The Skinny on Overweights

When I used to go to the races as a kid, I followed the same pattern. I’d get a program to supplement my tattered and marked-up edition of the Daily Racing Form, find a comfortable table and chair in the gazebo and accept the offer of an ice-cold Pepsi from my stepdad, Dennis.

Then, sipping on our overpriced and oversweetened beverages, swapping various sections of the Form like we were assembling a great puzzle (which, in a sense, we were), both Dennis and I would await track announcer Gary Henson to make his presence known.

“Welcome to Longacres,” Henson would drawl in that distinctive, dulcet voice of his. “Here are today’s overweights and changes.”

Like a dutiful servant, I would note the information in my program. Mostly I cared about the scratches, but I always wrote down the overweights as well.

“In the sixth race, number two, Scamparina, is three pounds over… no changes in the seventh race… in the eighth race, number five, My Satire, has been withdrawn…”

And so it would go until my program was filled with all the latest updates.

At that point, smiling smugly at my efficiency and attention to detail, I would proceed to ignore all but the scratches and go about my day — not once, in 30+ years of playing the races, have I ever altered my selections or list of contenders based on a late change in weights.

Recently, I’ve begun to wonder whether my dismissive attitude regarding this factor was/is wise, so I decided to find out.

Using my new, handy-dandy results chart database program, I examined 889 races — I didn’t have the stamina for 890 — from various tracks across the country. My query was simple: Do horses that are carrying more than their assigned weight, i.e. overweights, win as often as they should?

Now, before we go any further, let me explain the phrase “as often as they should.”  Simply put, this means that horses showing a specific characteristic — in this case, a greater race weight than assigned weight — win their rightful share of races.

In his masterful work “Winning At The Races,” Dr. William Quirin introduced the concept of an “impact value,” which he calculated by “dividing the percentage of winners with a given characteristic by the percentage of starters with that characteristic.”

“An IV of 1.00 means that horses with a specific characteristic have won no more and no less than their fair share of races,” Dr. Quirin explained.

Similarly, an IV of greater than 1.00 denotes that a particular factor is producing more than its fair share of winners, while an IV under 1.00 means that it is producing less.

In addition to Dr. Quirin’s original impact value, I will also offer an odds-based impact value when I present my findings later in this piece. The OBIV is based, not on field size, but on the average odds of the horses meeting the criteria of the study. The advantage of such an approach is that it better isolates the factor being tested by equalizing the winning chances of the horses that show it.

For example, if one were to test what effect the letter “a” in a race favorite’s name had on its performance, it is abundantly clear to those who understand statistics that the impact value would be high… not because the horse’s name contained the letter “a,” mind you, but because the horse was favored.

In fact, when I tested just such a proposition (race favorites with the letter “a” in their name) using my database, 579 horses qualified and the impact value was a staggering 2.78. Yet, the OBIV was only 0.81, which is right in line with expectations for race favorites as a whole (see graphic below).

(Click on image to enlarge)

So, with all of that out of the way, let’s get on to the fun stuff — the study itself.  The first thing I’ll post is how ALL of the horses in my survey of 889 races performed, just so we have some baseline statistics:

Horses – 6,790

Winners – 889

Rate – 13.1%

IV – 1.00

OBIV – 0.81

And now — drum roll, please — we’ll see what happens when a horse is asked to carry more than its assigned weight:

Horses – 1,010

Winners – 109

Rate – 10.8%

IV – 0.85

OBIV – 0.73

The results are stunning — at least to me — as they appear to confirm that additional, unassigned weight is, in fact, detrimental to a horse’s performance. Not only is the IV less than 1.00, but the OBIV is also below the 0.80 “fairness” threshold.

I guess I should have paid more attention to Gary Henson.

Weekend Handicapping Reports

SATURDAY (8/4)

2012 West Virginia Derby (MNR)
2012 Prioress Stakes (SAR)
2012 Whitney Stakes (SAR)
Fair Meadows Tulsa Win Factor Report
The Meadowlands Win Factor Report