The Kentucky Derby: Grading the Road Ahead
Since our inception in 2007, HorseRaceInsider always has encouraged readers to engage honest dialogue with HRI authors and with each other. More than just a great opportunity to vent, it’s humbling to learn that this game boasts many passionate experts, not just those fortunate enough to derive a living from Thoroughbred racing.
From time to time, a reader will write a position paper so interesting that we are compelled to share it with our loyal followers. We also hope that this is a way for our fans, as well as paid journalists, to give back to the sport they love. After all, the game’s practitioners need all the help they can get.
This recent contribution written by well recognized racing websites contributor Indulto is a studied approach to a more equitable Kentucky Derby qualification system, a methodology that is much more studied than the present graded earnings model. Hopefully it will help provide a method to insure that only the most worthy 3-year-olds punch a ticket into the Churchill Downs starting gate on May’s first Saturday.
The piece is lengthy and its weighty material. We think you’ll agree that it’s worth your time and effort. Part 2 tomorrow, Friday.
When the 2010 Belmont winner, Drosselmeyer, powered home to win the 2011 Breeders' Cup Classic at Churchill Downs following a second-place finish in the Jockey Club Gold Cup (also at 1-1/4 miles), it further emphasized the continued failure of the Kentucky Derby's eligibility system (and 20-horse fields) to produce a Triple Crown winner that many believe would revive the sport's popularity.
Based on graded stakes earnings, that system denied Drosselmeyer a start in the Derby despite his third-place finish in one of the major late Derby preps; demonstrating form that improved with increasing distance. Of course there's no guarantee he would have won all three events, but it's clear that earlier recognition of contenders with his undeniable stamina and competitiveness is required.
Drosselmeyer's victory also frustrated the BC's Win-And-You're-In (WAYI) policy which some see as a less effective stimulus for subsequent success than simply participating in traditional recent BC preps. In 2011 only 2 of the 6 WAYI race winners made it to the Classic, whereas all starters came out of preps during September and October. A list of BC WAYI races can be found at Breederscup.com.
Proponents of these systems contend that they more accurately - and, generally, more fairly -- reflect the relative achievements, competition level, consistency, and potential for further success, among the ranked entities.
Eligibility based on earnings has made virtual WAYI events for the Kentucky Derby out of the BC Juvenile races, the Delta Jackpot, the Sunland Derby, and the UAE Derby. This is also true of the more traditional preps at 9 furlongs within 6 weeks of the Derby, whose starters have produced multiple winners of the main event.
One could argue that preferring participants from multiple graded stakes over a mile during March and April including one of the final seven 9-furlong preps is a better strategy. The preferred entrants would be ranked according to their Win, Place, and/or Show (W-P-S) finishes in graded stakes for three-year-olds over a mile.
The Breeders' Cup employs a ranking system that assigns 10-6-4, 6-4-2, and 4-2-1 points, respectively, to W-P-S finishes in G1, G2, and G3 races. Note that the place and show points at each grade level match the win and place points at the next lower level. When pre-entered fields are oversubscribed, preference is given to winners of specific WAYI races, but the next seven available slots go to the highest-ranked among the remainder. Further details regarding their application by the BC are available at Breederscup.com.
The Triple Crown Bonus has not been offered in recent years, but Magna introduced a bonus for the 2011 Preakness winner who had competed in Derby preps at other Magna tracks. I believe we'll see more of this in the future since it fuels interest for customers as well as horsemen. Perhaps a futures pool for every bonus offered would not only extend, but also enhance enthusiasm for the culminating event.
The now defunct ACRS bonus awarded 10-7-5-3-1 points for 1st through 5th place finishes, respectively. The point leader at mid-series earned a bonus of $150,000 and the final point leader earned $550,000. The runner-up earned $225,000 and 3rd place was worth $125,000. The effective final distribution ratio was 21-11-5.
I find it interesting to compare both bonus points and awards with the most common graded stakes purse distribution percentages of 60-20-10-5-3 used in NY, CA, and FL as well as the 62-20-10-5-3 split in KY and the 54-18-10-5-3 ratio for BC races. The Grade I Hollywood Futurity is an example of an uncommon distribution with a $375-140-100-60-45-30K split of its $750K purse thereby yielding only 50% to the winner.
The discontinued Triple Crown bonus awarded 10-5-3-1 points for 1st through 4th place finishes. The point leader would receive $5,000,000 for sweeping all three legs or else $1,000,000 with points earned in each leg. In 1987, Belmont Winner, Bet Twice, who finished second in both prior legs, earned the bonus over Derby and Preakness winner, Alysheba, who earned no points in the Belmont.
An important source of innovative thinking regarding point systems is Greg Brandow at thoroughbredreport.com. For several years he has been ranking "horses of the month" by assigning 10-8-6, 7-5-3, and 4-2-1 points to W-P-S finishes, respectively, for G1, G2, and G3 races. Unlike the above BC system, finishing 2nd is also worth more than finishing 1st at the next lower grade level. That each value is unique permits easy identification of the top associated grade/placing. We'll refer to this system as PTS-TR, and to the BC system as PTS-BC.
In his blog piece, "A Graded Stakes Points System," Mr. Brandow offered a 3-stage methodology for creating systems to rate graded stakes races: 1) Assign a basic rating to each race according to grade, age, sex, and distance 2) Adjust the basic value for each race according to purse level and actual field size. 3) Distribute the resultant value for each race to W-P-S finishers according to fixed percentages
The system he implemented using that methodology is worth reviewing; in particular, his tables at the link which display: A) Basic point values for all categories of graded stakes on dirt and turf B) Adjustment percentages for all defined factor categories C) Top-three finish distribution percentages
For the purpose of this discussion, examples will be confined to open route races on dirt. Mr. Brandow's system assigns 100-65-45 points, respectively, to G1-G2-G3 races on dirt for three-year-olds and up, and 90-55-35 when restricted to three-year-olds.
Considering adjustments at their boundary values, ratings for fields with fewer than 5 starters would be reduced by 20% while those with 12 or more would be increased by 10%. Ratings for races with purses of $2M or more would be increased by 50% while those with less than $500K would not be adjusted.
Adjusted ratings would be distributed using percentages of 60-25-15 to W-P-S, respectively, which approximate the purse distributions mentioned earlier. For 2011, this system assigned the most points to the campaign of Acclamation who did not compete in the Breeders' Cup, but was a finalist for Horse of the Year honors.
Perhaps an adjustment for the margin of loss to the winner and/or the preceding finisher - in place of the adjustment for purse value -- would provide more independence from earnings if such a system were applied to determine eligibility for TC and BC races.
"A New Pony Express" by Troy Record columnist, Nick Kling, helped put point systems in proper perspective for me:
"Points are much more equitable than raw earnings or wins. Points reward consistency and activity."
It also addressed the desire of many fans to establish alternative measurements for objectively comparing the accomplishments of horses at all levels of competition:
"Don't stop following three-year-olds after the Belmont Stakes. If there was a points series tying preps, Triple Crown races, summer events like the Swaps and Travers, then leading into fall races against older horses, fans would stay involved all year long."
"Don't stop with stakes-caliber horses. Why shouldn't there be similar series for horses competing at various claiming or allowance levels?"
"Imagine if there had been something like this in place in 2011. How many points would Rapid Redux have earned versus one-hit wonders like Animal Kingdom?"
Extending Mr. Brandow's system(s) to include non-graded stakes, as well as some subset of allowance races, does not appear overly difficult. However, I suspect the remainder will pose a major challenge considering that slots-fueled purses and the proliferation of optional claiming races have skewed the traditional relationships among non-stakes categories, which in turn may have compromised speed ratings as well.
Class pars on which commercial speed figures are based could be an appropriate starting point for establishing non-graded stakes race categories, rankings, and relative ratings -- assuming they have been adjusted by their creators to accommodate said altered relationships. Hopefully, a respected product vendor will make such proprietary data available for this purpose.
If a suitable point system reflecting all competition levels could be implemented, the next step would be to include total points accumulated as of each race on past performance lines. That would save a lot of time for those of us whose handicapping relies less on how "fast" a horse ran than on how consistently well it ran at which levels.
Mr. Kling's quoted acquaintance will be happy to learn that the throughbredreport.com website also maintains an on-line data base containing both the scheduling and results of all U.S. graded stakes with which it tracks point leaders among horses, jockeys, and trainers competing at those levels. It's a good place to continue tracking three-year-olds after the Belmont Stakes until a universally accepted system is in place. It also makes the task less formidable of accumulating total points earned by W-P-S finishers in graded stakes over various periods.
Further, this data base contains readily accessible dead heat finish data. Dead heats undervalue the achievements of those involved when eligibility is based on earnings. Assuming that tied finishers split the awards for the current and next lower placing(s), two dead-heating winners earn approximately a third less while point systems credit both in full.
I'd like to see the contents expanded to include 1) graded stakes in Canada and the U.A.E. where U.S.-based horses frequently compete, and 2) actual cumulative earnings for each horse tracked. Accurate earnings-based rankings cannot be generated from W-P-S earnings alone, and the data base currently maintains entries only for horses finishing in the top three (or four in a dead heat for show). Currently, only race winners are available prior to 2009.
Actual cumulative earnings figures for the upcoming Derby are available at several websites including Linemakers.net.http://www.linemakers.net/. Of the top 20 eligible and 4 also-eligible horses on that list at the time this was published, 9 were fillies who are more likely to run in the Oaks than the Derby. As the past 3 Horse-of-the-Year winners were female, however, perhaps the next Triple Crown winner will be also.
Such sources of ongoing pre-Derby eligibility rankings should be encouraged to present a side-by-side comparison with points-based rankings. I propose using PTS-BC point totals with the top single-race PTS-TR value serving as primary tie-breaker, and total graded stake W-P-S finishes as secondary.
Tomorrow: Putting It all Together
Do you have a take on the state of the game, a favorite race horse, trainer or jockey? Share your ideas. Show the racing industry and media the folly of their ways. The HRI Readers Blog: "When you lose it, Use it." Submit your blog to HRI Webmaster.