Should you bet favorites or underdogs?

On MLB moneylines both sides came back close to fair, and the sample is not large enough to say which side is dearer. Favorites returned about -0.8% and underdogs about -3.1%. Neither figure separates from zero: the standard error is about 2.5 points on the favorite side and 3.6 on the underdog side. The split points in the direction the research literature expects, but on 2,222 graded sides it is a direction, not a size.

Written by Jesse, NegativeEV. Last updated 12 August 2026.

Why do underdogs cost more?

The mechanism has a name: the longshot bias, one of the most consistent findings in betting-market research. Bettors systematically overpay for long odds, so books can shade underdog prices thin while favorite prices stay honest. The MLB moneyline sample is too small to show it on its own, but larger censuses do. Across 36,345 completed ATP matches the favorite side returned -4.9% and the underdog side -12.0%, both far outside their noise floors, and the sign held in all 19 seasons. On the MLB prop board the same shape holds: anytime home run props, mostly +200 and longer, realized about -23% across 18,468 graded props. The full grading behind that prop figure is published as 18,468 home run props, graded.

Does that make favorites automatically good bets?

No. A fair average still contains mispriced games in both directions. A heavy favorite at a steep price is a bad bet whenever it wins less often than the price implies, and some underdogs are genuinely underpriced. Which side of the split tonight's game sits on is a matchup question - the starters, the lineups, the park, the weather - not a rule of thumb.

What does this page not establish?

Which of the two sides is dearer on an MLB moneyline.

Both averages sit inside their own noise floors, so the split between them is a direction and not a size. The figures that do clear their noise floors come from other markets, tennis match winners and baseball home run props, and those say the longshot bias is real where it was measured, not that it costs any particular amount on a baseball moneyline. Every price here is the best one posted across books, so a bettor who cannot reach the book holding that number is not looking at the same figure. Nothing here reads a single game either: the averages describe what two whole sides of a market charged, not which way tonight's price is wrong.

How can a moneyline be checked?

Compare the posted price to the true win probability from a simulation of the actual game. NegativeEV simulates each game thousands of times, play by play, with the real starters and lineups, and reports the win probability and expected value for either side, free at negativeev.com, with no account.

More on the number: https://negativeev.com/about/expected-value The measurement behind this page: https://negativeev.com/research/mlb-moneylines