Thread Rating:

DrawingDead
DrawingDead
Joined: Jun 13, 2014
  • Threads: 9
  • Posts: 2233
March 12th, 2016 at 8:35:45 PM permalink
Better to be lucky than good? That race was not run as I was expecting. Neither did my horse. I was just plain wrong. I had Destin across the board, but expecting something different than the kind of (successful) trip he got. In track record time. But I get zero brownie points for horseplaying prowess, and if there was any justice in the racing world I should've lost my money. There isn't; I don't plan to give it back. Sue me.

"I'm against stuff like crack and math" --AxelWold
Keeneone
Keeneone
Joined: Aug 16, 2014
  • Threads: 13
  • Posts: 996
March 12th, 2016 at 9:18:02 PM permalink
Congrats on the Destin win. The $2 double with Tepin was also a very nice $17.40 (for those who had it).
I am going to go out on a limb...
His assigned Beyer in the TB Derby will be the highest yet for any of the Derby hopefuls.
lilredrooster
lilredrooster
Joined: May 8, 2015
  • Threads: 148
  • Posts: 2501
March 13th, 2016 at 1:26:17 AM permalink
Mr. Drawing Dead: Thank you for answering my question. Since you were so kind to do so I would like to impose upon you with another question in order to solicit your opinion which I regard as being an expert opinion based upon the depth of knowledge about horse racing that you continually show.


Being by nature a suspicious person I have always questioned the value and legitimacy of speed figures both
Beyer's and those from Equibase. The reason: In comparing the average (or par) times of races at the same
level at the same track to final times of today's races they then assign a track variant (relative slowness or
fastness of the track) from which each horse 's speed figure is then derived. The flaw in my opinion: To do
this they are using data from only 9 or 10 races, however many the track had on its card. Come on: 9 races
or what a mathematician would call trials or resolutions!!! Gee whiz, when pro blackjack players test a strategy
against a new rule change they will run a computer simulation that gives results based on one BILLION hands.
This is relatively easy with today's powerful computers. So blackjack experts need a billion hands to come to
a conclusion but speed figure sellers only need 9 or 10? I have a sneaking suspicion that if a mathematician
were to analyze the way speed figures are calculated that he would call it voodoo mathematics. In writing this
and asking this question I acknowledge the fact that I may be wrong or missing something. Thanks for your
help if you are willing to answer this question for me. If you are not or don't have time that's fine, I totally
can understand that. I also realize that there has to be some way to account for the drastically different
final times at different tracks. If I remember correctly Golden Gate use to have 15,000 claimers running
6 furlongs at 1:08 and change more than good enough to win stakes races as we move eastward.
So it has to be in some way accounted for. But speed figures as they now are: I just don't know.
"that ball hit his hands. that's a foul ball. the hands are part of the bat"........... "O.K. you think the hands are part of the bat then go down to Dick's Sporting Goods and buy a bat with hands attached." former MLB umpire Jim Evans
DrawingDead
DrawingDead
Joined: Jun 13, 2014
  • Threads: 9
  • Posts: 2233
March 13th, 2016 at 3:32:00 AM permalink
I agree with you about 63.472%. But definitely not more than 64.192% after the mid-afternoon renovation break when the discussion track was harrowed but insufficiently watered by the grounds crew.

You can just call me Dead, if you like. And I will be exactly that before trying to re-play the evolution and regurgitate in excruciating detail all the relevant sagging bookshelves on it here in message board format, so I won't be deep sea diving in that ocean. Two full shelves not far from my elbow right now are hardbound Beyer on Speed, Quinn on Beyer, Beyer on Ludes, Brohamer on Sartin on Quin on Beyer on a Pedestal With A Partridge in a Pear Tree.... I may have just now revised some of those book titles slightly, but not the author names in case you want to follow the published trail of the topic. You've heard the name Beyer? Start there, all his books are still in print, and from that jumping off point you'll find enough material to occupy many months, if you like. But I don't want to just brush your thoughts off with a "go read" and leave it totally at that.

While I happen to think a good bit of your speculative criticism of speed figures is well founded, I also think it may have carried the skeptical critique a bit far. If someone somewhere really fervently started to imagine that an analogy to something like a precise statistical analysis of the effect of variance on the very tiny presumed underlying fractional edge or loss occurring in methods for card games or video poker play is a highly releveant cousin to this, then one has apparently had the misfortune to hang out overmuch with some especially foolish math-THEMED (but judgment deprived) geeks who need to get out more and contemplate the vast range of imprecise decisions that a human being continually makes and MUST make in the course of living in the world daily, if they are to avoid becoming a hazard to themselves and others, or for that matter tie their shoes without assistance. But a skeptical eye is called for when using them, as they are representations in numeric form that are at best roughly estimating athletic performance by living animals. They are useful, and do have some degree of predictive value. If you want to dive into that statement, I really do suggest exploring on your own the many volumes (supplemented with a 40 year indoor blizzard of academic papers) published on the topic, beginning in the early 1970s. I'd rather use the time for THAT to scrub behind the fridge and vacuum the heating ducts than re-invent that whole wheel here.

But having said they are useful, that's not the same thing as consistently accurate and equally so in all circumstances, and they aren't. And any use of them that imputes an illusion of precision to them is a gross misuse of the tool, as well as just plain foolish. I could expand on your criticism and add to it quite a bit. Consider the effect of trip, and pace for example, on final time. And what about bias across the surface at a given time? Aqueduct is sometimes notorious for getting a "golden rail" that acts like a rocket sled for inside speed at times, while for decades the old Hollywood Park main track had an infamous "dead rail" that had an effect similar to running in a tar pit for those riders incompetent and unobservant enough to tuck their mounts in near it (as many still did) while those taken out to mid track just glided along relatively effortlessly. It is a multi-variant living analog environment, making it impossible to parse out the exact role of general overall surface condition with any great degree of precision in arriving at the quantitative endpoint of digital final time.

Leaving raw times un-adjusted is a decision to wear a blindfold on one eye and a pink lens on the other, because the sight of that grey squishy stuff over there induces headaches, so pretending or ignoring is just easier. But making good quality figures (I use to do my own for myself years ago before all the published stuff that exists now) properly involves more than was described. Purely mechanistic semi-automated figures will tend to be nearly as useful as slot machine seduction and dice influencing schemes. Done well, par times SHOULD be only a starting point in a process of arriving at variants to adjust raw times. The individual animals and their specific history at the particular point in their life-cycle is considered, among other things, and not just for the one crossing the wire first but for all or nearly all the 100+ running at all the points of call with their many internal fractional splits, and contextual judgments are made, if one is doing a real quality job of it. They will work better or less well in some circumstances than others for comparing changes in conditions and venues. Deciding which is what and when and and how much utility a figure does or doesn't have involves judgment rooted in experience. It isn't too unusual for a set of figures applied to a day's racecard to be reasonably good, and then totally not, becoming utter crap after, say, 2:32 pm, when humidity and wind conditions changed. It is all about estimating probabilities in conditions of incomplete and uncertain information, and that's not just for brewing speed figures.

That's a lot a words of "on the one hand, and the other hand" but before I shut up on it I do want to include one less mealy-mouthed straightforward unqualified thing about speed figures. A few things everyone needs to teach the little ones before they are old enough to box an exacta are: Don't squat upstream from where you drink, don't eat yellow snow, don't p*ss into the wind, and for freakin' Dogsakes don't pay the slightest bit of attention to any supposed speed figures based on final times of races on turf; to the extent they can sometimes be useful, speed figures are appropriate for dirt racing only.
Quote: lilredrooster

Mr. Drawing Dead: Thank you for answering my question. Since you were so kind to do so I would like to impose upon you with another question in order to solicit your opinion which I regard as being an expert opinion based upon the depth of knowledge about horse racing that you continually show.


Being by nature a suspicious person I have always questioned the value and legitimacy of speed figures both
Beyer's and those from Equibase. The reason: In comparing the average (or par) times of races at the same
level at the same track to final times of today's races they then assign a track variant (relative slowness or
fastness of the track) from which each horse 's speed figure is then derived. The flaw in my opinion: To do
this they are using data from only 9 or 10 races, however many the track had on its card. Come on: 9 races
or what a mathematician would call trials or resolutions!!! Gee whiz, when pro blackjack players test a strategy
against a new rule change they will run a computer simulation that gives results based on one BILLION hands.
This is relatively easy with today's powerful computers. So blackjack experts need a billion hands to come to
a conclusion but speed figure sellers only need 9 or 10? I have a sneaking suspicion that if a mathematician
were to analyze the way speed figures are calculated that he would call it voodoo mathematics. In writing this
and asking this question I acknowledge the fact that I may be wrong or missing something. Thanks for your
help if you are willing to answer this question for me. If you are not or don't have time that's fine, I totally
can understand that. I also realize that there has to be some way to account for the drastically different
final times at different tracks. If I remember correctly Golden Gate use to have 15,000 claimers running
6 furlongs at 1:08 and change more than good enough to win stakes races as we move eastward.
So it has to be in some way accounted for. But speed figures as they now are: I just don't know.

Last edited by: DrawingDead on Mar 13, 2016
"I'm against stuff like crack and math" --AxelWold
lilredrooster
lilredrooster
Joined: May 8, 2015
  • Threads: 148
  • Posts: 2501
March 13th, 2016 at 6:03:07 AM permalink
Thanks Dead. Relevant to what we were talking about you pointed that Destin set the track record for Tampa Bay Downs at this distance.
He beat Outwork by one length which means Outwork was only about .2 seconds off the record himself.
Hmmm. Two 3 year olds in the same race and one sets the track record and one almost equals it.
I'm aware Tampa is not in the same league as Churchill or Saratoga but still.
I don't think I have the brain power to figure out the relative importance of this.
"that ball hit his hands. that's a foul ball. the hands are part of the bat"........... "O.K. you think the hands are part of the bat then go down to Dick's Sporting Goods and buy a bat with hands attached." former MLB umpire Jim Evans
DrawingDead
DrawingDead
Joined: Jun 13, 2014
  • Threads: 9
  • Posts: 2233
March 13th, 2016 at 6:35:40 AM permalink
Quote: lilredrooster

Thanks Dead. Relevant to what we were talking about you pointed that Destin set the track record for Tampa Bay Downs at this distance.
He beat Outwork by one length which means Outwork was only about .2 seconds off the record himself.
Hmmm. Two 3 year olds in the same race and one sets the track record and one almost equals it.
I'm aware Tampa is not in the same league as Churchill or Saratoga but still.
I don't think I have the brain power to figure out the relative importance of this.

I'm positively sure you've got more than enough to spare. Maybe not enough sleep if you're responding to me at this hour, but I think you kid if you pretend you can't crush this one over the left field wall. Same thing occurred to me as soon as I heard the announcement "track record time."

A) What were formerly two relatively immature young adolescent animals which their human connections may hope to develop into something that can eventually compete outside of 3 y/o age restricted Grade 2 company at a second tier track, suddenly and miraculously became epic superhorses capable of running rings around the very best fully mature late four year old and five year old veteran Grade 1 racehorses, and by coincidence it just happens that two of them went through this same astounding transformation at the same instant in the same race;

-or-

B) They actually did not magically go from the equivalent of two pimply faced 17 year-old high school stars to suddenly waking up on Saturday as a pair of 30 year-old big-league all-stars, but instead the track superintendent's folks went out with their equipment and watered and rolled it and had their track tight as hell when they ran their big race of the whole year at Tampa, and a large adjustment or other allowance in some form needs to be made when looking at that.

Pick one.


EDIT to add: Andy Beyer just put up his number for Destin's TAM Derby race this morning: a 100 for his time of 1:42.82 for 8.5 furlongs. And at Santa Anita Danzig Candy's time of 1:43.04 in the San Felipe at the same 1-1/16 mi. distance got an identical Beyer figure of 100. And now good night, or morning, or something.
Last edited by: DrawingDead on Mar 13, 2016
"I'm against stuff like crack and math" --AxelWold
Keeneone
Keeneone
Joined: Aug 16, 2014
  • Threads: 13
  • Posts: 996
March 13th, 2016 at 7:33:39 PM permalink
Quote: DrawingDead

-snip-
EDIT to add: Andy Beyer just put up his number for Destin's TAM Derby race this morning: a 100 for his time of 1:42.82 for 8.5 furlongs. And at Santa Anita Danzig Candy's time of 1:43.04 in the San Felipe at the same 1-1/16 mi. distance got an identical Beyer figure of 100. And now good night, or morning, or something.


-editing mine-

Thanks for the detailed explanations, it is very interesting reading.
----------

3rd pool Kentucky Derby Future wager now closed:
https://www.kentuckyderby.com/wager/future-wager/derby-pool-3
Mohaymen favored @ 7/2. Field @ 5/1.
At 12/1 yesterday I was about to plunk down a little money on "all others", but suspected it might come down. It dropped all day, so I did not wager in this pool.
----------

DD, if you please...Do these prospects show any potential in their AWDs?
Cupid (Rebel prospect)
Flexibility (Wood Memorial/Blue Grass prospect)
Siding Spring (Rebel prospect)
Cherry Wine (Rebel prospect)
DrawingDead
DrawingDead
Joined: Jun 13, 2014
  • Threads: 9
  • Posts: 2233
March 13th, 2016 at 8:30:05 PM permalink
Good that you edit down when quoting some of mine, so the server for the site doesn't suffer a seizure from character overload.
Quote: Keeneone

...<SNIP>...

DD, if you please...Do these prospects show any potential in their AWDs?
Cupid (Rebel prospect)
Flexibility (Wood Memorial/Blue Grass prospect)
Siding Spring (Rebel prospect)
Cherry Wine (Rebel prospect)


I get:

  • 7.3f Cherry Wine* (Paddy O'Prado [El Prado] - Unbridled's Song)

  • 7.2f Cupid (Tapit - Beau Genius)

  • 7.1f Flexibility (Bluegrass Cat - Vindication)

  • 6.9f Siding Spring* (Warrior's Reward [Medaglia d'Oro] - Peace Rules)

*NOTE RE: Cherry Wine & Siding Spring - Both Paddy O'Prado & Warrior's Reward are young sires with limited direct data on their progeny. My calculation of the pedigree AWD used a diminished weighting of the sire & was supplemented with a partial weighting of data on the progeny of the paternal grandsire.

I should also note that the pedigree of Cherry Wine is a bit grassy, with significant turf influence. That's not a bad thing for the main track surface at Churchill, if he ends up there.
"I'm against stuff like crack and math" --AxelWold
Keeneone
Keeneone
Joined: Aug 16, 2014
  • Threads: 13
  • Posts: 996
March 14th, 2016 at 12:44:03 PM permalink
Thanks DD. I almost bet on Cherry Wine before the Fountain Of Youth @ Gulfstream but he did not enter. I think he may have spiked a temp or something. He is interesting with the turf influence but he must do well in his next to get some Derby points.
Flexibility already has 15 Derby points and has run some solid races (notably against Moyhaymen). His last was uninspiring (Withers), but he has been given a little time off for his next start. Training good and available @ around 85/1 in the futures.
DrawingDead
DrawingDead
Joined: Jun 13, 2014
  • Threads: 9
  • Posts: 2233
March 14th, 2016 at 4:14:39 PM permalink
Quote: Keeneone

Thanks DD. I almost bet on Cherry Wine before the Fountain Of Youth @ Gulfstream but he did not enter. I think he may have spiked a temp or something. He is interesting with the turf influence but he must do well in his next to get some Derby points.

...<SNIP>...

If he's fit and healthy but doesn't end up making it into this, pretty good chance we'll see him running in something at Churchill that weekend. Trained by Dale Romans, a Louisville native and that's his home base.
"I'm against stuff like crack and math" --AxelWold

  • Jump to: