Go Back   Pace and Cap - Sartin Methodology & The Match Up > RDSS > RDSS Info, Reference
Mark Forums Read
Google Site Search Get RDSS Sartin Library RDSS FAQs Conduct Register Site FAQ Members List Search Today's Posts

RDSS Info, Reference How to obtain RDSS2, get started and learn to use it. (Video and other tutorials)

Reply
 
Thread Tools Display Modes
Old 09-07-2022, 12:56 PM   #21
Bill Lyster
Grade 1
 
Bill Lyster's Avatar
 
Join Date: Feb 2006
Location: Escondido CA just 25 minutes from where the turf meets the surf - "...at Del Mar"
Posts: 2,418
Mitch:

Wouldn't a horse that only beat six horses out of 65 get a 10% rating - [1 - (65-6)/65] = 9 or 10% rating (9.23 actual)?
Bill Lyster is offline   Reply With Quote
Old 09-07-2022, 01:18 PM   #22
Ted Craven
Grade 1
 
Ted Craven's Avatar
 
Join Date: Jun 2005
Location: Nanaimo, British Columbia, Canada
Posts: 8,853
Quote:
Originally Posted by Bill Lyster View Post
Mitch:

Wouldn't a horse that only beat six horses out of 65 get a 10% rating - [1 - (65-6)/65] = 9 or 10% rating (9.23 actual)?
RDSS uses the formula: 100 - 6/65*100 = 100 - 9.23 = 91% rounded.

A horse with 6th fastest workout time out of 65 horses that day and distance worked faster than 91% of the rest.

Then again, a horse who finished 1 out of 11 also gets about a 91%. Of these 2 different horses - or the same horse on 2 different days - if they both worked over 5 furlongs but 6/65 got a speed rating (measure of final workout time) of 80 (slower time) and 1/11 got a SR of 90 (faster time) -- which indicates the better effort? The 6/65 effort worked faster than 59 other horses, while the 1/11 worked faster than only 10 others. Does it matter?

I ask because this is a discussion Mitch and I have been having and I am wondering if the relative presentation of such figures can be tweaked. This readout has been in RDSS for years and Mitch is presenting new ways of using it to measure improvement, particularly in Maiden and longer layoff scenarios.

Ted
__________________

R
DSS -
Racing Decision Support System™

Last edited by Ted Craven; 09-07-2022 at 01:23 PM.
Ted Craven is offline   Reply With Quote
Old 09-07-2022, 02:06 PM   #23
Mitch44
Grade 1
 
Join Date: Nov 2014
Location: The Villages, Fl.
Posts: 3,705
No Bill it's not. There are a couple different ways of doing this.
In simple terms:
a. 65 horses
b. highest value is 100
c. 100 divided by 65= 1.538 value of each horse
d. value 1.538 times (X) number of horses beat which is 59 = 90.742 or rounded is 91

the actual value is 1.538461538461538 X 59 = 90.7969 rounded is 91 I actually use this as it saves a move cutting it down for simple understanding. Note different calculators will vary very slightly which is insignificant.

Another way of doing it this and simpler is: 65 horses and it beat 59, therefore 59 divided by 65 = 90.769 or rounded = 91
Note this one is a must it more than 100 horses worked. I.e. 6/130

Mitch44
Mitch44 is offline   Reply With Quote
Old 09-07-2022, 02:46 PM   #24
Mitch44
Grade 1
 
Join Date: Nov 2014
Location: The Villages, Fl.
Posts: 3,705
The Speed rating using time is insignificant as it obtains more faults. Because the times are different may be a result of the track that day or the variant and or may have been accomplished on a different and slower track for that workout. None of these conditions, variant or track are considered, in a Speed Rating based on time. Better than nothing but fraught with faults.

The % of horses beaten is better tool because it eliminates those two problems, as it takes place on the same track under same conditions of variant, at similar period of time that day etc.

Now a horse can get as Ted states a same rating of 91 in his example of (1/11 & 6/65).
Are they equal? Absolutely not because the 6/65 workout horse competed against 65 horses, therefore the degree of difficulty is much greater to accomplish a 91 than against only 11 horses. It's an unknown what the 1/11 horse could have accomplished up against 65 horses. May have finished 11 or whatever. WE can't determine that, a complete unknown however there's no doubt it's a much harder task or degree of difficulty.

One way to solve this problem is analysis and putting these % rating into content. Is this horse entered today at the proper class, distance and surface. Yes, some work but easily accomplished by perusing the PP;'s If either are entered wrong their 91 rating is meaningless. Note: there are exceptions to this that I can think of, but they are an exception not the norm.

I'm sure there's a formula that can compensate for those differences however it's too much for my little brain. The 6/65 should get additional points for degree of difficulty but how many? Anyone have an answer? without a definitive scientific answer its better left alone or as is.

Mitch44

Last edited by Mitch44; 09-07-2022 at 03:02 PM.
Mitch44 is offline   Reply With Quote
Old 09-07-2022, 03:06 PM   #25
Bill Lyster
Grade 1
 
Bill Lyster's Avatar
 
Join Date: Feb 2006
Location: Escondido CA just 25 minutes from where the turf meets the surf - "...at Del Mar"
Posts: 2,418
Yeah but Mitch's example was "A horse that beats only 6 horses would get 50% and earn nothing", it was not a horse that had the 6th FASTEST time, it was a horse who had the 59th WORST time. Your example was for a horse with the 6th fastest time, and for that example, I agree with your math.

another point is that Mitch's example was based on a horse that finished 6th, not a horse that beat only 6 of 65. I get it now, was confused earlier.


Quote:
Originally Posted by Ted Craven View Post
RDSS uses the formula: 100 - 6/65*100 = 100 - 9.23 = 91% rounded.

A horse with 6th fastest workout time out of 65 horses that day and distance worked faster than 91% of the rest.

Then again, a horse who finished 1 out of 11 also gets about a 91%. Of these 2 different horses - or the same horse on 2 different days - if they both worked over 5 furlongs but 6/65 got a speed rating (measure of final workout time) of 80 (slower time) and 1/11 got a SR of 90 (faster time) -- which indicates the better effort? The 6/65 effort worked faster than 59 other horses, while the 1/11 worked faster than only 10 others. Does it matter?

I ask because this is a discussion Mitch and I have been having and I am wondering if the relative presentation of such figures can be tweaked. This readout has been in RDSS for years and Mitch is presenting new ways of using it to measure improvement, particularly in Maiden and longer layoff scenarios.

Ted

Last edited by Bill Lyster; 09-07-2022 at 03:12 PM.
Bill Lyster is offline   Reply With Quote
Old 09-07-2022, 03:36 PM   #26
Bill Lyster
Grade 1
 
Bill Lyster's Avatar
 
Join Date: Feb 2006
Location: Escondido CA just 25 minutes from where the turf meets the surf - "...at Del Mar"
Posts: 2,418
My second issue:

"Now a horse can get as Ted states a same rating of 91 in his example of (1/11 & 6/65)."

Well, I agree that if you are comparing horses on different days or distances that you should give more credit to the horse that "ran" against the greater number of horses. However, if 1/11 is meant as an approximate mathemical reduction of 6/65 that is one issue, I do not accept that both horses generated a 91% rating.

If at another day and/or distance a particular horse was best of 11, or 1/11, would not that horse for that set of circumstances get a 100% rating? So you could have the 6th of 65 horse (91%) running against a horse that was first of 11 (100%) and perhaps need to develop a way to separate them (?)

Then almost every day we see horses that work the more common distance of 4 furlongs, than work 5 furlongs, or even 6 or more furlongs, so balancing a 6/65 horse at four furlongs versus say a horse who worked one of the longer distances becomes even harder.

I know you guys are working out the bugs in this feature, but is there anything to suggest that horses that work longer distances should be upgraded when facing horses that predominantly work shorter distances.

I'd be happy to help evaluate any current avenues of thought on this subject if someone gave me some parameters to work with.

Not to put a damper on this, but all of this might go out the window if trying to make sense of how Chad Brown works his mighty stable. It seems that he rarely works them more than 4 f no matter what.
Bill Lyster is offline   Reply With Quote
Old 09-07-2022, 03:39 PM   #27
Ted Craven
Grade 1
 
Ted Craven's Avatar
 
Join Date: Jun 2005
Location: Nanaimo, British Columbia, Canada
Posts: 8,853
Quote:
Originally Posted by Mitch44 View Post
I'm sure there's a formula that can compensate for those differences however it's too much for my little brain. The 6/65 should get additional points for degree of difficulty but how many? Anyone have an answer? without a definitive scientific answer its better left alone or as is.

Mitch44

While working on and testing such a formula, anecdotally and statistically, since we agree the greater the number of horses beaten the more impressive the Rank% is - we can start with some kind of 'shading/augmentation' of the Rank% number with a markup like we use for other Workout Patterns, for example, when a Rank% reaches 80% or more, when it worked faster than say, 20 other horses it gets a + (80+) and when it worked faster than 50 other horses it gets a ! (80!). At least with this, your eye is drawn to competing 80s (or 90s, etc) and can elevate and further investigate.

Ted
__________________

R
DSS -
Racing Decision Support System™
Ted Craven is offline   Reply With Quote
Old 09-07-2022, 03:42 PM   #28
Ted Craven
Grade 1
 
Ted Craven's Avatar
 
Join Date: Jun 2005
Location: Nanaimo, British Columbia, Canada
Posts: 8,853
Quote:
Originally Posted by Bill Lyster View Post
My second issue:

"Now a horse can get as Ted states a same rating of 91 in his example of (1/11 & 6/65)."

Well, I agree that if you are comparing horses on different days or distances that you should give more credit to the horse that "ran" against the greater number of horses. However, if 1/11 is meant as an approximate mathemical reduction of 6/65 that is one issue, I do not accept that both horses generated a 91% rating.

If at another day and/or distance a particular horse was best of 11, or 1/11, would not that horse for that set of circumstances get a 100% rating? So you could have the 6th of 65 horse (91%) running against a horse that was first of 11 (100%) and perhaps need to develop a way to separate them (?)

Then almost every day we see horses that work the more common distance of 4 furlongs, than work 5 furlongs, or even 6 or more furlongs, so balancing a 6/65 horse at four furlongs versus say a horse who worked one of the longer distances becomes even harder.

I know you guys are working out the bugs in this feature, but is there anything to suggest that horses that work longer distances should be upgraded when facing horses that predominantly work shorter distances.

I'd be happy to help evaluate any current avenues of thought on this subject if someone gave me some parameters to work with.

Not to put a damper on this, but all of this might go out the window if trying to make sense of how Chad Brown works his mighty stable. It seems that he rarely works them more than 4 f no matter what.
Yes to upgrading horses working 5f+ - in progress . Bill, you are in the testing group, so when the next update is ready for testing - have at it, your reports and thoughts welcomed!

And - all this Improvement/Fitness analysis should naturally be sensitive to one's knowledge of Trainer workout patterns, as well as other ratings from actual races which horses may have (e.g. CSR, CR, CR+, PL, PP etc) as well as CHANGES which might be occurring (equipment, Trainer, Jockey, etc). Keeps the puzzle multi-dimensional and frustrates easy answers!

Ted
__________________

R
DSS -
Racing Decision Support System™

Last edited by Ted Craven; 09-07-2022 at 03:45 PM.
Ted Craven is offline   Reply With Quote
Old 09-07-2022, 04:10 PM   #29
ranchwest
Grade 1
 
Join Date: Mar 2010
Posts: 3,489
My logic says that a horse with a bullet work should have a rank pct of 100, having worked faster than 100% of the horses with slower times.

So, my formula is 100 * ( ( nCount - nRank ) / ( nCount - 1 ) )

Of course, programmatically, an exception must be included where nCount is 1 because it is not legal to divide by zero. I then count the score as 100 for 1st of 1, but the argument could be made for that score being 0.

I am sure some will disagree with me, but that's okay, too.
ranchwest is offline   Reply With Quote
Old 09-07-2022, 04:35 PM   #30
Mitch44
Grade 1
 
Join Date: Nov 2014
Location: The Villages, Fl.
Posts: 3,705
Bill,
The horse itself doesn't count and can't get 100 because he's not counted as a beaten horse. It's based on how many the horse beats, and it can't beat itself.

As far as Chad Brown perhaps that why he loses sometimes. This does not capture every situation, there are exceptions to everything. As Sartin said there are no rules. Our guidelines aren't going to win every race or situation however it is a big improvement.

There are trainers that never work their horses fast. They are the exception not the norm of most trainers and some work their horse away from the track. Every facet of horseracing has an exception however some have more than others. We can't worry over the un-getable. To a certain extent form belongs to the connections and even more so in MDN. races. There are also times that where smarter than them and know their best distance, surface etc. We must capture the majority; nothing is perfect particularity in horseracing.

Mitch44
Mitch44 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Original Energy Program Lord John Classic Sartin Programs - Support, Discussion 0 01-18-2013 12:10 AM
energy program problems amydlarz Classic Sartin Programs - Support, Discussion 2 02-23-2011 11:27 PM
program to a different computer shrink1 RDSS 2 07-21-2009 12:39 PM
C++ manual entry program smilingtiger General Discussion 4 06-24-2009 05:38 PM


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.

All times are GMT -4. The time now is 08:27 AM.