Go Back   Pace and Cap - Sartin Methodology & The Match Up > Sartin Methodology Handicapping 101 (102 ...) > Previous 'Handicapping Discussion' Forum
Mark Forums Read
Google Site Search Get RDSS Sartin Library RDSS FAQs Conduct Register Site FAQ Members List Search Today's Posts

Previous 'Handicapping Discussion' Forum General Handicapping Discussion

Reply
 
Thread Tools Display Modes
Old 05-28-2012, 12:38 PM   #1
ginag2
AlwNW3X
 
Join Date: May 2012
Posts: 28
spn and ent

Hello all -

I'm new to the forum - old timer with Doc's stuff - having started in the late 70s with Phase 3. Very glad to have found this group of serious folks.

Anyway: I see that spn (Sartin pace number) and ent (entropy) are shown and tracked in the export data of RDSS, but for the life of me - I can't find them in any of the program screens? Am I just missing something - or?

Thanks - Gary
ginag2 is offline   Reply With Quote
Old 05-28-2012, 01:00 PM   #2
Bill V.
The egg man
 
Bill V.'s Avatar
 
Join Date: Jul 2008
Location: Carlsbad, California
Posts: 10,005
Entropy and SPN

Hi Sir
welcome to pace and cap

Funny you asked this question. I happen to have asked this question to Ted in a private email and here is his reply
Maybe I can save Ted some time here


Bill,



It took me a little while to find those calculations ...



PF3 is not SPN and TS+F3 is not Entropy. They are their own numbers: TS+F3 is from the TS+F3 readouts on the Segments screen: Average of True Speed+Fraction 3. PF3 is Projected Fraction 3, from Pizzolla’s PPF (Projected Power Fraction) which is Fraction 3 with 2nd call beaten lengths added back in.



SPN was simply BL/BL+some random number to make it hard for ‘pirates’ to reverse engineer it. You can test this by seeing in RDSS1 or Val4 if there was EVER a difference in ranks of BL/BL and SPN. There should be none (as it is simply the same number, disguised). It was redundant, so I eliminated it. Just like FX in the Supplementary was redundant.



I added DCL (Deceleration) to the BLBL screen (which was not in Val4 – at least not on the surface). DCL = F3 / 2ndCall velocities. Guy Wadsworth first displayed this in Speculator.



Entropy was: the average of DCL + SPN, or – the average of the DCL factor + BL/BL.



Since SPN is redundant to BLBL and DCL is stated separately, I chose to eliminate it. There are numerous combinations of factors similar to what was listed as ‘Entropy’ that people use and have talked about: F1+TS, F2+TS, F1+BLBL, F3+BLBL and they may all have some value – more or less – at certain track/surface/distance combos which will be revealed by modeling them.



Doc told me that there were several such factors left in Validator that were carry-overs from earlier software that he would liked to have eliminated because it was simply ‘too much information’ and thus potentially confusing overload (Fractals, SCBL, Entropy, SPN, even %Median) but that people liked them and wanted them displayed, so he continued displaying them. I guess I only felt comfortable going part-way to Doc’s view on readout simplification, in eliminating the Entropy readout in RDSS2.



This is the first time since RDSS2 Beta appeared over a year ago that anyone has commented on SPN and Entropy. I do have a method for easily creating ‘custom screens’ in RDSS2 – either for horse data or for Analysis data, which people can show or hide by the PP and Analysis Layout Tab. Perhaps some of these custom combinations of factors can appear on such screens for those who want them.



However, I thought you really didn’t use RDSS for analysis, merely for export to DOS programs – so, would these various readouts really matter to you?



yours,



Ted


Hope this helps
Bill V. is offline   Reply With Quote
Old 05-28-2012, 01:27 PM   #3
ginag2
AlwNW3X
 
Join Date: May 2012
Posts: 28
Thanks so much for responding Bill - appreciated.

I guess I'm still a bit 'unfulfilled' - have been modeling the readout results of my recent actual races wagered and because spn showed up as the most valid variable of the lot - I then looked for it on the RDSS readouts.

I guess I am still wondering how (from where?) it shows up on the exported data if it is not included in original readouts? AND - there is definitely a significant difference in the comparison of SPN to BLBL rankings. Whatever is being used - it does not appear to be simply BLBL with a random "fake-out" number added.

Perhaps Ted will clarify.

Thanks much - Gary




















ginag2 is offline   Reply With Quote
Old 05-28-2012, 04:53 PM   #4
Bill V.
The egg man
 
Bill V.'s Avatar
 
Join Date: Jul 2008
Location: Carlsbad, California
Posts: 10,005
Blbl - spn

Hello Gary


I went back and checked the last 20 races I used with Synthesis and Validator
I only used races in which I had 5 contenders.
I then compared the BLBL and the SPN number.

These races were all run between May 1st and May 24th 2012
The tracks used were Belmont CD, Parx, Hollywood Woodbine Delaware
Will Rodgers Downs and Turf Paradise
Sprints, Routes, Dirt, Turf, I used no maiden races.

The BLBL and SPN are very close. There are slight differences.
These usually happen when there are ties in BLBL,
Of the 20 races there was only 1 tie in SPN
When there was a tie in BLBL I noted it with a ( )
I have shaded the races with differences in BLBL and SPN
Name:  BLBLSPNRACES.PNG
Views: 1367
Size:  12.4 KB
The two readouts look comparable to me
Bill
Bill V. is offline   Reply With Quote
Old 05-28-2012, 05:50 PM   #5
Ted Craven
Grade 1
 
Ted Craven's Avatar
 
Join Date: Jun 2005
Location: Nanaimo, British Columbia, Canada
Posts: 8,853
Gary,

Good to see you digging into the models. I agree that there can be a slight difference in rank order between BL/BL and SPN, but never by more than 1 rank. In fact, SPN is a ranking of an intermediate state of BLBL before BLBL is adjusted in some cases to make adjacent values equal depending on the corresponding Primary Line Score value (and that is probably both too much, and unnecessary information). I believe that is precisely what Bill's short chart above is showing.

That said, I would be curious to see, from a model listing of your races so far, whether you have found a more significant difference than Bill in impact of the ranks of winners in your model, comparing BLBL rank to SPN rank. If you feel like either posting a spreadsheet here, or sending it privately to me, I would love to see it before blithely dismissing SPN further as irrelevant (though I indeed think it is).

The reason it appears in the model export is an oversight on my part - it should be removed so as to not cause this very confusion you are reporting. Internally in RDSS, there are numerous linked data structures for each race and horse, and values from these data structures are displayed in multiple places (i.e. on different screens), or on no screens at all in the case of obsolete factors or purely internal calculations. SPN is in the former category.

None the less, please let us know if you believe you have a case for the preservation of SPN ('Sartin Pace Number') and I will definitely give it respectful attention.

Welcome to the neighbourhood, and best wishes!

Ted
__________________

R
DSS -
Racing Decision Support System™
Ted Craven is offline   Reply With Quote
Old 06-04-2012, 09:56 PM   #6
kenr3138
AlwNW2X
 
Join Date: Nov 2008
Posts: 22
SPN & Entropy

Quote:
Originally Posted by Bill V. View Post
Hi Sir
welcome to pace and cap

Funny you asked this question. I happen to have asked this question to Ted in a private email and here is his reply
Maybe I can save Ted some time here


Bill,



It took me a little while to find those calculations ...



PF3 is not SPN and TS+F3 is not Entropy. They are their own numbers: TS+F3 is from the TS+F3 readouts on the Segments screen: Average of True Speed+Fraction 3. PF3 is Projected Fraction 3, from Pizzolla’s PPF (Projected Power Fraction) which is Fraction 3 with 2nd call beaten lengths added back in.



SPN was simply BL/BL+some random number to make it hard for ‘pirates’ to reverse engineer it. You can test this by seeing in RDSS1 or Val4 if there was EVER a difference in ranks of BL/BL and SPN. There should be none (as it is simply the same number, disguised). It was redundant, so I eliminated it. Just like FX in the Supplementary was redundant.



I added DCL (Deceleration) to the BLBL screen (which was not in Val4 – at least not on the surface). DCL = F3 / 2ndCall velocities. Guy Wadsworth first displayed this in Speculator.



Entropy was: the average of DCL + SPN, or – the average of the DCL factor + BL/BL.



Since SPN is redundant to BLBL and DCL is stated separately, I chose to eliminate it. There are numerous combinations of factors similar to what was listed as ‘Entropy’ that people use and have talked about: F1+TS, F2+TS, F1+BLBL, F3+BLBL and they may all have some value – more or less – at certain track/surface/distance combos which will be revealed by modeling them.



Doc told me that there were several such factors left in Validator that were carry-overs from earlier software that he would liked to have eliminated because it was simply ‘too much information’ and thus potentially confusing overload (Fractals, SCBL, Entropy, SPN, even %Median) but that people liked them and wanted them displayed, so he continued displaying them. I guess I only felt comfortable going part-way to Doc’s view on readout simplification, in eliminating the Entropy readout in RDSS2.



This is the first time since RDSS2 Beta appeared over a year ago that anyone has commented on SPN and Entropy. I do have a method for easily creating ‘custom screens’ in RDSS2 – either for horse data or for Analysis data, which people can show or hide by the PP and Analysis Layout Tab. Perhaps some of these custom combinations of factors can appear on such screens for those who want them.



However, I thought you really didn’t use RDSS for analysis, merely for export to DOS programs – so, would these various readouts really matter to you?



yours,



Ted


Hope this helps
Hey Ted:
"I do have a method for easily creating ‘custom screens’ in RDSS2 – either for horse data or for Analysis data, which people can show or hide by the PP and Analysis Layout Tab. Perhaps some of these custom combinations of factors can appear on such screens for those who want them".
I think this is a great ideal for those want them including VDC in the 3rd segment screen.
Thanks,
Ken
kenr3138 is offline   Reply With Quote
Old 07-08-2012, 09:54 PM   #7
ginag2
AlwNW3X
 
Join Date: May 2012
Posts: 28
thanks re spn

Thanks Ted - apologies for the tardy reply.

I have much more data now, and indeed, though spn and blbl differ maybe 10% of the time - it is never more than 1 rank as you stated, and the results after enough data bring the totals/averages very close to each other.

THanks again - and to you others who responded as well. Gary
ginag2 is offline   Reply With Quote
Reply

Tags
ent, spn

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


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

All times are GMT -4. The time now is 07:20 PM.