side of the ProphetX line, and that gap is the value. The window closes the moment Fanatics catches up. Scrub across the chart to see it open and close:

So we tested it directly. We took books across every tier (prediction markets, exchanges, sharp offshores, softer domestic books) and ran them against each other in every possible pairing. For each pairing we scanned the full dataset for every MLB player prop crossed market, then graded every one against the actual result, tracking what would have happened historically if you had only ever bet one book's side.

The pattern was consistent: across a crossed market, the book with the higher historical return on its side is the softer book, the one giving away value, and the book with the lower or negative return is the sharper reference point. Over a large enough sample, that separation is meaningful. It is worth stressing this is a backward-looking measurement, not a prediction, and any individual bet can still win or lose.

The book-by-book verdict

Kalshi and ProphetX are the sharpest books in the dataset. Across nearly every matchup, their side of a crossed market produced negative or near-zero returns, meaning there was very little to gain betting against them. These are the lines that track true probability most closely, and the ones to devig against when you are deciding where fair value sits.

Kalshi had the tightest pricing across the board, which tracks: it is one of the fastest-moving, most liquid prediction markets in the US right now, with a large and growing pool of participants trading MLB. You can price its markets directly with our Kalshi odds converter. ProphetX actually edged Kalshi head to head, a mild surprise, likely because Kalshi's player prop markets are still newer and thinner. ProphetX recently earned CFTC approval as a regulated prediction market, and both exchanges should only sharpen as volume grows. For now, treat them as the two gold-standard reference points for MLB props.

Novig is sharp, but it trails the exchanges. Novig posted negative returns against most books it was paired with, confirming it as a solid sharp reference, but ProphetX cleared it head to head and Kalshi beat it comfortably. It is meaningfully better than any traditional sportsbook, just a notch below the two prediction markets.

DraftKings performed better than expected. DraftKings was more competitive than a typical recreational sportsbook, beating FanDuel in several pairings. It handles enough MLB prop volume that its lines are tighter than they are sometimes given credit for. Not a sharp reference book, but not as soft as Pinnacle or Bookmaker either.

FanDuel is not as sharp as its reputation suggests, at least on MLB. FanDuel is often called the gold standard for props, and that reputation is probably deserved for NBA and NFL where it handles enormous volume. MLB props are a different story. It does not post lines as early as other books, its sample in our Brier analysis was a fraction of competitors, and in crossed-market testing the sharper side consistently beat it. It still clears Pinnacle and Bookmaker, but any exchange or prediction market clears it cleanly.

Pinnacle is slow and soft on props. Pinnacle's side of a crossed market was consistently the one you wanted to be on, which matches what we see day to day. Its player prop limits are tiny compared to its mainline offerings, a signal it is not putting serious modeling into these markets the way it does for totals and spreads. If you have access to Pinnacle props, they are worth watching. For those in the US without direct access, Bet105 is a Pinnacle clone available domestically with essentially the same lines.

Bookmaker was the softest book in the test. Taking Bookmaker's side of a crossed market gave the highest historical returns of any book, meaning it gave away the most value relative to true probability. Bookmaker used to be one of the sharpest offshore books, but it has clearly stepped back on player props: slow to move lines, lowered limits, and reportedly cutting sharp bettors off from props entirely, which usually means it was getting beaten badly enough to become a problem. If you still have props access there, that is where the softest lines in this study lived. Just expect the limits and restrictions that follow anyone who wins.

What this means for betting MLB props in 2026

Is there still value to find on MLB props? Yes, but it has gotten harder, and this data explains why.

The rise of exchanges like Kalshi and ProphetX has changed how MLB prop lines form. They are fast, sharp, and increasingly liquid. They move first, the market follows, and traditional sportsbooks that used to set the pace are now playing catch-up. Blind positive EV strategies that relied on a simpler market hierarchy are less effective than a few years ago, because the sharp end of the market has gotten sharper.

The soft end has not disappeared, though. Bookmaker and Pinnacle still post lines that give away value in crossed markets, and FanDuel lags the exchanges more than its reputation suggests. These books are still slow to react to news, lineup changes, and injury updates. That is where the edge lives.

The most practical strategy this points to is one built around an information hierarchy:

• Use Kalshi and ProphetX as your reference prices, and weight Novig just behind them. This is the same logic behind our Smart Money signal: follow where the sharp money sets the number.
• When you see either exchange crossed with a domestic book, that gap exists for a reason, and the exchange side is almost always the sharp line.
• If you would rather not place the arb itself, taking the soft book's side has, across this historical sample, been the higher-value side far more often than not. That is a backward-looking result on our own data, not a prediction. Results vary, and any single bet can lose.

The most systematic way to execute is to pull all these books into an arbitrage tool and hunt crossed markets with a clear sense of which side to take. Our Arbitrage Finder surfaces those crosses in real time, and the hierarchy above tells you which side is the value. That hierarchy alone (exchanges at the top, Novig behind them, DraftKings and FanDuel in the middle, Pinnacle and Bookmaker on the soft end) is a workable filter for finding value on MLB player props this season.

Prefer to test all of this yourself? The full graded dataset is public on Hugging Face, so you can rebuild the hierarchy from the raw line movements and check every claim in this piece.

Frequently asked questions

Which sportsbook is sharpest for MLB player props? Kalshi and ProphetX are the sharpest books for MLB player props in 2026, based on a study of over 600 million line movements. Their prices track true probability most closely, so their side of a crossed market rarely leaves value on the table. Novig is the next-sharpest, and it sits just below both prediction markets.

Is Pinnacle sharp on MLB player props? Pinnacle is not sharp on MLB player props, despite its sharp reputation on totals and spreads. Its side of a crossed market was consistently the value-giving one in this study, and its tiny prop limits signal it does not model these markets seriously. It is a book to bet against on props, not to use as a reference.

Are prediction markets like Kalshi better than sportsbooks for props? Prediction markets like Kalshi and ProphetX post sharper MLB prop lines than traditional sportsbooks, because they are faster, more liquid, and move first while sportsbooks follow. Use them as your fair-value reference, and treat any domestic book crossed against them as the soft side of the trade.

Does a good closing-line Brier score mean a book is worth following? No. A good closing-line Brier score only means a book's final number was accurate, which by first pitch nearly every book achieves because the market has consolidated. It says nothing about whether the book led or merely copied sharper books late, which is why crossed markets are the better test of sharpness.

How do I actually use this to bet? Price off Kalshi and ProphetX, then look for their lines crossed against softer books like Bookmaker and Pinnacle. You can bet both sides as an arb, or simply take the soft book's side. A tool like the Arbitrage Finder automates the search so you are not scanning books by hand.

Is the dataset public? Yes. The full graded dataset behind this study, over 600 million MLB prop line movements with outcomes attached, is available on Hugging Face for anyone who wants to reproduce the results or run their own analysis.

Ready to put the hierarchy to work? Start hunting crossed markets with SmartStake and let the tool surface the crosses while you take the sharp side.