Actuarial Review Return to Main Page

Nonactuarial Pursuits of Casualty Actuaries

Horse Sense

by Marty Adler

Colt by Cherokee Run (sire) and Arctic Assay (mare).

The challenge of beating the odds must appeal to many with a mathematical bent, casualty actuaries included, of course. And one obvious place to face this challenge is at the racetrack. The late Lew Roberts, for whom I worked early in my career, claimed that he had developed a system that would beat the odds, but only to the extent of neutralizing the house advantage. So he turned his active mind to other pursuits.

But one of our Fellows has progressed beyond the handicapping of mere horse races and breeds thoroughbreds, though on a much smaller scale than the giant Kentucky stables in bluegrass country. Both activities require statistical analysis; both have a wealth of data available. Handicapping horse races is the "short-tail" aspect; you know the results very soon. Breeding is the "longer tail" end of the game. When a breeder buys a mare and selects a stallion to breed to her it will take far longer to learn if the breeder made a good choice.

Our Fellow has followed racing since age 12, when his uncle took him to the races at Saratoga. His first interest was in handicapping, a combination of art and science. Among the many characteristics that can affect the outcome of a race are age and gender, distance of the race, and track condition—all subject to objective analysis. Professional handicappers constantly study their statistics, looking for something that the others haven't yet found—that hidden betting factor that can yield upset payoffs. The more rigorous handicappers will compute speed figures and track variants, attempting to rate each horse's prior efforts to help determine who might win today. Thus handicappers are using prior results to predict future performance—sounds like a pursuit that's not all that foreign to actuaries.

Our Fellow ventured into breeding in the mid-1980's, when he bought a small share of a thoroughbred broodmare. In exchange for paying the stud fee for the stallion, he received a half interest in any resulting foal. That mating worked out well. They sold the foal, a beautiful chestnut filly, at one of the many public thoroughbred auctions.

Filly by Gold Fever (sire) and Storm Struck (mare).Like the giant Kentucky breeders, he must consider carefully the matings for each of his mares. He looks for pedigree matches that will increase the probability that the resulting foal will be a successful racehorse. There are many theories about breeding: inbreeding, crossbreeding, and the like. A wealth of statistics is also available for the conscientious breeder to study.

One example of the analysis involved is found in a comprehensive report by an Australian breeder who studied the success of horses who were in-bred to certain successful female horses. That analysis concludes that such horses have a significant statistical edge over their competition. While relatively few horses have this breeding characteristic, the racing results of these horses, as a group, shows about a 30 percent advantage. Following these principles, this spring he bred a mare and a stallion that share a common female ancestor. The mare has Secretariat in her pedigree; and the stallion, Sir Gaylord. Secretariat and Sir Gaylord are half brothers, that is they share the same dam, the mare Somethingroyal. The resulting foal is expected to be born early next spring, and our actuary will closely follow its development over the years to come.

He does not intend to rely exclusively on the work of others. There are various bloodstock statistical services that provide data for breeders to use to test their own theories. He has a few such theories supported by anecdotal evidence. Once he finds the time, he hopes to put these theories to a more analytical test.

He spends about 3-5 hours a week studying the pedigrees and race records of the thousands of stallions. He wants to discover the new stallion that will become the next successful producer, for once a stallion proves himself, the stud fees are increased beyond a level that he can afford. In years past this approach was successful when he got Cherokee Run and Storm Cat well below their current stud fees. Both have produced major winners in North America and in Europe. Some of his other choices were not so good. These experiences are helping him develop a profile of the type of racehorse that has a greater-than-average chance of becoming a successful stallion.

Currently he owns or shares in six broodmares, four of whom are in foal (pregnant), one two-year-old who is in training, one yearling, and three weanlings (horses born earlier this year). He plans to sell the weanlings in yearling sales next year, and one of the expected foals at a weanling sale next November. For the remainder of this year he will study the pedigrees of the six mares, trying to decide which stallion to breed to each one. He hopes to have breeding contracts finalized in December.

Generally, the mare's owner pays the stud fee and assumes all the risk that the foal will not be good. Once the foal is born there is no refund, or adjustment on the stud fee, if the resulting foal proves to be inferior. For example, if the foal suffers from any of a number of defects that may preclude it from withstanding the required physical training—a not uncommon situation; approximately 30 percent of foals born never race—auction prices will not recover a mare owner's costs. He purchased a mare last year and bred her to the Kentucky Derby winner, Charismatic. Not wanting to risk the high stud fee, he negotiated a "foal-sharing" arrangement with the owners of that stallion. The horses were bred earlier this spring and the contractual arrangement calls for the foal to be sold at public auction in November 2002 (a few months after the foal will have been weaned) with the proceeds of that sale to be split 50/50 between the mare owner and the stallion owner. He retains the right to pay the advertised fee at any time prior to this sale in order to retain full ownership of the foal. He may elect that option if the foal looks spectacular, if the market for auction yearlings is strong, and if he has the money. That's a lot of ifs. If, on the other hand, the foal does not look promising, he has significantly hedged his downside risk by sharing it with the stallion owner.

Russ Fisher likes to think that his experience evaluating risk in our profession has helped him assess the range of risk he is taking on each horse transaction, helping him create a variety of risk-sharing strategies. He considers himself fairly lucky so far, but is not ready to give up his day job, designing and directing an e-commerce strategy for the worldwide treaty business of GeneralCologne Re.