Why are post position stats misleading, and how can Impact Values help?

Post position statistics can be among most confusing metrics you’ll find in the sport of horse racing.

At first glance, they seem simple. Let’s saying you’re reviewing stats for a given racetrack, and you discover post 1 wins 15% of the time over a particular distance. Meanwhile, post 10 wins at a 10% rate. Surely post 1 is more favorable than post 10?

Not necessarily. With rare exceptions, post 1 is always occupied, which means the win percentage for post 1 benefits from races with small fields. If only five horses start in a race, post 1 has a 20% statistical chance at producing the winner.

In contrast, post 10 only joins the fray when 10 or more horses are entered. This means random chance never provides post 10 with more than a 10% chance at producing a winner, so its lower average win percentage can be forgiven.

How can we put post position stats to use when they’re prone to being misleading? Brisnet Ultimate Past Performances can help. A wealth of data under the “Track Bias Stats” section explores post position stats in granular detail, providing a more insightful view of post position stats.

Specifically, we’re interested in the “Post Bias” section of Brisnet Ultimate Past Performances, as illustrated below. Stats can be found applying to specific race distances from the current race meet (which may cover months of data) and from the previous week of racing.

The Post Bias stats are broken into four categories. From left to right, they are:

• RAIL: Stats pertaining to the 1 (rail) post position.
• 1-3: Stats pertaining to post positions 1, 2, and 3.
• 4-7: Stats pertaining to post positions 4, 5, 6, and 7.
• 8+: Stats pertaining to post positions 8, 9, 10, and beyond.

Underneath the categories, you’ll find two rows of stats. They are:

• Impact Values: A number indicating how each category performs relative to expectations. A value of 1.00 is average, with numbers above 1.00 reflecting better than average performance and numbers below 1.00 reflecting worse than average performance. If pure math suggests the rail post should win 20% of races, but it’s only producing 10% winners, the Impact Value will be 0.50, or 50% lower than the 1.00 average.
• Avg Win %: Indicates the average win percentage for the post(s) in each category. If the 1-3 category boasts a win percentage of 20%, then the combined average win percentage of post 1, post 2, and post 3 is 20%.

The average win percentage stats can be useful, but the impact value stats are more valuable. Since they compare the actual performance of each post position category with the expected results for the category, they recognize how post 10 is expected to produce a lower win percentage than post 1 and avoid unfairly penalizing post 10.

The next time you’re handicapping a race and grow curious about post position statistics, be sure to give the data in Brisnet Ultimate Past Performances a close examination. The impact values can make all the difference in finding positive angles.