PDA

View Full Version : Par Times


Milleruszk
05-26-2003, 08:00 PM
Where can I get par times for Belmont Park? Thanks!

Brian Flewwelling
05-27-2003, 01:46 AM
I am not clear on the meaning of Par Times, and my reading on this board has not helped.

Pars refer to "Classes" of races, right? That is there would be a Par for Clm20000 and a different one for Stks100000, etc. Thus there need be n different pars for n different classes.

When we take distance into account, there needs to be a Par Time for each Class, at each distance, say d distances.

Tracks, like Main Dirt , Inner Turf ... lead to s surfaces

So Par Times require n.d.s Pars... quite a few IMHO.

Alternatively, if we had workable Speed Figures, a Par Speed for a class would apply to all distances and surfaces, so we would need only n different Pars.

Can anyone clarify this for me?

Brian

GameTheory
05-27-2003, 02:58 AM
Well, for handicapping purposes, a speed figure par is ok.

The problem is, to make speed figures in the first place, you need those n pars for s surfaces at d distances, etc.

Here's some material:

http://www.turfpedia.com/Speed%20Handicap/pars.html

(Also look at the links for Quirin & Beyer speed figures.)

hurrikane
05-27-2003, 07:06 AM
Miller,

check up with Dave Swartz...I think he makes them.
Also Gordon Pine..not sure of the website.

I would suggest making your own. Then you have something noone else has.
You may also be crazy as a loon when your done. :D

Milleruszk
05-27-2003, 12:31 PM
Hurricane,

You are correct.....it is a task that can make you go crazy! My wife suggested that I look into buying them before I go completely over the edge. The process of retreiving the data and posting to a spread sheet is very tedious. Thanks,

Tom

Lefty
05-27-2003, 12:46 PM
Dave Schwartz makes the best pars around, in my opinion. Don't think Gordon Pine makes them anymore as his software Netcapper does not use them and he no longer makes them for Cyntha Pub.
Contact Dave at
www.horsestreet.com

Brian Flewwelling
05-27-2003, 12:55 PM
Tom

I have all the times from Equibase for the past 4years for most tracks. If you were to explain the track differences (Inner, outer) and the classes you want, i will make a table of par times for you.

In fact, that is an open offer to anyone ... if you regularly handicap a track/circuit, I would be pleased to create Par times tables for your track.

For myself, I want some experienced 'cappers to 'approve' my methods, so i will be generous to the first couple working each circuit.

Brian

sjk
05-27-2003, 02:01 PM
Don't know how the formatting will come out, but here are some Belmont par times:

SURFACE DISTANCE 1ST 2ND 3RD 4TH 5TH
DIRT 6 FURLONGS 22.51 45.88 58.15 70.96
DIRT 6 1/2 FURLONGS 22.61 46 71.09 77.71
DIRT 7 FURLONGS 22.82 46.16 71.13 84.19
DIRT 1 MILE 23.19 46.5 71.58 84.63 97.72
DIRT 1 1/16 MILES 23.46 46.8 71.77 97.61 104.32
DIRT 1 1/8 MILES 23.65 47.06 71.94 97.78 111.11
TURF 7 FURLONGS 23 46.13 70.53 83.16
TURF 1 MILE 23.28 46.69 71.23 83.65 96.12
TURF 1 1/16 MILES 24.23 48 72.43 97.07 103.37
TURF 1 1/8 MILES 24.36 48.73 73.19 97.92 110.23

Buddha
05-27-2003, 02:07 PM
Originally posted by Brian Flewwelling
Tom

I have all the times from Equibase for the past 4years for most tracks. If you were to explain the track differences (Inner, outer) and the classes you want, i will make a table of par times for you.

In fact, that is an open offer to anyone ... if you regularly handicap a track/circuit, I would be pleased to create Par times tables for your track.

For myself, I want some experienced 'cappers to 'approve' my methods, so i will be generous to the first couple working each circuit.

Brian

I would be interested in some Par times for Mountaineer. I play there frequently, and have never tried to make a par time for any of the races. If you Brian, or anyone else could help, that would be appreciated.

sjk
05-27-2003, 02:13 PM
For MNR:

SURFACE DISTANCE 1ST 2ND 3RD 4TH 5TH
DIRT 5 FURLONGS 22.14 45.68 51.94 58.26
DIRT 5 1/2 FURLONGS 22.26 45.76 58.28 64.88
DIRT 6 FURLONGS 22.21 45.43 58.03 71.04
DIRT 1 MILE 23.79 47.36 72.6 85.96 99.23
DIRT 1 MILE AND 70 YARDS 22.93 46.85 72.3 99.1 103.55
DIRT 1 1/16 MILES 23.71 47.63 72.78 99.16 106.04
DIRT 1 1/8 MILES 23.88 48.2 73.16 99.51 112.92
DIRT 1 3/16 MILES 23.83 48.16 73.31 99.9 119.88

Dave Schwartz
05-27-2003, 02:50 PM
SJK,

MNR
===
What class level do these pars represent?

Your pars are almost 2 seconds faster at most distances than mine!


BEL
===
Your pars are mostly about 1 tick faster, except for the routes which are some slower/some faster.

If you need some help in making pars, give a call. I'd be happy to spend a little time with you.


Regards,
Dave Schwartz

sjk
05-27-2003, 02:55 PM
They represent an 80 (Beyer scale) level. I have been using them for years with good success (I play Mnr much more often that Bel) , but I thank you for your offer.

I think the absolute time would depend on what length of time you have looked at the track. I see that Mnr has been on average 15 points slower the past two years that it was the previous three years. That wouldn't explain 2 seconds, but would explain a part of it.

Buddha
05-27-2003, 03:06 PM
another bad thing about mountaineer times that i have noticed is when they resurface the track in late sept, early oct, the track tends to run about 2 seconds slower for sprints. they were running 1:13-1:16 6F, and now seem to be around 1:10-1:11. it changes with the seasons i guess :)

hurrikane
05-27-2003, 03:29 PM
sjk
Am I missing something here?

Are your pars just for the distance without any consideration for the quality of horses running?

sjk
05-27-2003, 03:54 PM
As I understand and try to apply the theory which others have developed and written about, there are 3 interrelated sets of values you need to make speed figures and so forth. You need the par times for a fixed (80) class of horses, you need condition pars for each class of race run (these would be in relation to the 80) and you need daily track variants. Given any two of the three, you can try to calculate the third, making your information self-consistent.

The times I posted related to the first bit of information above. For times of other class horses, you would adjust based on the difference of the class par from 80.

As Dave pointed out to me above, it is possible to have a different perception of the absolute speed to the track (I think based on the time-frame of your data). From that standpoint, the times above are only useful in making distance comparisons. I think I am OK as long as my values are self-consistent.

kenwoodall
05-27-2003, 04:33 PM
Sometimes I am a contrarian thinker so please humor me as I would like a serious answer to the following__ How relevant would par times of the 2nd place horse be to accurate average speed comparisons?

Storm Cadet
05-27-2003, 04:51 PM
I've used Bris speed ratings for a couple of years and on the top of each sheet they place the pars average pace/speed ratings for the leader/winnner of races at todays class level/distance E1-E2-Late-Final...they seem to projest well as as long as the track surface is fast.

hurrikane
05-27-2003, 05:25 PM
gottcha ...thanks

Dave Schwartz
05-27-2003, 06:41 PM
Ken,

That's not contrarian. That is good thinking! By making a par of the top 2 finishers you overcome two of the major obstacles in making pars:

1. The horse that wins by 15 lengths because he is really 3 levels better than the race he is entered in.

2. Small samples at some distances (because you have twice as many horses).

Dave

Milleruszk
05-27-2003, 07:57 PM
SJK,

Just a few questions regarding your times from Belmont.
What class level recorded these times? How many races do they represent? Are the Turf times from the Widener or the inner course? Thanks for help!

Tom

Zaf
05-27-2003, 08:03 PM
Does anyone have a set of par times for Turf at Belmont, Hollywood & Churchill that they would be willing to share?

Merci Mille Fois

ZAFONIC

sjk
05-27-2003, 09:23 PM
Milleruszk

The pars are from 1700 or so races over the last 4 years. I don't play turf races so I may have given you garbage for the turf pars. If they run the same distance on both courses that would certainly make a mess of the turf numbers so please ignore.

The times are adjusted to an 80 Beyer scale number. I show the 14000 claimer running at 82.

kgonzales
05-27-2003, 10:38 PM
Originally posted by sjk
Milleruszk

The pars are from 1700 or so races over the last 4 years. I don't play turf races so I may have given you garbage for the turf pars. If they run the same distance on both courses that would certainly make a mess of the turf numbers so please ignore.

The times are adjusted to an 80 Beyer scale number. I show the 14000 claimer running at 82.

Does the $14000 claimer refer to turf races? I've never seen a $14000 claiming race on the turf at NYRA. I think the lowest claiming price they card on grass is $3500.

Amazin
05-27-2003, 11:52 PM
SJK

Nice of you to put up those numbers,but they're far from complete.For example,You show times for what I assume are older male horses somewhere in the 12.500 class.But most races are run by other classifications of horses.Age:2YO's,young and old 3yo's,newly turned 4yo's run different than established older horses.What about fillies and mares?You have no par for them,and their respective age classifications.What about Maidens and their respective class and and age classifications.And the the whole point of pars to me is to be able to make your own track variant in order to gain an edge over the betting public.You certainly can't do that with the limited information you provided.

lousycapper
05-28-2003, 12:38 AM
The 10th @ Pim [05-17-03]? The Citgo Dixie Stks. My top 3 horses were #5, #8 & #10. #10 had the highest SR-TV of all the starters. Just curious to see what your numbers show.

-L.C.

Shacopate
05-28-2003, 01:18 AM
Here are some homemade par times.

Adjusted to an 80 Beyer speed figure that can be used to build your own chart.

Sprints

6f. 45.60 111.00
6 1/2 117.30
7 124.10

One turn mile

111.20 137.00

Routes

1 1/16th 112.60 144.40
1 1/8th 151.20

I've found that pace figures are often much higher than speed figures, which is why I don't allow the extra tenth per half furlong that most do.

Speed Figure
05-28-2003, 02:44 AM
wrong post sorry!

jotb
05-28-2003, 08:11 AM
Originally posted by sjk
Don't know how the formatting will come out, but here are some Belmont par times:

SURFACE DISTANCE 1ST 2ND 3RD 4TH 5TH
DIRT 6 FURLONGS 22.51 45.88 58.15 70.96
DIRT 6 1/2 FURLONGS 22.61 46 71.09 77.71
DIRT 7 FURLONGS 22.82 46.16 71.13 84.19
DIRT 1 MILE 23.19 46.5 71.58 84.63 97.72
DIRT 1 1/16 MILES 23.46 46.8 71.77 97.61 104.32
DIRT 1 1/8 MILES 23.65 47.06 71.94 97.78 111.11
TURF 7 FURLONGS 23 46.13 70.53 83.16
TURF 1 MILE 23.28 46.69 71.23 83.65 96.12
TURF 1 1/16 MILES 24.23 48 72.43 97.07 103.37
TURF 1 1/8 MILES 24.36 48.73 73.19 97.92 110.23


Hello:

I have a question regarding your "pars". Did you use "raw times" or "adjusted times" from your sample? I'm assuming that you have created "condition pars" on top of the "distance pars" because it would be extremly difficult to calculate an accurate variant each day just working with "distance pars". I found out when creating the "conditions pars" one must first adjust the raw times for a specific timeframe in order to produce an accurate "condition par table". When I calculate my variant each day, I don't have many differences as most variant makers encounter. If I have a 9 race card and all distances were run at 6F and the track was playing faster than par, I believe this should be the same for most (at least 7 ) of the races. I remember reading one of Beyer's books and when he provided a sample day producing a variant, every race seem to fit like a glove. I wish he would have provided a full year of "variant making" because he made it seem easier said than done. JMHO

Best regards,
Joe

andicap
05-28-2003, 09:50 AM
Originally posted by Amazin
SJK

Nice of you to put up those numbers,but they're far from complete.For example,You show times for what I assume are older male horses somewhere in the 12.500 class.But most races are run by other classifications of horses.Age:2YO's,young and old 3yo's,newly turned 4yo's run different than established older horses.What about fillies and mares?You have no par for them,and their respective age classifications.What about Maidens and their respective class and and age classifications.And the the whole point of pars to me is to be able to make your own track variant in order to gain an edge over the betting public.You certainly can't do that with the limited information you provided.

Somewhere on the board recently -- maybe even in this thread, someone posted how to make adjustments in this regard. Brohamer does so in his book as well I believe.

Amazin
05-28-2003, 11:10 AM
I've seen the adjustments regarding par times for other classifications of horses,as an offshoot from an older male claiming race.The problem is that it is too general and the error margin gets greater with 2 and 3yo's.It doesn't make sense for me to do pars half assed(pardon my french)if you are going to do them in the first place.That's why they really are alot of work(i.e.Beyer and Jack Daniels).I prefer to either do them thoroughly or not do them at all.IMO,you really need that extra percentage of accuracy to gain that edge over the betting public.You don't get that with the generalized method.I also think HOW you come up with your pars is very important or you may sacrifice your accuracy.For example do you take samples on off tracks or what do you do with horses winning by several lengths.That is just as important in making accurate pars.It's a bit of an art.

sjk
05-28-2003, 06:10 PM
jotb,

I use adjusted times to make the pars (note that I only play dirt races and have distanced myself from the turf pars above). I have condition pars and daily variants just like you do. In my case the condition pars are the weakest element. I use mainly projections to make variants, so there is a lot of averaging and smoothing of pars in making the daily variant.

amazin,

I don't use 2yo races or 3yo races (unless I have to and never maidens) to make variants so I don't make pars for these conditions. I don't think that my condition pars are particularly accurate on an individual basis (I have nearly 8000 of them). If they are close, that works for my purposes.

Milleruszk
05-28-2003, 08:02 PM
Amazin wrote...............

I've seen the adjustments regarding par times for other classifications of horses,as an offshoot from an older male claiming race.The problem is that it is too general and the error margin gets greater with 2 and 3yo's.It doesn't make sense for me to do pars half assed(pardon my french)if you are going to do them in the first place.That's why they really are alot of work(i.e.Beyer and Jack Daniels).I prefer to either do them thoroughly or not do them at all.IMO,you really need that extra percentage of accuracy to gain that edge over the betting public.You don't get that with the generalized method.I also think HOW you come up with your pars is very important or you may sacrifice your accuracy.For example do you take samples on off tracks or what do you do with horses winning by several lengths.That is just as important in making accurate pars.It's a bit of an art.


Well your post has inspired me to attempt my own pars. Your points are right on. I'm not convinced that the formula used by most books on the subject is correct. ie that if you know the par of 10 claimers you can adjust that par upwards to get pars for the rest of the race classifications. Now some questions. I intend to have different classifications for age and sex. (no joke intended!) Should off tracks be included in the general stats or as a seperate classification? Why wouldn't you use samples of races that are won by open lengths? This could keep me busy for a while! Thanks,

Tom

jotb
05-28-2003, 08:24 PM
Originally posted by Milleruszk
Amazin wrote...............

I've seen the adjustments regarding par times for other classifications of horses,as an offshoot from an older male claiming race.The problem is that it is too general and the error margin gets greater with 2 and 3yo's.It doesn't make sense for me to do pars half assed(pardon my french)if you are going to do them in the first place.That's why they really are alot of work(i.e.Beyer and Jack Daniels).I prefer to either do them thoroughly or not do them at all.IMO,you really need that extra percentage of accuracy to gain that edge over the betting public.You don't get that with the generalized method.I also think HOW you come up with your pars is very important or you may sacrifice your accuracy.For example do you take samples on off tracks or what do you do with horses winning by several lengths.That is just as important in making accurate pars.It's a bit of an art.


Well your post has inspired me to attempt my own pars. Your points are right on. I'm not convinced that the formula used by most books on the subject is correct. ie that if you know the par of 10 claimers you can adjust that par upwards to get pars for the rest of the race classifications. Now some questions. I intend to have different classifications for age and sex. (no joke intended!) Should off tracks be included in the general stats or as a seperate classification? Why wouldn't you use samples of races that are won by open lengths? This could keep me busy for a while! Thanks,

Tom

I would include all races regardless of track condition unless someone can explain to me the difference between a time of 109 flat on a sloopy track vs a time of 109 flat on a fast track when all other factors are the same. (AGE, GENDER, CONDITION, CLASS and VARIANT).

Joe

Show Me the Wire
05-28-2003, 10:10 PM
jotb:

Track composition can make a difference. A mostly sand based track can be faster when wet (sloppy) than when dry. Conversly, clay is faster dry than wet.

Regards,
Show Me the Wire

perception is realty.

Amazin
05-28-2003, 11:20 PM
Milleruszk wrote:

"Should off tracks be included in the general stats or as a seperate classification? Why wouldn't you use samples of races that are won by open lengths?"

This is part of the art of making pars.Since off tracks can produce biased and very unpar times,you have to personally decide if you can make sense out the results or scrap them.If you can make sense out of them then I would use them in the classification.

Horses that win by mega lengths are suspect for pars.This is the main reason you still have to do pars manually rather than with a computer.If a horse goes wire to wire,his/her time will be inflated.If you have a maiden classification and they are all running 1:11,then you get this race out of the same classification running 109 and he won by 12 lenghts wire to wire,you need to either adjust it or throw it out.My general rule is to cut the margin of victory to 2 lengths.The leftover lengths I add to the final time and get 1:11.Another way I verify suspect times is look at the first ,second and third place finishers and note what their time is.Then I look at these horses PP's with the same distance and see if it jives.If it does,I'll use it if not,I'll take another sample.Also when you take samples for pars make use you view the times in hundreths,not fifths because 109.19 is rounded off to 109 in fifths and once again you lose accuracy

Brian Flewwelling
11-27-2003, 02:40 PM
Ken suggested using the Place Horse in determining Pars!

Originally posted by Dave Schwartz
Ken,

That's not contrarian. That is good thinking! By making a par of the top 2 finishers you overcome two of the major obstacles in making pars:

1. The horse that wins by 15 lengths because he is really 3 levels better than the race he is entered in.

2. Small samples at some distances (because you have twice as many horses).

Dave

I was digging around for exactly this!!

Based on comments by MV MCKEE, i started calculating Class Pars for Win, Place, Show ... and using the Median rather than the Average of the Speeds

I am thinking of using the 3 placings to calculate the Daily Track Variant. Has anyone experience with this? Any suggestions?

Like should one only use Show Par if there is a minimum number of horses? Should one Weight the placings differently by giving more weight to the winner than ...?


Fleww

sjk
11-27-2003, 08:44 PM
I use any horse 1st, 2nd, or 3rd or within 8 lengths of winner.

traveler
11-29-2003, 10:38 PM
Unless you are going to make speed or pace figures from your speed/pace pars, what good are they? I have never seen the use for them. A race has an 80 speed par - be it a BRIS #, Beyer# or whatever, and the paceline you choose for each horse has a speed figure of 75, except one has an 80. Or in this same hypothetical race, they all have an 80 and one has an 85, the highest rated horse is the best horse based on the #. What difference does it make what the race par is, the horses are running against each other, not par. If I am missing something please let me know. By the way I am a lurker here and have never posted. Thanks.

Dave Schwartz
11-29-2003, 11:15 PM
Traveler,

The theory would be that horses who have proven they can run to par are better placed. It addresses trainer intent to a great degree.

Oh, and welcome. <G>


Regards,
Dave Schwartz

cj
11-29-2003, 11:17 PM
Here are a couple theories on use of pars...I know the first one works pretty good from experience. The second, in my opinion, is even better.

Theory #1

In a maiden special weight race, if the horses who have run before have equalled or exceeded par, the 1st timers are in deep trouble.

Theory #2

In races in which none of the horses have recently matched par for today's race class, figure handicapping (speed figures) is a very poor method of finding the winner,

Tom
11-30-2003, 12:24 AM
I second CJ's comments.
Also, I use the pars to get an idea what claiming price allowance horses might drop to successfully.

traveler
11-30-2003, 12:28 PM
cj - I have read (Quinn?) your 2nd point that if no horses have recently equalled par, then speed figures are suspect at best for today's race. Again, I don't use them so I defer to your experience, but for the sake of discussion, if the par is 80 - a $10,000 clm. race - and all horses show recent failures against this classs speed figures are not the way to go. However, if all the same horses with the same figures, were entered in a $8,000 race - par of 76 let's say, which is a level they have had recent success at, then speed figures can be used to handicap that race. The horses haven't changed only the race conditions. Not trying to argue, if it works for you great, but I mentally can't buy into it so I
don't use it. I do pass races with too many unknown horses in it - wrong dist. for horse, layoffs, bad recent form, shipping from cheap to better track. If a race has too many of these horses in it, I don't think figures are the way to go, which may be arriving at the same point as you but via a different route.

Tom
11-30-2003, 12:41 PM
I th ink the issue is that none of the horses are really up to today's class leve so you should look for other things besides speed to pick one. A class drop, pace advantage, trainer move, close fit to conditions if there are restrictions, a sharp horse moving up calss.
Trouble with the bottom level claimng ranks is it compresses everything into one class. Aqueddcut has a lot of $3500 claimers on the grounds - they all run for $10,000 because that is the lowest they can.

Figman
11-30-2003, 12:49 PM
They're moving closer to that $3,500 level at Aqueduct as the next month will see $7,500 claimers written by the race secretary for the first time in recent memory in New York!

Also, all horses that finish a race will share in the purse money....not just the top five finishers. The 3rd and 4th place horses will race for one per cent less each and the horses that finish 6th thru last will share equally in that aggregate 2% taken from the 3rd & 4th place horses.

jotb
11-30-2003, 01:42 PM
Originally posted by Tom
I th ink the issue is that none of the horses are really up to today's class leve so you should look for other things besides speed to pick one. A class drop, pace advantage, trainer move, close fit to conditions if there are restrictions, a sharp horse moving up calss.
Trouble with the bottom level claimng ranks is it compresses everything into one class. Aqueddcut has a lot of $3500 claimers on the grounds - they all run for $10,000 because that is the lowest they can.

"Condtion pars" are extremly difficult to create. Grouping is a must because of sample size unless you have 10 years of race results because there are just too many conditions for each distance and class level. You are bound to run into overlapping if one is to seek a class heirarchy. Claiming races if not grouped will cause overlapping (pars). For example, on the Aqueduct inner dirt you will find races with a claiming value from 10k to 75k but the inbetween claiming races are as follows: 12.5, 14, 16, 17.5, 18, 20, 22.5, 25, 27.5, 30, 32.5, 35, 37.5, 40, 45, 50, 55, 60, 65, and 70. (This is for ages 3up and 4up.) Now if we have a par for each claiming value we would end up with about 22 different pars and I can assure you there would be a tremendous amount of overlapping. Even if you incorporate some of these claiming values and broke it down to maybe 6 claiming levels you still will have overlapping. There are several reasons for this. The first and most important reason is the raw times. If you have 25 races at the 10k claiming level, I'm sure you will have a top time of 109:00 and a bottom time at 114:00 from the sample. We already know that either one of these times are close to par for that level. So I believe before you try to create condition pars you must first compare raw times using some type of procedure to calculate your variant and adjust each raw time from there so that you would not encounter this 5 second differential. Most condition par makers believe that it's best to throw out all races unless they are run over a fast track. I don't think this is the answer either because we go back to the problem of sample size.
Here is another example which creates the overlapping effect. You have 2 claiming races on the same day and one is valued at 25k and the other 18k (same gender and same age). The 18k race runs 6F in 111.20 and the 25k runs in 111:60. It should be the other way around but this happens on many occasions. The reason for this is because of dropdowns and a rising in class. It's possible for the 18k race to attract 6 runners who are dropping from 25k down to 18k and the winner from the 25k is rising in class from 18k. All of this must be taken into consideration if one seeks to create accurate condition pars. Even if one can overcome the above you still need to deal with each surface and track. For instance the Aqu inner dirt has claiming races for 10k, 12.5k, 14k and 16k but Belmont does now run a claiming race under 14k. Can we conclude the 14k at Belmont is equal to the 14k's on the inner? The same is true for Saratoga. The bottom claiming level at the Spa is 25k for older horses. Can we say the 25k at Sar is equal to 25k claimers over at Belmont? You will see that at Saraotga those 25k claimer races attrack horses that were running over at Belmont for 18 and 20k. This is another reason why many horses that ship over to Saratoga from Churchill Downs and run at the bottom claiming level 25k are snatched from these races by the NY trainers especially the Churchill horses that are dropping from 30k down to 25k. There make great prospects for racing over the Aqu fall meet and Aqu inner winter meet. The bottom line is creating "condition pars" is tricky business.

Joe

Dave Schwartz
11-30-2003, 03:12 PM
Joe,

Wow! You have sure come a long way since we began talking about pars last year.

What an awesome understanding you have of the "par problem." I look forward to running my New York pars past you before I ship them in January.


Regards,
Dave Schwartz

Tom
11-30-2003, 04:29 PM
Originally posted by Figman
They're moving closer to that $3,500 level at Aqueduct as the next month will see $7,500 claimers written by the race secretary for the first time in recent memory in New York!

Also, all horses that finish a race will share in the purse money....not just the top five finishers. The 3rd and 4th place horses will race for one per cent less each and the horses that finish 6th thru last will share equally in that aggregate 2% taken from the 3rd & 4th place horses.

Sounds like welfare to me. Penalize horses that try to reward slugs.
:(

Tom
11-30-2003, 04:47 PM
Very interesting post. Years ago agoat FL, I was setting up pars and for the $10,000 claiming group, I had 8 sprint races.
the problem was, that of those 8 races, 6 of them were won by the same horse!
What I do now is to list out the Beyer numbers for each class, $20K, $22.5K, $25K, $27.5K, etc, in Excel and put each into cloumns next to each other so I can look at all the data at once.
From thereI will group where it makes sense, but I don't rely too much on median or average anymore, rather I look for a minimum figure for the classs-usuallu a number that will include 75-80% of the races. I break the sample down into quarters, and the low quarter are bad races, the top quarter are excellent races, and the middle two quarters are what to expect.
I operatre under the assumption that there is no variant unless there is compelling evidence to the contrary.
So if a par for a race is 80 and a horse runs 82, I am likely to assume the variant is 0. If all the races for the day (or days-I group days whenver possible, especially turf races) cluster around -2 and +2, I don't bother with a variant.
Pace on the other hand, is another animal. I have a scatter sheet set up with the raw final times converted to beyers for each distance.

45.0 85
45.1 86
45.1 87
45.2 88
45.3 89


When Sim Weekly comes out, I compare the raw pace time to the actual Beyer and use that as my race variant. Actually, I have it set up in the spreadsheet to do it for me-all I do is key in the Beyer each week.
I convert the Beyer to a quirin number, unless a race or two are out of wack, much like CJ does. Then I adjust for those if I think Beyer is wrong (might presumptuous of me!).
I end up with quirin numbers, like 100-103, 98-95, etc.
I find using pace pars based on final time or figs is better than on a time par. Having said that, I do buy Dave's pars and will do so again in January - they are great and I just cannot do this for every track-not an keep my job and sanity.
I find a lot of days with significant pace variants, especially at Aqueduct. I am starting to use the wind data in the Sim Weekly as a reference source for this work and it is making things clearer to some degree. Sometimes a 47 11.1 is a faster early race than a 46 111.1.

Tom
11-30-2003, 04:57 PM
We talked earlier about how to use a par. One way I use them is to evaluate horses stepping up in class.
Say the race is Alw nw2X
Horse A earned an 83 Beyer last time, at this class, which is in the low quarter for this class.
Horse B ran an 82 in a AlwNW1x last out, which is in the high quarter. B is stepping up and earned a slightly lesser number than A. But the hidden thing here is that horses should improve from nw1 to nw2. I expect A to run better today because he is outperforming pars for his normal developemnt cycle.
Lots of time, a horse will be in the top quarters of the MSW, NW1, NW2, and then jump up into stakes, open or conditioned, and these are the kinds I look for to handle the jumps.
Having comparabel figures for FL and AQU, I have founds horses that earned spectacular numbers for the class at FL ship in to NYRA and have so-so numbers in comparison, maybe 75-80 at FL while the AQU horses have 80-85 numbers that are in the low end of the class here. 10 points is not too much to overcome for a sharp horse, especially when the locals are stale, coming off layoffs, at the wrong distances, etc.

Tom
11-30-2003, 05:05 PM
Right after I posted the above, I got this email- it is my daily California "tip" from Speedfigures.com for Hollywood Park.
They are using pretty much the same thing I was talking about:

quote:
"Today's 8th race in an allowance N2X dirt route at the 107 level. For these particular conditions, I like to look at how each runner faired (figures wise) upon clearing the N1X level. Only those horses that earned better than par figures for the 105 level should be considered true contenders at the 107 N2X level. Based on that criteria, only KEWEN (4-1) and MENUHIN (3-1) fit the better than par winning N1X profile."
unquote.

Let's see what happens.

:D

jotb
11-30-2003, 07:21 PM
Originally posted by Dave Schwartz
Joe,

Wow! You have sure come a long way since we began talking about pars last year.

What an awesome understanding you have of the "par problem." I look forward to running my New York pars past you before I ship them in January.


Regards,
Dave Schwartz


Study Study Study and then Study some more.....It can be quite nerve racking but you try to cover all bases when creating "condition pars" but it seems everytime you think all is well another obstacle stands in front of you. I've read tons of material on the creation of "condition pars" and for the most part the authors seem to make it an easy job but my conclusion is, they leave out many missing pieces of the puzzle.

Best regards,
Joe

cj
11-30-2003, 07:55 PM
Joe,

No disagreements on anything you said. When playing the races, I was talking about speed figure pars though. Raw time pars mean absolutely nothing to me. They are valuable to those making the speed figures, but when evaluating horses in today's race, I want the par Performance Figure, not a raw time.

VetScratch
12-01-2003, 10:08 AM
I agree with Joe and others... but we should also recognize that "sports medicine" is changing the claiming game.

The classic view of class ladders in the claiming ranks has become distorted primarily because risk/reward dynamics have changed. Of course, circuit variances, grossly inflated racino purses, and lucrative state-bred incentives to beat open company do influence the way horses are managed. However, advances in legal and illegal sports medicine have also impacted horse management.

It is a lot easier to convince owners to run them where they belong (i.e., to risk losing their horses) when purse/tag ratios are favorable. A concurrent influence is that owners are more inclined to accept risk when horse replacement carries less risk.

After you lose a horse, it is comforting to know that you can probably claim off paper with less risk than ever before. Clenbuterol, EPO, joint injections, hoof-repair treatments, and a host of other "tools" make it more likely than ever that a talented trainer can help you wiggle out of a bad claim.

Brian Flewwelling
12-01-2003, 12:21 PM
Originally posted by sjk
I use any horse 1st, 2nd, or 3rd or within 8 lengths of winner.

I take it you use a Projection Method rather than a Class based system for your Pars. Am i correct?

But if a horse romps to a 15 length lead, you then ignore the remaining horses?

Fleww

sjk
12-01-2003, 02:12 PM
Brian,

I take the horses qualified as I mentioned above (1st, 2nd or 3rd or within 8 lengths; this includes the horse 15 back if he was 2nd or 3rd) and then look at any races (same surface) within 60 days (before or after) in which they met the same qualifications. Then I calculate a daily track variant based on the difference between today's unadjusted speed rating and the speed rating in the other qualifying race.

I also calculate the variant based on the par for the track and condition, although I only use that variant for the winner and as a bound for the other calculation (I limit the projected variant to be no more than 20 points above or below this par variant).

I take these and then average by runner and then by date, track, surface and track condition.

You may be sorry you asked, and I'm sure no one else is still reading, but there it is.

highnote
06-10-2014, 12:02 AM
Brian,

I take the horses qualified as I mentioned above (1st, 2nd or 3rd or within 8 lengths; this includes the horse 15 back if he was 2nd or 3rd) and then look at any races (same surface) within 60 days (before or after) in which they met the same qualifications. Then I calculate a daily track variant based on the difference between today's unadjusted speed rating and the speed rating in the other qualifying race.

I also calculate the variant based on the par for the track and condition, although I only use that variant for the winner and as a bound for the other calculation (I limit the projected variant to be no more than 20 points above or below this par variant).

I take these and then average by runner and then by date, track, surface and track condition.

You may be sorry you asked, and I'm sure no one else is still reading, but there it is.

It's 11 years later and I'm still reading. :D

Found this post while doing a search engine query on Belmont Park par times. :)

sjk
06-10-2014, 03:58 AM
And I am still doing it the same way.

SK

highnote
06-10-2014, 11:46 AM
And I am still doing it the same way.

SK

:ThmbUp:

If it ain't broke....

Harvhorse
06-10-2014, 05:37 PM
yOU DONT MENTION WHICH CLASS THES PARS ARE FOR. aRE THEY 12.5 CLAIMERS, A1Z STK. oNE PAR CANT BE VALID FOR ALL CLASSES.

Milleruszk
06-10-2014, 07:57 PM
This thread proves that nothing vanishes from the internet. :eek:

sjk
06-11-2014, 06:06 PM
I think it is funny as hell and probably no one else is still reading.

highnote
06-11-2014, 09:33 PM
This thread proves that nothing vanishes from the internet. :eek:


Sometimes that is a good thing -- especially when it is classic information! :ThmbUp: