Between a Rock and a Hard Place

What is also interesting is that “A” nicks far outweigh any other category. This is perplexing, as it seems to me, that an “A” nick should be the nick that occurs least frequently. We certainly know that superior runners occur least frequently. It seems that by sheer numbers alone the “A” nicks are doomed to failure. What am I missing?

The above is a comment from Ned Williams. Here is my reply:

You are not missing anything. You are correct that A nicks being 40% of the entire sale (as opposed to 13%-15% of the overall population) sticks out like a sore thumb. And you are probably also correct that A was doomed to failure by its sheer numbers.

What people have a hard time grasping is that the Keeneland September sale is a pretty good sale overall. Even the pedigrees offered on the final day of the sale are not bad pedigrees (relative to the entire population, which is a comment on the lack of quality in the entire population). Sales foals are generally better than the entire population. The Keeneland September sale is better than most sales. Therefore, it is not really surprising to me that 40% of all nags sold in the 2010 Keeneland September sale were A nicks.

I have secured the links to the original cheatbooks for the 2010-2011-2012 OBS August yearling sales. That sale is much lower down on the scale than Keeneland September. Should not have nearly as many A nicks (although you never know). Might be a better population to study for that reason. A nicks might fare better in that environment. Will see. Worth a look.


Some other foolosophical reflections on the subject of nick ratings. As I understand it, TrueNicks tries to base their nick ratings on the sire-broodmare sire combination. If the computer finds enough foals to fit that description, they do base their ratings on that combination. Below is a quote from the TrueNicks explanation in the 2010 Keeneland September cheatbook:

“The TrueNicks program is designed to establish a rating within a minimum number of
generations from the sire/broodmare sire cross. As soon as a cross has met statistically significant thresholds, a rating is calculated. The farthest the TrueNicks program will go to calculate a rating is to the third generation of the horse (or prospective mating) in the sire line, and the fourth generation of the horse (or prospective mating) in the broodmare sire line. Ratings beyond those parameters would be meaningless and will yield a No Rating score.”

This written explanation does not specify what constitutes “statistically significant thresholds” (perhaps it varies), but I am guessing about ten foals. The only problem with that is that ten foals is really not a very big sample upon which to base a rating. And not many sire-broodmare sire combinations have 50 or 100 foals or more upon which to base a rating.

If the computer does not find enough foals, it goes back another generation or two (or more) to find enough foals and uses the total number of foals found for that particular combination. The only problem with this is that going back another generation or two (or more) makes any predictions based on that combination much more weakly correlated to the foal in question. If you have learned anything at all from this blog, I hope it is that “influence” recedes much more rapidly than most people think with each generation farther back you go.

So if you are in the business of making nick ratings scientifically (which not all of them do; some of them just skip the scientifically), you are in between a rock and a hard place. You can make a rating based on ten-plus foals, which may not be every reliable because that is really not enough foals. Or you can go back another generation or two and make a rating that way, which may not be very reliable because it is several generations removed from the foal in question.

In extreme cases you may not be able to make any rating at all because even going back several generations you can not find enough foals bred similarly to the foal in question. It must be pretty galling to discover that this not rated group actually had pretty good results (especially since buyers paid the least amount of money for them).

Here is something people may not understand about nick ratings. They are basically past performances. And as sellers of mutual funds are fond of saying, past performances are no GUARANTEE of future results.

Here is where I differ foolosophically from the makers of nick ratings. The latter might say, “This is an A+++ nick. It should improve your chances of getting a stakes winner by 100%.”

And I might reply, “Don’t think so. Maybe 10%-20%. Nowhere near 100%.”

So it is a question of degree. It makes perfectly logical sense to expect a mating (or a similar mating) that has worked well in the past to work well again. The question is HOW MUCH better results should you expect????? I think the makers of nick ratings exaggerate the amount of improvement they can deliver (not to mention its predictability). They exaggerate the importance of nicks.

I do not deny the existence of nicks. I merely think that their ability to deliver better results (not to mention the predictability of those results) has been exaggerated. Their importance has been exaggerated by people who are trying to make $$$$$$$$ doing so.

Those are some thoughts you might want to keep in mind as you go about your business of buying and/or selling at the Keeneland September yearling sale (or any sale, for that matter). Good luck at the sales!!!!!!!!!!!!!!!


That concludes this five-part series on nick ratings. Many of you seem to have missed the most important one, post number two. So here is a link back to that post if you want to use it.

Posted in Uncategorized | Tagged , , | 8 Comments

Off the Beaten Path

I thought it might be helpful to list the best stakes winners (those with 1,000+ Performance Points) for the various nick categories. Listed for each stakes winners is its name, pedigree (sire–dam, broodmare sire), price for which it sold at the 2010 Keeneland September sale ($ omitted), and the number of Performance Points it earned. The stakes winners are listed in descending order (best ones first, those with the most Performance Points).

A Stakes Winners (1,000+ Performance Points)

Dullahan (Even the Score–Mining My Own, Smart Strike), 250,000, 2,936 Performance Points.

Creative Cause (Giant’s Causeway–Dream of Summer, Siberian Summer), 135,000, 2,039.

Daddy Nose Best (Scat Daddy–Follow Your Bliss, Thunder Gulch), 35,000, 1,757.

Lady of Fifty (After Market–K. D.’s Shady Lady, Maria’s Mon), 40,000, 1,693.

Blueskiesnrainbows (English Channel–Cho Cho San, Deputy Minister), 33,000, 1,572.

Paynter (Awesome Again–Tizso, Cee’s Tizzy), 325,000, 1,502.

Sabercat (Bluegrass Cat–Miner’s Blessing, Forty Niner), 120,000, 1,198.

Jimmy Creed (Distorted Humor–Hookedonthefeelin, Citidancer), 900,000, 1,139.

Northern Passion (First Samurai–A Touch of Glory, Golden Gear), 95,000, 1,134.

Macho Macho (Macho Uno–Dazzling Contrast, General Meeting), 160,000, 1.082

Evelyn’s Dancer (Songandaprayer–Seraphic Too, Southern Halo), 4,500, 1,028.

Pianist (More Than Ready–Red Piano, Red Ransom), 100,000, 1,009.

Dance Card (Tapit–Tempting Note, Editor’s Note), 67,000, 1,002.

B Stakes Winners (1,000+)

I’ll Have Another (Flower Alley–Arch’s Gal Edith, Arch), 11,000, 4,194.

My Miss Aurelia (Smart Strike–My Miss Storm Cat, Sea of Secrets), 550,000, 4,147.

Elusive Kate (Elusive Quality–Gout de Terroir, Lemon Drop Kid), 70,000, 3,284.

Bodemeister (Empire Maker–Untouched Talent, Storm Cat), 260,000, 1,705.

Liaison (Indian Charlie–Galloping Gal, Victory Gallop), 290,000, 1,667.

Isn’t He Clever (Smarty Jones–Sharp Minister, Deputy Minister), 45,000, 1,237.

Hero of Order (Sharp Humor–Ocean Sprite, Ocean Crest), 3,000, 1,048.

C Stakes Winner (1,000+)

Contested (Ghostzapper–Gold Vault, Arch), 110,000, 1,634.

D Stakes Winners (1,000+)

Fed Biz (Giant’s Causeway–Spunoutacontrol, Wild Again), 950,000, 1,710.

Drill (Lawyer Ron–Cat Dancer, Storm Cat), 300,000, 1,450.

Belle Gallantey (After Market–Revealed, Old Trieste), 10,000, 1,271.

Chips All In (North Light–Maria’s Mirage, Maria’s Mon), 4,500, 1,151.

In looking at the lists above I recommend paying particular attention to the sires and broodmare sires involved (the basis for most nick ratings). I can not really tell at a glance why the A stakes winners are rated highly and the D stakes winners are rated lowly. It is not evident to the naked eye (at least to my naked eye). Perhaps you can see something I do not.

(That brings up a point. It is relatively easy to to see with the naked eye how good a pedigree is. A foal by A.P. Indy out of a G1 winner by Theatrical is obviously a pretty good pedigree. Ditto for a foal by Storm Cat out of a G1 winner by Unbridled.

Yet the former might be a D nick and the latter an A nick. The nick rating purports to provide insight that can not be determined by the naked eye. That is what they are selling. Unfortunately, their “insight” is not nearly as good as they want it to be when it comes to predicting future racetrack success.)

Because the small not rated group had such good success, I decided to list all 11 of their stakes winners. I do have some comments on this group and will discuss them after the list.

Not Rated Stakes Winners (all)

Grace Hall (Empire Maker–Season’s Greetings, Ezzoud), 95,000, 2,946. (????)

Judy the Beauty (Ghostzapper–Holy Blitz, Holy Bull), 20,000, 2,312. (now rated A)

Gypsy Robin (Daaher–Feisty Princess, Indian Charlie), 27,000, 1,499. (now rated A++)

Private Zone (Macho Uno–Auburn Beauty, Siphon), 15,000, 1,317. (still not rated)

Silverette (Street Sense–Holy Lightning, Holy Bull), 290,000, 698. (still not rated)

Mucho Mas Macho (Macho Uno–A. P. Andie, Star de Naskra), 10,000, 656. (now rated A++)

Raging Daoust (Kafwain–Champagne Forever, Caveat), 10,500, 537. (????)

Street Life (Street Sense–Stone Hope, Grindstone), 130,000, 438. (now rated B)

Rocket Twentyone (Indian Charlie–Symphonic Lady, Blare of Trumpets), 95,000, 336. (????)

Lulu Wong (Badge of Silver–Lovely Cool, Indian Charlie), 27,000, 278. (now rated A++)

Window Boy (Include–Window Woman, Skip Away), 12,000, 217. (now rated A++)

(I have added a notation to these 11 stakes winners as to what their nick ratings are now. Two are still not rated. Three are unknown. Of the six who are now rated, four are now A++, one is A, and one is B.)

The first thing that jumps out at me about these 11 stakes winners is that their broodmare sires mainly hail from odd sire lines. The mainstream for both male lines and broodmare sire lines is Northern Dancer and/or Mr. Prospector. Only one of the broodmare sires above is from the Northern Dancer male line. That is Ezzoud, broodmare sire of Grace Hall. Ezzoud is by Last Tycoon, by Try My Best, by Northern Dancer. Ezzoud, Last Tycoon, and Try My Best are not exactly mainstream Northern Dancers, at least in terms of North American pedigrees.

Only one of the broodmare sires above is from the Mr. Prospector male line. That is Grindstone, broodmare sire of Street Life. Grindstone was by Unbridled, by Fappiano, by Mr. Prospector. That is more in the mainstream than Ezzoud, but Grindstone is not exactly one of the leading broodmare sires.

The other nine broodmare sires involved above (Holy Bull twice, Indian Charlie twice, Siphon, Star de Naskra, Caveat, Blare of Truumpets, and Skip Away) are all pretty far off the Northern Dancer/Mr. Prospector beaten path. And Indian Charlie shows up again as the sire of Rocket Twentyone.

So I can see why a computer might have a tough time generating  nick ratings on these 11 pedigrees. They are generally somewhat obscure pedigrees, especially off the beaten path in their broodmare sire lines. Not many foals have been bred similarly.

Another thing I noticed about these 11 pedigrees is that four of them (including the top three) were out of dams who were stakes winners. Season’s Greetings (dam of Ezzoud) was a stakes winner in France. Holy Blitz (dam of Judy the Beauty) was a listed stakes winner. Feisty Princess (dam of Gypsy Robin) was a stakes winner with a record of 16-5-4-3 for earnings of $141,613. A. P. Andie (dam of Much Mas Macho) was a listed stakes winner and an earner of $262,402.

Additionally, Raging Daoust and Rocket Twentyone were out of stakes-placed mares (Champagne Forever and Symphonic Lady, respectively). So six of the 11 were out of black-type mares. That seems to me to be a pretty good percentage, better than that for the sale overall.

I am suggesting that class in the dam might have been the overlooked factor in these 133 not rated foals generally and these 11 stakes winners specifically.

Posted in Uncategorized | Tagged , , | Leave a comment

Put Up or Shut Up

Ned Williams commented on my post from two days ago:

I can already hear the wailing and gnashing of teeth….”but my nicks are better than the those found in the cheat books.” My reply to the purveyors of nicks, If so please provide your Nick ratings (of the above horses) to Boojum so he can run the numbers and prove your ratings are better predictors of racetrack success.

To which I replied:

Better yet, send me your lists of nick ratings for the entire 2014 Keeneland September sale (after the sale is over, if that makes you more comfortable). And about three years down the road, about 2017 or so (if I am still alive), I will go through the sale again and report on how it all turned out.


That was Wednesday. On Thursday Byron Rogers of TrueNicks graciously provided me with their nick ratings for the 2014 Keeneland September yearling sale.

Think your nick ratings are better???? ENicks is the only other company I am aware of that provides these ratings for a profit, but there are probably others. Let me hear from you if you accept this challenge. Send me your 2014 Keeneland September nick ratings by commenting on this post.

Put up or shut up.

Posted in Uncategorized | 4 Comments

An ENTIRELY Different Can of Worms

As promised, below are the results for the 2010 Keeneland September yearling sale by nick ratings. APPPSW stands for average Performance Points per stakes winner, a measure of the quality of the stakes winners involved. Usually it is a little over 600. For this group it is 710, which is not too surprising, considering that Keeneland September is a pretty good sale in terms of quality of pedigrees and horseflesh offered (and in this case sold).

Nick Rating          Foals          Stakes Winners          %          APPPSW           PPI (Result)

A                            1,230                    58                    4.72             701                      0.94

B                               628                     29                    4.62            876                       1.15

C                               529                     25                    4.73            490                       0.66

D                               539                     29                    5.38            631                       0.96

Not rated                 133                     11                     8.27          1,021                     2.40

Totals                    3,059                  152                     4.97             710                     1.00

Over 4,800 yearlings were cataloged in this sale, but only 3,059 actually sold, and those 3,059 constitute the sample group. Almost 5% (152 of 3,059, 4.97%) of those sold became stakes winners. The A, B, and C nick groups are fell below that benchmark at 4.72%, 4.62%, and 4.73% respectively. Extremely interestingly, the D group was better than all of the above at 5.38%. And most interesting of all, the small not rated group was best of all at 8.27% (11 stakes winners from only 133 sold).

The not rated group was also best by APPPSW at 1,021. Included among their 11 stakes winners were Grace Hall (2,946 Performance Points), Judy the Beauty (2,312), Gypsy Robin (1,499), and Private Zone (1,317). The B group was next at 876. Included among their 29 stakes winners were champions I’ll Have Another (4,194) and My Miss Aurelia (4,147). I’ll Have Another was sold for only $11,000 near the end of the sale.

A, C, and D were below the 710 benchmark at 701, 490, and 631, respectively.

So overall, not rated was by far the best at 2.40, and B was also positive at 1.15. D actually nosed out A for third (0.96 to 0.94), and C was a distant last at 0.66.

These results would have been interesting if the prices for all five groups had been about the same. But as demonstrated in my last post, the prices were NOT nearly the same for all five groups. The chart below summarizes prices versus results.

Nick Rating          Foals          Price Index           PPI (Result)

A                            1,230               1.12                         0.94

B                               628                0.98                        1.15

C                               529                0.94                        0.66

D                               539                0.86                        0.96

Not rated                 133                0.79                        2.40

A was a big disappointment (price of 1.12 and result of 0.94). B was almost the opposite of A (price of 0.98 and result of 1.15). C was miserable (price of 0.94 and result of 0.66). D was not bad (price of 0.86 and result of 0.96). And not rated was too good to be true (price of 0.79 and result of 2.40).

You have probably heard the handicapping aphorisms: “A nag is usually not as good as his last good race. A nag is usually not as bad as his last bad race.” I throw that in here to concede that not rated is probably not as good as it appears to be here on paper. C is probably not as bad as it appears to be here on paper. A may just have thrown in a bad race in 2010 (a little skeptical about that). But since A was by far the most expensive group, the most important comparison to be made here is A versus all others. The next three charts summarize the data.

Nick Rating          Foals          Average          Maverage        Price Index

A                             1,230        $78,281              218.67                 1.12

All others               1,829        $55,783              180.14                 0.92

A was clearly a lot more expensive than all others.

Nick Rating          Foals          Stakes Winners          %          APPPSW           PPI (Result)

A                            1,230                    58                    4.72             701                      0.94

All others              1,829                    94                    5.14             715                      1.04

A was clearly not as successful on the track as all others.

Nick Rating          Foals          Price Index           PPI (Result)

A                            1,230               1.12                         0.94

All others              1,829               0.92                        1.04

Therefore, A is burdened with the dreaded double whammy. A cost more than average and produced racetrack results lower than average. All others collectively did the opposite.

Even conceding that not rated is probably not as good as it appears here on paper, its excellent performance cannot be ignored (especially since it was the CHEAPEST of the five groups). The 133 not rated foals were products of matings that had been tried so few times (if ever) that they literally could not be rated by a computer.

That they succeeded so well suggests that nicks are vastly overrated. After all, nick theory preaches that certain combinations of sires and dams exceed their opportunities by large margins. Therefore, knowing the correct nicks is important.

The success of the not rated group casts some extreme doubt on that. It suggests that the important thing is not so much the combination of sire and dam, but the overall quality of sire and dam. I will expand on this theme with relation to the not rated group in another post.

And a final observation. This is sheer speculation on my part. I have nothing concrete to back it up except my impressions as I mined the data. It seemed to me that the C group was mainly foals whose sire-dam combinations had been tried often enough without much (if any) success. The D group was mainly foals whose sire-dam combinations had not been tried very frequently at all.

The most important thing to take from this study is that A nicks cost more and produced less. It is possible that A nicks would fare better in a study of a different sales population. I have another such study in the works and hope to present it before the end of the year.

A nicks from the 2010 Keeneland September yearling sale had very little predictive value of future racetrack success. That is about as politely as I can put it. As some of you know, I can be much more BLUNT than that if I feel like it. But I am feeling relatively mellow and polite this morning. Therefore, I will merely repeat that polite conclusion.

A nicks from the 2010 Keeneland September yearling sale had very little predictive value of future racetrack success.

Posted in Uncategorized | Tagged | 1 Comment

Nick Ratings–A Different Can of Worms

“So how come you haven’t written anything about nicks????”

That question was posed to me back in January of this year. I was treating a friend to lunch for her birthday. I answered her question extensively, perhaps too extensively. I will try to boil it all down to its essence here.

I told her that it was not because I actually believe in nicks. Nicks are just as much BS as any other pedigree theory. The problem is how to go about proving that nicks are just as much BS as any other pedigree theory.

Specifically, the problem is getting hold of the original data, the original nick ratings. If I could do that, I said, I would love to take one entire Keeneland September yearling sale, for example, divide the nags that sold into categories by their nick ratings (A, B, C, D, et al), and let the chips fall where they may.

The only problem was that I could not finger out how to get hold of those original ratings. Even if I wanted to pay the exorbitant price demanded for the current nick ratings of these nags, the current nick ratings are irrelevant. What is important is knowing the nick ratings AT THE TIME THESE NAGS WERE SOLD as weanlings, yearlings, or two-year-olds (my preferred mining grounds). That was the true test of reality for nick ratings, seeing if they had any predictive value whatsoever.

In the back of my mind I thought I had seen nick ratings published in some of the sales “cheat books” (buyers’ supplements) over the years. I talked to some people about it. None of the leads panned out.

Then early one Sunday afternoon I was merrily plugging along mining data (now into sales foals of 2008-2011). I saw an ad in the Bland-Horse for their cheat books. “EXCLUSIVE: Auction Edge now includes TrueNicks ratings!” this ad proudly proclaimed. This was the issue of January 2, 2010 (the auction review issue naturally).

EUREKA!!!!!!!!!!!!!!! Looks like they started including nick ratings in their cheat books for auctions of 2010. So I called up the Bland-Horse the next day. I fully expected them to say that they no longer had any cheat books to sell me from the 2010 Keeneland September yearling sale. And that’s exactly what they did say, explaining that all such books were recycled after each sale.

But they also mentioned that they could sell me a link to those old books. So for a measly $10 I procured a link to the cheat books with nick ratings for the 2010 Keeneland September yearling sale. HOT DAWG!!!!!! I was in business.

The 2014 Keeneland September yearling sale begins in less than a week, on Monday, September 8. So I have to say that the timing of this post is pretty good. Whether you are a buyer or a consignor, these nick ratings have some effect on the amount of money you pay (in the case of the former) or receive (in the case of the latter) for Keeneland September yearlings.

So without further ado, here are the price breakdowns by nick ratings for the 2010 Keeneland September yearling sale.

Nick Rating          Sold          Average          Maverage          Price Index

A                           1,230        $78,281               218.67                    1.12

B                              628        $62,506               192.53                   0.98

C                              529        $58,054               183.08                   0.94

D                              539       $49,385                168.99                   0.86

Not rated                133       $40,518                155.12                   0.79

Totals                   3,059      $64,811                 195.63                  1.00

For the sake of clarity, it should be pointed out that A ratings include A+ and A++, B ratings include B+ and B++, etc.

If you like your numbers to make sense, you should love the chart above. A is higher than B in all three categories, B is higher than C, C is higher than D, and D is higher than not rated.

The overall average for all 3,059 yearlings sold was $64,811. The overall maverage was 195.63. I am treating this is an independent population. So that overall maverage of 195.63 corresponds to a Price Index of 1.00.

Only the A group ($78,281) was higher than the overall average. Only the A group (218.67) was higher than the overall maverage. The other four groups were all below the overall average and maverage.

If you need a translation for this, it means that if you were the consignor of a yearling with a A nick rating, you love this stuff. If you were the consignor of a yearling with a nick rating below A, you hate this stuff. If your yearling had a nick rating below A, you were probably penalized in terms of price received.

These results are probably not at all surprising to anyone. As long as the racetrack results followed the same pattern, everything is hunky-dory. If the racetrack results deviated from this pattern, that is a whole different can of worms. If, for example, A is NOT the best of the five groups in terms of racetrack results, that is an ENTIRELY different can of worms.

Racetrack results (the reality test) are in my next post. Stay tuned.

Posted in Uncategorized | 4 Comments

The California Chrome Criterion (PPPP)

What proportion of modern Thoroughbred pedigrees does NOT trace to Phalaris in the male line???? I estimated an answer to that question awhile back at about 5,000 of the 70,714 sales foals of 2003-2007, or about 7%.

That estimate could have been a lot better. By actual count I came up with 6,616 foals NOT tracing to Phalaris in the male line. That is a little over 9% of all 70,714 sales foals of 2003-2007. The main sources of NON Phalaris in the male line these days are In Reality, Damascus, and Ribot.

The pedigree of California Chrome is an example of another phenomenon. The four sires in his third generation are A.P. Indy, Cozzene, Mr. Prospector, and Polish Numbers. All four are Phalaris-line sires.

This is not exactly a rare phenomenon. In fact, it is becoming more of the norm than any other combination of Phalaris among the four sires in the third generation. Of the 13,099 sales foals of 2003, 3,862 (29.5%) met this (California Chrome) criterion. Of the 14,796 sales foals of 2007, 6,409 (43.3%) met this criterion. That increase from 29.5% to 43.3% in only four years suggests to me that having all four sires in the third generation trace to Phalaris in the male line is becoming the norm, if not the absolute majority, in modern Thoroughbreds.

It was suggested to me that I should study this phenomenon more closely. I decided to do so. There are 16 possible combinations, ranging from PPPP (all four sires in the third generation tracing to Phalaris in the male line) to OOOO (none of the four sires in the third generation tracing to Phalaris in the male line).

I examined all 16 combinations by their prices and results, but I have decided not to bore you with all 16 results. Instead I combined them into five groups: PPPP (all four Phalaris), three of four Phalaris, two of four Phalaris, one of four Phalaris, and zero of four Phalaris (OOOO). The prices are summarized below.

Combination               Foals          Average         Maverage       Price Index

Phalaris 4X                 24,968       $67,984             183.29                1.12

Phalaris 3X                 30,063       $50,699            161.01                0.99

Phalaris 2X                 13,259        $40,981            139.40               0.85

Phalaris 1X                   2,321         $26,656            116.19               0.71

No Phalaris                      103         $15,835              81.99              0.50

The prices are all pretty much as expected. The more male-line Phalaris in the pedigree, the higher the prices. Only 103 foals had no Phalaris-line sires in their third generations, and they were the cheapest of all. Phalaris 3X actually had more foals than Phalaris 4X, but I suspect that those results will have reversed in more recent crops.

The racetrack results are summarized below. APPPSW stands for average Performance Points per stakes winner, a measure of the quality of stakes winners involved, with 620 being average.

Combination               Foals          SWs          %           APPPSW          PPI (Result)

Phalaris 4X                 24,968       977          3.91              613                   1.13

Phalaris 3X                 30,063    1,023         3.40             648                   1.04

Phalaris 2X                 13,259       356          2.68             587                   0.74

Phalaris 1X                   2,321         57           2.46             437                   0.51

No Phalaris                      103          2            1.94             386                   0.35

The results pretty much mirror the prices. The more male-line Phalaris in the pedigree, the better the results. The only glitch is that Phalaris 3X was better by APPPSW (648) than Phalaris 4X (613). That result was reversed by overall PPI, 1.13 to 1.04 in favor of Phalaris 4X over Phalaris 3X.

The table below compares prices with results.

Combination               Foals          Price Index          PPI (Result)

Phalaris 4X                 24,968             1.12                       1.13

Phalaris 3X                 30,063            0.99                      1.04

Phalaris 2X                 13,259             0.85                      0.74

Phalaris 1X                   2,321             0.71                       0.51

No Phalaris                      103            0.50                       0.35

Phalaris 4X was slightly positive (price of 1.12 and result of 1.13). Phalaris 3X was a little more positive (price of 0.99 and result of 1.04). The other three groups were increasingly negative. Not too many surprises overall.

One group that had pretty good results was PPPO (Phalaris-Phalaris-Phalaris, with the sire of the second dam NON Phalaris). There were 12,779 such foals, with an average of $54,657, a maverage of 166.80, and a Price Index of 1.02. Included among those 12,779 foals were 440 stakes winners with average Performance Points of 692 for a PPI (result) of 1.13. A price of 1.02 and a PPI of 1.13 is a pretty good result.

Those 6,616 foals without Phalaris in the male line also had pretty good results. They were pretty cheap, selling for an average of $36,264, a maverage of 133.01, and a Price Index of 0.82. Included among those 6,616 foals were 214 stakes winners with average Performance Points of 642 for a PPI (result) of 0.98. A price of 0.82 and a PPI of 0.98 is a pretty good result.

Since the 6,616 foals without Phalaris in the male line were overperformers, it follows that the other 64,098 foals with Phalaris in the male line were underperformers. Those 64,098 foals sold for an average of $55,985, a maverage of 166.22, and a Price Index of 1.02. Included among those 64,098 foals were 2,201 stakes winners with average Perfomance Points of 618 for a PPI (result) of 1.00 (1.002 if you want to get technical). A price of 1.02 and a PPI of 1.002 is a slightly negative result.

To recapitulate, the PPPP combination was not exactly magical, with a price of 1.12 and a result of 1.13 (very slightly positive). The totality of the Phalaris male line (all combinations starting with P) was a bit of an underperformer (see paragraph above).

But these things have a momentum of their own. It seems almost inevitable that the Phalaris male line will go to 100% one of these days, perhaps sooner than you think, wiping out all other male lines.

Some of you will bemoan the passing of all other male lines except Phalaris. I do not particularly do so. Male lines are a human construct. It is you who are attaching so much importance to male lines, more importance than they really deserve. The male line is only a tiny portion of a pedigree, especially if you trace it back a century to 1913, when Phalaris was foaled. Pedigrees are infinitely larger than just their male lines.

Posted in Uncategorized | Leave a comment

The Confusion of Quantity With Quality

I am going to resort to Racehorse Breeding Theories, Chapter 14, pages 273-276, to continue this discussion of the Bruce Lowe Figure System.

“Lowe traced the winners of the English Oaks, Derby, and St. Leger through their female families back to the first registered mare found in the English Stud Book. . . .

“Lowe found that some families had produced many more classic winners than others. The family that had produced the most classic winners at that time . . . he labeled the Number 1 Family; the family with the second-highest total was the Number 2 Family and so forth. . . .

“Estes’s work with statistics showed that the Figure System was actually a classic case of opportunity. The most populous female families were, in general, the most successful. The families with the larger number of classic winners were the same families who accounted for the larger proportion of the breed. And their success was in proportion to their representation. A family might account for 12 percent of the breed through the tail-female line, and its members would have won 12 percent of the classics, more or less. . . .

“Furthermore, the Lowe families from first to least, for instance, also follow their rankings in having the largest number of claiming race winners. The rankings, in the final estimation, are a measure of quantity but not quality.”

I would like to emphasize that point about the families with the most classic winners also having the highest number of foals. If we substitute stakes winners for classic winners and examine sales foals of 2003-2007, the same holds true today. The chart below illustrates.

Family Number          Foals          Stakes Winners

1                                    10,036                342

4                                     7,893                 255

9                                     6,640                 245

2                                     5,837                 188

The four most popular families (by number of foals) today are one (10,036), four (7,893), nine (6,640), and two (5,837). No other family has more than 5,000 foals.

Not coincidentally, those same four families also rank first through fourth, in the same order, by number of stakes winners: one (342 stakes winners), four (255), nine (245), and two (188).

The above also illustrates that popularity waxes and wanes over time. Among the top ten Lowe families, numbers six (only 1,201 foals), seven (1,503 foals), and ten (1,747 foals) have declined in popularity over time.

The chart below summarizes racing results for all 21 groups. APPPSW stands for average Performance Points per stakes winner, a measure of the quality of stakes winners involved, with 620 being average.

Family Number      Foals         Stakes Winners      %       APPPSW          PPI (Result)      Rank

19                               1,475                   47                 3.19         823                     1.24                 1

3                                 4,361                 147                 3.37        764                      1.22                 2

9                                 6,640                245                 3.69        694                      1.21                 3

22                               1,677                  57                  3.40        734                      1.18                 4

13                               2.258                  86                  3.81        650                      1.17                 5

11                                1,171                  48                  4.10       606                       1.17                 6

10                               1,747                   74                 4.24        544                      1.09                 7

1                               10,036                342                3.41         669                      1.08                 8

21                               1,292                  46                 3.56         643                      1.08                9

14                               2,523                  92                 3.65        622                       1.07               10

4                                 7,893                 255                3.23        608                      0.93               11

5                                 3,378                 101                2.99        651                       0.92               12

16                               3,430                 121                3.53        541                       0.90               13

all others                   4,584                 158                3.45        529                       0.86               14

8                                 4,637                 151                3.26         561                      0.86                15

23                              2,019                  70                  3.47        523                      0.86                16

2                                5,837                  188                3.22        558                      0.85                17

6                                1,201                    36                3.00        578                      0.82                18

20                              1,371                    41                 2.99        578                      0.82                19

7                                 1,503                   54                3.59        469                      0.80                20

12                               1,681                   56                3.33         486                      0.76                21

The groups are also ranked by their PPIs (results). Ten of the 21 groups have PPIs above 1.00. 11 have PPis below 1.00. That is about as expected. Only three of the ten groups with PPIs above 1.00 were from the top ten Lowe families: family number three finished second, number nine finished third, and number one finished eighth. The top ten Lowe families should have had at least five groups in the top ten by PPI (results); they actually had only three.

The chart below compares prices with results. For example, family number 19 had a Price Index of 0.98 and a PPI (result) of 1.24. The latter is 0.26 higher than the former. That is an excellent (positive) result. Positive numbers in the Comparison column are good. Negative numbers are bad.

Family Number      Foals         Price Index          PPI (Result)         Comparison

19                               1,475             0.98                       1.24                      +0.26

3                                 4,361            1.01                        1.22                       +0.21

9                                 6,640           0.99                        1.21                       +0.22

22                               1,677            1.01                        1.18                        +0.17

13                               2.258           1.08                        1.17                        +0.09

11                                1,171            1.05                        1.17                        +0.12

10                               1,747            0.95                        1.09                       +0.14

1                               10,036           0.99                        1.08                       +0.09

21                               1,292            1.05                        1.08                       +0.03

14                               2,523            1.00                        1.07                       +0.07

4                                 7,893            0.99                        0.93                     –0.06

5                                 3,378            1.04                        0.92                     –0.12

16                               3,430           0.98                        0.90                     –0.08

all others                   4,584          1.02                          0.86                     –0.16

8                                 4,637           1.02                         0.86                     –0.16

23                              2,019           1.04                          0.86                    –0.18

2                                5,837           0.98                          0.85                    –0.13

6                                1,201           0.87                          0.82                    –0.05

20                              1,371          0.89                           0.82                    –0.07

7                                 1,503          1.01                          0.80                     –0.21

12                               1,681          1.04                          0.76                     –0.28

Interestingly, all ten groups with PPIs above 1.00 also had positive Comparison numbers. All 11 groups with PPIs below 1.00 also had negative Comparison numbers.

You probably want to take those results for family number 19 with a grain of salt. Curlin (13,802) accounted for more than a third of the total Performance Points garnered by that family (38,696). Without Curlin the PPI of this group falls from 1.24 to 0.80.

Family number 12 was the worst in the chart above with a Price Index of 1.04, a PPI of 0.76, and a Comparison number of –0.28. I can not come up with a good explanation for this poor result.

But do not take these numbers too seriously. Do not take this theory seriously at all. As I said at the beginning of my last post, the Bruce Lowe Figure System has long been discredited. And for very good reasons.

I will conclude with another quote from Racehorse Breeding Theories (page 276):

“Lowe’s work was interesting because it urged breeders to take more thought about the bottom lines in their breeding plans, but the theory was not thought out with an understanding of statistics and opportunity. As a result, it would have little use for the practical breeder today.”

Many breeding theories are “not thought out with an understanding of statistics and opportunity.” Many breeding theories confuse quantity with quality. The Bruce Lowe Figure System was a classic example of this confusion.

I hope these posts have helped readers to a better understanding of the statistics of pedigrees. Specifically, NOT to confuse quantity with quality.



Posted in Uncategorized | Leave a comment