Does this mean we should jump on the favorite whenever we encounter a small field? We decided to test this idea by checking the results of 3102 races run at Santa Anita, Hollywood Park, Gulfstream, Fair Grounds, Aqueduct, Belmont and Churchill Downs between January and June, 2004. We are currently in the process of a far more extensive study of horses over a longer period and many more tracks, but this data is still interesting.

Overall, 1123 of the 3102 favorites won, or 36.2%. The average payoff was $ 4.78, giving a Return On Investment or ROI equal to 0.87. This means that for every dollar you invested on these favorites, you would get back only 87 cents, representing a loss of 13% of your money.

For example, consider the races in which only 4 horses were fielded. All things being equal, each horse would have one chance in four of winning, a 25% win expectation. In our study, the favorites actually won 53.1% of the races. The Impact is the ratio of the Actual% to the Expected%. In other words, the favorites won more than 2 times as often as expected on the basis of pure chance.

The average mutual payoff was $ 3.60. Since the favorites won 53.1% of 32 races, there were 17 winners. Since each paid $3.60, the total amount returned was $ 61.20. However, if we bet $ 2 on each of the 32 races, we put up $ 64. Since we lost $ 3.80, our ROI was 61.2/64 or 0.956.

So, in spite of winning more than half of the races with a field of 4 horses, the average mutuel is so low that we still wind up losing money. We found that the average mutuel increases as the number of horses in the field increases. However, the actual win percentage of the favorites decreases faster, making the ROI decrease with the number of horses in the field. For example, in the races having a 12 horse field, the average payoff jumped up to $ 5.75. However, the actual win percentage dropped to 29.2%, leaving us with a ROI equal to 0.84, a loss of 16 cents on the dollar! This was in spite of the fact that the Impact value tells us that the favorites won 3.5 times as often as would be suggested by pure chance.

Still, the ROI shows a fairly clear tendency to increase as the field size decreases. The implication is that we will lose money if we bet on the favorite just because it is the favorite. However, if we can add some handicapping capability and learn to determine when the public is wrong and select the true favorite, we do appear to have a better opportunity in smaller fields.

We did find one potential area for showing a profit. When the favorite went off at more than even money in a field of only 4 or 5 horses, the ROI was greater than 1.03. Even this small profit of 3 cents on the dollar is questionable because it was based on only 93 of the 3102 races.

We looked at one more variation, that of favorites which also had at least 2 more speed points than any other horse in the race. (See our article "Two Faces of Speed" for an explanation of speed points). Overall, 197 of the 480 favorites which also had a 2 speed point advantage won (42%) paying an average of $ 4.65 for an ROI of 0.95. We're still losing 5 cents on the dollar, but that is quite a bit less than the track take. However, the ROI seemed to increase as you move from a Field of 8 in either direction.

In particular, for fields of 10, 11, or 12 horses we found 41 winners out of 101 (40.6%) with an average payout of $ 5.31 for an ROI of 1.08. This is a bit more interesting, and appears to be at odds with the results shown earlier. Why do our results show an improving ROI in larger fields?

Perhaps this is simply a statistical anomaly caused by the small sample size of 101 races. On the other hand, it may well be that horses with a clear speed advantage do better than the others in crowded fields. It may well be worth the effort to track this potential relation between favorites with a speed advantage and field size until the sample size is clearly adequate. We expect to clear this up in our current, on-going study.In any case, it appears fairly clear that unless you can incorporate some additional (handicapping) factors into making betting selections, the size of the field, taken alone, does not justify betting on the favorite.