Every year during the Triple Crown, the subject of Dosage is dragged out of the closet and dusted off. No, I’m not referring to drugs. Dosage is a breeding theory that some handicappers swear by to predict the outcomes of certain races, such as the Kentucky Derby, Belmont Stakes and many maiden races.
The Dosage Theory divides race fans like no other subject in horse racing. You either subscribe to it… or you don’t. Heated and sometimes downright nasty debates rage between the believers and non-believers.
I have a confession to make: I used to be a believer.
When I initially studied pedigree theories, I explored how Dosage was used in breeding and, later, how it applied to handicapping. I can’t say that using Dosage as a handicapping tool helped my ROI any. But I figured if knowledgeable experts claimed that Dosage was the real deal and that it helped predict winners, then that was good enough for me.
So what, exactly, is Dosage? The sentence below, from, “Horse People: Thoroughbred Culture in Lexington and Newmarket,” by Rebecca Louise Cassidy, sums it up perfectly.
“Dosage theory attempts to predict ability based upon the analysis of superior male ancestors in a horse’s pedigree and is used by gamblers and breeders to forecast the likely distance over which a horse will excel.”
A Brief History of Dosage
Dosage was originally created as a breeding tool the early part of the 20th century. One of the top pedigree authorities of the day, Lt. Col. J. J. Vuillier, discovered fifteen stallions that were present in the majority of pedigrees and that the stallions passed along certain characteristics. He called these stallions the chefs-de-race or “chiefs of racing.”
He noted that every fifteen to twenty years, new chef-de-races (chefs) evolved and passed along their own set of traits. He calculated the importance of each of these sires and the position in the pedigrees of the stakes winners that he studied, applied a mathematical formula, and voilà, the rudiments of Dosage were created.
Halfway through the 20th century, Dr. Francesco Varola refined Vuillier’s system, shifting the emphasis from the chefs themselves to the attributes (speed, stamina) that these stallions passed along to their offspring. Instead of giving each chef a Dosage number, Varola classified the chefs by the traits they passed to their offspring into distance categories, known as aptitudinal groups: trans-brilliant, pure brilliant, intermediate, classic, stout solid, stout rough and professional.
In the 1970’s Dr. Steve Roman refined and combined the systems. Dr. Roman changed the aptitudinal factions to Brilliant, Intermediate, Classic, Solid and Professional. Then, he created a numbers system for these groups using a four-generation pedigree chart.
Roman assigned a greater number value to the chefs in the first generation, with descending numbers of influence to subsequent generations. Adding up these points, Dr. Roman created the Dosage Index (DI) and Center of Distribution (CD) to measure the speed and stamina in a horse’s pedigree. The more speed the horse carries, the higher the DI and CD. If you’d like to learn exactly how Dosage is calculated, check out Dr. Roman’s explanation of Dosage at www.chef-de-race.com/Dosage/review.htm
Dosage and Handicapping
The “Dosage for Handicapping” theory came about in 1981. Pedigree guru Leo Rasmussen wrote a three-part series presenting Dr. Roman’s work. Rasmussen noted that Dr. Roman’s research showed that no horse had won the Kentucky Derby with a Dosage Index of 4.00 or greater. Suddenly, everyone jumped on the bandwagon.
Using Dosage for handicapping became a big predictor of Triple Crown success. The theory seemed infallible until 1991, when Strike the Gold, with his DI of 4.20, tossed a hand grenade at the Dosage Theory.
After the race, Dr. Roman added Strike the Gold’s sire, Alydar, as a chef-de-race, which changed Strike the Gold’s Dosage to fit the “no horse over 4.00 DI” concept.
When more horses started breaking the 4.00 DI rule, the rule was changed to incorporate the higher 4.00 Dosage rate. This made sense because breeders focused on adding more speed to thoroughbred pedigrees.
Since 1991, eight Kentucky Derby winners haven’t fit the newly tweaked Dosage parameters which incorporated a 4.00 DI cut-off. Our latest Derby hero Nyquist has a DI of 7.00, the highest of any Derby winner. Do we switch the formula again to fit the facts? As a predictive tool for picking the Triple Crown races, Dosage has many flaws.
So, Dosage can’t be used to help figure out who will win the Classics. They are weird races with too many factors, anyway. But we can still use Dosage to handicap maiden races can’t we? Umm, not so fast.
Here’s the major flaw in using Dosage for handicapping: it only accounts for particular stallions (chef-de-races) in a pedigree. The female family isn’t taken into consideration at all. So, right off the bat, the system is 50 percent inaccurate. Dosage numbers don’t factor in non-chefs either. Non-chefs make up at least 40-75 percent of the modern pedigrees. If I told you that I had a system for picking winners that worked only 15-25 percent of the time, would you use it?
Let’s use our most recent 2-Year-Old Champ and Kentucky Derby hero as an example of what a pedigree chart looks like if you use Dosage to determine a horse’s distance capabilities for handicapping. Remember that the farther back in the pedigree (3+ generations), the less influence these ancestors have since they passed along whatever influence they had long ago.
I don’t know anyone who would be able to accurately predict a horse’s distance range using this chart, yet there are still people who fanatically and vigorously defend using Dosage for handicapping, despite being shown proof that the system is incredibly inaccurate.
Dr. Roman admitted that Dosage was unreliable as a handicapping tool in 2011 when he stated that “Dosage is not a breeding theory, nor a handicapping system, or betting scheme for the Kentucky Derby.”
He further defined Dosage as, “A methodology applied to large populations of Thoroughbreds for classifying pedigrees by aptitudinal type, and as a research tool correlating type with real world performance.”
When is a Chef not a Chef?
Fast forward to 2016. Addressing critics, Dr. Roman admits that one of the issues with Dosage is that it ignores non-chef-de-race stallions. To rectify that, Dr. Roman created a list of eighty-seven non-chefs that are grouped into either speed or stamina categories, as determined by the average winning distance of their major stakes winners. The list is intended to help “fill in the gaps” and be used to adjust the Dosage numbers for individual horses.
These eighty-seven stallions, fifty-five in the speed section and thirty-two on the stamina side, haven’t been classified as chefs because, according to Dr. Roman, the sires, “have not been identified as attitudinally prepotent.”
In other words, the sires don’t fit into one or more of the specific factions of Brilliant, Intermediate, Classic, Solid or Professional, so they’re lumped into broad speed or stamina categories.
So, now, the scientific Dosage method of focusing on specified stallions noted for passing along certain attributes has been changed once again. Now, non-chefs that might contribute speed or stamina are added to shore up the theory.
Using the progeny’s average winning distance to determine where to categorize the non-chefs is also sketchy, at least for the stallions based in America. Races at the classic distances of 1 ¼ miles to 1 ½ miles make up less than one percent of the events in the U.S., so naturally, the average winning distance is shorter. Many offspring of stallions with stamina-oriented pedigrees never get an opportunity to race at 1 ¼ miles.
What’s a Handicapper to Do?
Dosage, when used to plot the changing trends of speed and stamina over the course of years, is very helpful; using it to determine the specific attributes of individual horses is not. Recognizing the value imparted by chef-de-race stallions in a single pedigree is useful. Blindly tossing the rest of the pedigree, minus the “non-chefs” that might contribute attributes, is not useful and leads to erroneous and misleading assumptions about the horse’s distance abilities.
When it comes to the Dosage Theory in handicapping, be mindful of the chefs and where they are on the pedigree chart, but give more credence to the entire first two generations in your pedigree handicapping and toss the Dosage numbers.
The views and opinions expressed herein are those of the author and do not necessarily state or reflect those of US Racing.