One of the biggest debates among sports bettors is whether betting markets are truly efficient.
If sportsbooks set accurate odds, then long-term results should closely match the implied probability of the betting line.
To test this idea, we analyzed thousands of MLB games since 2004 and compared favorite win rates across several odds ranges.
Favorites -150 Or Higher
SU: 13,945–7,823
Win Rate: 64.1%
ROI: -0.9%
Heavy favorites win roughly 64% of the time, which is close to what their implied probabilities suggest.
Favorites -120 To -150
SU: 12,747–9,973
Win Rate: 56.1%
ROI: -1.7%
Mid-range favorites win about 56% of games, again producing results close to sportsbook expectations.
Favorites -110 To -120
SU: 5,925–5,680
Win Rate: 51.1%
ROI: -4.5%
Near pick’em games become much more volatile, and the edge from the sportsbook’s vig becomes more visible.
Pick’em Games
SU: 1,630–1,654
Win Rate: 49.6%
ROI: -0.7%
As expected, pick’em games produce results close to a true 50/50 split.
What The Data Suggests
Across thousands of games, win percentages across different odds ranges track closely with expected probabilities.
Key observations:
• Heavy favorites win roughly 64% of games
• Mid-range favorites win around 56%
• Pick’em games are essentially coin flips
These results suggest that the MLB betting market is highly efficient over long sample sizes.
