Abstract
We examine the efficiency of the Major League Baseball over-under betting market. Previous research in this area was unable to correctly test market efficiency because a data-set including the odds at which wagers were placed was unavailable. Using a data-set that includes both run lines and money lines, we find little evidence that the over-under market is inefficient.
Acknowledgement
The authors thank John Gandar for his comments on a draft of this manuscript.
Notes
1 Computer Sports World stopped providing the odds archive after 2007.
2 The variance of the binomial distribution is . The normal approximation to the binomial is very good when this variance is greater than 10. Thus, if , then n should be greater than 40.