Keeneland Book 1: The Kids are All Right….

Posted on September 8, 2011 by Rob Bingel

The catalog for Book 1 is okay!

Yes, the prior sentence deserved an exclamation point. I don’t know how the yearlings will look, nor how they will scope, or even what buyers were present. What I do know is that this catalog compares very favorably with Book 1 from last year.

My model uses the following variables: current stud fees, a dam scoring system (racetrack performance of the dam, the dam’s offspring, the second dam, and the second dam’s offspring), colts vs. fillies, birthdays, and first-time sires.

My modeled scores for prices basically showed the sale to be flat with last year. This may not seem like a glowing endorsement of the 2011 sale, but please remember that stud fees have fallen from 2010 to 2011. Stud fees have consistently been the most dominant variable in modeling past sales prices. Keeping modeled prices flat year-to-year in spite of market-wide falling stud fees is thus impressive. In fact, median stud fee has actually increased by $13,000 per yearling in 2011.

The dam-side pedigree is the same as last year on average, but slightly lower on a median basis.

Coupled with the fact that the S&P 500 is actually higher than it was a year ago, Keeneland Book 1 should be all right…

Posted in Uncategorized | Tagged , , | Leave a comment

Comparing the 2010 and 2011 Fasig-Tipton Saratoga Catalogs

Posted on July 27, 2011 by Rob Bingel

What factors are you using to compare the catalogs?

Current stud fee, my Bingel scoring system for the yearling’s dam side pedigree, and a couple of minor variables in the catalog.

Why current stud fee?

Current stud fee has consistently been the best variable in auction catalogs for predicting sales price.  I have run many regressions on different racehorse sales (thoroughbreds and standardbreds), and current stud fee always pops up as the best predictive variable in a catalog.  Obviously, physicals and vetting aren’t known until the sale occurs and can’t be included in any pre-auction catalog analysis.  Current stud fee matters to buyers, not the stud fee when the yearling was bred, because current stud fee reflects the market’s attitude about a particular sire.

How do the 2011 stud fees compare with the stud fees in the 2010 catalog?

The average stud fee in the 2011 catalog is down 13% when compared with the average stud fee in the 2010 catalog.  The median stud fee is actually flat at $50,000.

How will this affect prices at the auction?

All else being equal, buyers have an idea of what multiple of stud fee that they are willing to pay.  Just because the stallion masters have lowered their stud fees doesn’t mean that buyers are willing to pay the same amount of money, with the benefit going to the breeders rather than the stallion masters.  Thus, all else being equal, average sales prices will face downward pressure from the lower stud fees, although the breeders may be less willing to sell.

How about the dam side of the pedigree?

The average Bingel Score evaluating racetrack performance for two generations on the dam side of each yearling’s pedigree shows the average quality has increased by 5% while the median score has decreased by 5%.

Combining the stud fee impact and the Bingel Score, how do the catalogs compare in total?

There are a few additional minor variables including birthdates (January = good, May = bad) and a premium paid for colts versus fillies, at least at the August Sale.  When everything was included, the average modelled sales price was down 11% while the median sales price was down 13%.

How can the modelled score be lower???

There are mix issues.  Lower median and average modelled prices are, in my opinion, due to how the best sires and dams are matched with each other. Stud fee and dam score are remarkably independent at times, which seems counterintuitive.

In 2010, there were more yearlings with stud fees over $100,000.  Thus, the average stud fee is lower while the median stud fee is flat.

Are you predicting lower average and median prices, then?

No.  Other variables cannot possibly be measured at this time, such as physicals and vetting.  The buyers present can really change the outcome of an auction, especially when such wealth and egos are involved.  The economy is actually better year over year, if you can believe it and pending the debt ceiling deal.  Finally, the catalog is smaller this year, which may or may not lead to a bump in prices.  Given the long haul to Saratoga, I’ve got to think that it would be hard for any wealthy owner to make it to the sale without purchasing anything.

We shall see…

 

 

Posted in Uncategorized | Tagged , , , , , , | Leave a comment

FT post-analysis

Posted on July 12, 2011 by Rob Bingel

Median price was up 20%. My catalog-based model showed the prices to be flat, so that meant that the increase was due to better physicals, the smaller catalog, better buyers present, or good macroeconomic conditions. The point is, the increase was NOT due to differences in the catalog itself. Real value was delivered by the FT team.

The other good news is that the market appears to be correcting by adjusting the supply.

Posted in Uncategorized | Tagged | Leave a comment

Graded Stakes Winners and the Bingel Score

Posted on June 7, 2011 by Rob Bingel

The Bingel Score methodology was discussed in the previous post. To recap, the Bingel Score captures the quality of the dam for each yearling sold at a Thoroughbred auction. The natural question is: how do we know that the methodology is any good?

First, a brief note on how the methodology was developed. Scores were developed for the racetrack performance of the dam, her other offspring, the second dam, and the second dam’s other offspring. The points awarded for each performance were adjusted via regression analysis to provide the best possible fit for the market’s preferences as a whole.

How did the model determined in the paragraph above fare? Obtaining information from the Jockey Club is cost prohibitive, but The Thoroughbred Times (11 SEP ’10) published a list of graded stakes winners purchased at the Keeneland September Yearling Sale from 2006 to 2009.  The name of the winners could be traced back to the catalog pages found in the Keeneland archives.

The value of the graded stakes winners data was twofold:  first, the importance of the dam pedigree as quantified by the Bingel Score became apparent when comparing the graded stakes winners with past general population entries in yearling auctions.  Second, a relationship between number of yearling foaled and minimum Bingel Score needed was noted.

Approximately eighty percent of the graded stakes winners satisfied the following conditions for minimum Bingel Score as a function of the number of offspring foaled from the dam:

Bingel Score Criteria, Based on Graded Stakes Winners*
# Offspring Criteria % Higher # in Group
1 0.5+ 81% > 31
2 0.75+ 80%> 40
3 0.75+ 82%> 28
4 1.25+ 78%> 27
5,6 1.5+ 86%> 38
7+ 2+ 79%> 42
*Graded Stakes Winners purchased at the Keeneland September Yearling Sale
from 2006-2009, as reported in Thoroughbred Times Today, 11 September 2010.

Why 80%?  For a normal probability distribution, 80% is equal to approximately everything at one standard deviation below average all the way up to the higher echelons of the Bingel Score.  Why the dependence on number of yearling foaled?  With each foal, the performance of its peers becomes more important relative to the performance (in the past) of its dam, second dam, and second dam’s other offspring.

A couple of final thoughts.  First, the higher the Bingel Score does not mean the higher the chances of the dam having a graded stakes winner with the yearling entered into the auction.  Other factors like conformation, sire, vetting, and so on come into play.  The Bingel Score is merely a way to grade the dam side of the pedigree.  Second, there is a chance that a yearling with a Bingel Score not at the clip levels mentioned above becomes a graded stakes winner after all.  However, the odds become worse with Bingel Scores outside the levels noted in the chart above.

Please contact me with any questions at bingel@equinistics.com.  Future articles will look at the upcoming Fasig Tipton and Keeneland auctions.

Posted in Archive | 1 Comment

Scoring the Dam

Posted on May 31, 2011 by Rob Bingel

 

Yearling sales are sexist.

Everybody knows the value of a yearling’s sire based on his stud fee. For new sires, stud fee reflects performance on the racetrack. For established sires, stud fee reflects top offspring produced, percentage of stakes winners, both, or another factor. In any case, the market expresses its opinion of a particular sire at the auctions, and the sire’s stud fee is adjusted thereafter accordingly.

No such grading system exists for the dam. Buyers may look at the dam’s racetrack performance, her offspring, or even the lineage from which the dam came. The problem is that all “grading” is qualitative, not quantitative, and therefore completely subjective.

The Bingel Score was developed to solve the qualitative dilemma by providing quantitative grading of the dams. (“Bingel Score” proved much easier to type and understand than “Dam Side Score”, the original moniker. No ego inflation intended.) What goes into the scoring system? Lots of variables, most of which are not surprisingly related to racetrack performance.

The best, not all, of the racetrack performances of the dam, the dam’s other offspring, the second dam, and the second dam’s other offspring are all taken into consideration. The numerical scores are highest for a G1 win, then a G1 placing, then a G2 win, and so on, all the way down to non-stakes wins, then non-stakes placings. Never placing in a race leads to a score of zero. If the dam is unraced, then she gets a portion of her dam’s score; all others receive a score of zero for being unraced. Multiple graded and non-graded stakes placers receive a 15% uplift to their base scores.

The dam’s other offspring racetrack performance gets weighted most heavily, followed by the dam, then the second dam, and finally the second dam’s other offspring. The Bingel Score doesn’t penalize the dam for the last three offspring that she has produced, including the yearling in the sales ring, because they are deemed too young. Any positive two- or three-year old racetrack performances are included.

The key is that the Bingel Score changes with each yearling that the dam produces. The racetrack performance of the dam, the second dam, and the second dam’s other offspring are all really important when considering the first yearling that the dam produces. With each offspring that the dam produces, however, these influences gradually wane so that the racetrack performance of the dam’s other offspring dominates as a variable.

How was the Bingel Score tested? Please stay tuned for the next post.

Posted in Archive | Tagged , , , | Leave a comment