Pace and Cap  - Sartin Methodology & The Match Up

Pace and Cap - Sartin Methodology & The Match Up (http://paceandcap.com/forums/index.php)
-   Previous 'Handicapping Discussion' Forum (http://paceandcap.com/forums/forumdisplay.php?f=9)
-   -   selecting contenders (http://paceandcap.com/forums/showthread.php?t=8101)

Bill V. 01-28-2012 01:40 PM

Phase 1
 
1 Attachment(s)
Hi
here is phase 1 TPR
I agree with Jeff using a horses best ever line is risky
The basic pace line guidelines are to avoid the best ever unless its the last line. In the Sam Huston Race 6 while picking running styles, I basically called
the whole field NF or off too many days for a cheap claiming race were current form is vital

In this case I decided to use Phase 1. I used the last line for every horse
except for horse 1. I had to use line 3 because my phase 1 program will
not rate 5.0 lines and line 1-2 were 5.0 lines
Anyway
The 1 and 4 would be my bet as far and away the best TPR horses
Attachment 25886
GS
Bill

shoeless 01-28-2012 02:16 PM

Hi Bill,

Thanks for sharing and good to hear from you

Jeff

Dorianmode 01-28-2012 02:46 PM

Decision process example..from this morning
 
First of all, I AM NO EXPERT ON ANYTHING. I have only recently returned to racing, so PLEASE, feel free to tell me I'm nuts. You'll be in good company.


Anyway, I thought I would use today as a concrete example of my "system" so it can be understood, (and criticized....)


Here we go.


My initial search for races came up with 3 race candidates, all within a few minutes of 11:30 this morning, the 1st races at Aqueduct, Phili, and Gulf Stream. That's too many to do. So which to keep ? I toss the Phili race because it looks to be too contentious, and I respect the ML/odds/fan base in Florida and New York more.


(I thought it would not be possible to do BOTH the remaining races, but there WAS enough time between them, so I did them both). Since my preference WAS going to be the turf route at GS, because it looked like my choices were going to be going off at better odds, I will go through my logic process for that first).


1,2,3,5 are out by NP elimination, 4 is out because he's 20/1 in the ML, 40/1 win at GS just never happens, and 50/1 in the TwinSpires data base is a bad sign. The 9 was a coming off a maiden win, so I was skeptical of him, and the excellent fan base at GS didn't seem to like him very much. That left 6,7, and 8 The 8 looked very good just by old fashioned pace handicapping, so the choice came down to 8, and either 6, or 7. The 6 is on top of RDSS, so that had to be the 2nd choice in the dutch. The dutch payoff estimate was 0.76, which is ok for me, so I bet it. I also had a few minutes to check out the exotic tables, so decided to try those too. The result was the 8 won, ($12.80), the 2 placed, (($35.20 (2$) exacta)), and the 7 showed, ($139.80)(2$) trifecta. So far so good.


Before I hopped to Florida, I did the #1 race at Aqueduct. It was a little easier.


2,6,8 are out by NP elimination. (If they were to win, it would mean, that what would have to happen, is that one of them would have to jump out of it's NP range, AND, simultaneously, the 1,3,4,7 would ALL, simultaneously, have to run below their lower NP ranges, and the odds on THAT happening, are astronomical).......So I'm left with 1,3,4,5 and 7 to look at. 5 is out... just plain out, because of poor performance, it's also out by APV/Class. (Actually, it looks like the 3 wasn't really going to be competitive either, but I left it in just because the NP range was "close enough" to take a look in RDSS.). The 1 is out because the Aqueduct fan base would never let a good horse go off at 40/1, so I'm left with 3, 4, and 7. The 3 has never really been "in the hunt" since it's maiden win, so even before I look at RDSS, I have only 2 left, the 4 and the 7. The dutch payoff estimate was only 0.534, which is about my minimum, but I thought I'd try it anyway, just because I knew I was going to be posting this example race.

Result was :

https://mail.google.com/mail/?ui=2&i...8fffe6572b1&zw





Have a great day at the races ! :p

shoeless 01-28-2012 02:57 PM

1 Attachment(s)
Pino and I yesterday were talking about selecting contenders for races and
using the best of the last 3.Here is a horse from Sat at Parx in the 3rd race.

Race was CLM 6 FUR 7500N2Y.

I was really surprised it paid 7.80 even though there were dropdowns in
the race.


By the way I took Jim G sound advice from aways back and looked at
pacelines in a quiet setting

Dorianmode 01-28-2012 03:34 PM

Saturday 1/28 13:00 (1:00PM), (Central Time) races ..OP#1, PRX#5
 
OP #1.... 3,4,5,6,8 out by NP. Leaves 1,2,7. Someting is wrong with #1. Leaves 2 and 7. Dutch does not work...pass. (2 wins, 7 places).

Phili #5....3,5,6,7,8 out by NP elimination. Leaves 1,2,4,9,10. 1 not competitive. The rest, 2,4,9,10 are ok in APV and class rank. 10 out by TE, and too much early energy for this track's route model. 2,4,9 remain. 2 is top of RDSS, and must stay in dutch. ? = which for second in dutch. 4 or 9. Very tough choice. In TPR, (RDSS) screen 9 is better in LP, so THAT was the choice. Dutch .726 is OK. Played. 2 won, 9 placed.

shoeless 01-28-2012 04:49 PM

I find that your posts are very good and at least you explain how you came about getting your contenders and selections.

Dorianmode 01-28-2012 04:58 PM

One more, just to clarify the race selection process. At about 14:30, (2:30 CT), Saturday, 1/28, there are 3 race possibilities coming up. Phili 8, Laurel 7, Turfway 6.
Laurel 7, from NP looks very contentious.

So I looked at Phili 8. The number 4 was a clearly outstanding animal, and it was pretty obvious what was going to happen, so I did a single place bet, didn't wait to watch the race, (he won), and went on to Turfway 6.

This one was a little odd, because, 4,5,6,7 are out by NP. Leaves 1,2 and 8. 1 is on top of RDSS, BUT the top 3 have #1's in RDSS all over the place. I thought of checking out the All Ways readout in Brisnet, but didn't have time. So what to do ? Back to good old pace handicapping. #8 looked pretty good by eyeballing the pace numbers. #1 barely won in December here, so I decided to use the 2 and the 8. Dutch payoff was a no-go, (at .351). So had to pass. Didn't have time to look at the exotics. The 8 led all the way, (#1 in EPR), and won.

Dorianmode 01-28-2012 08:30 PM

Another, somewhat "quirky" race. #8 Santa Anita Saturday, 1/28/12.
6.5 F $40K Claimer.

8 is out, (30/1), 6 can't possibly win. 3 and 2 are out by NP. Leaves 1,4,5,7. I fired up my RDSS 1.0 to more carefully look at the pace of the race. 4 looks very dangerous, but runs too fast too early. Went back to NP, took out the 4, and the three left need a very careful look. TE looks OK for all. Then I noticed that the 7 was going off at 7/1, and was thinking of doing a one horse bet because it was #1 in RDSS AND #1 in LRP/DCL. But, since my second choice, the 5 was #2 in LPR/DCL was also going off at 7/1, it made a nice dutch, so I went with that.

I did have time to look at the exotics, and they were very nice ... and ta da...it was a nice way to end the day.

7 won.. $15.60
1 placed, 5 showed.

$2.00 exacta box : $95.40-$12.00 = $83.40 net
$2.00 trifecta box : $369.20- $24.00 = $325.00 net

shoeless 01-28-2012 10:03 PM

DM,

Very nice way to end the day and thanks for the explanation of you doing
the races .

pktruckdriver 01-28-2012 10:13 PM

Wow , Guess I missed some fun, well at least I can read about it, thanks guys

patick

mikesal57 01-29-2012 08:39 AM

"Just returned to racing" and " am no expert"

You sure fooled me!!!

Nice going and hope you continue your analysis

Mike

P.S.- Welcome back

shoeless 01-29-2012 09:35 AM

Maybe NP works better as DM is doing it as a way to sort out non contenders rather then a black box type of thing.


Also TPR numbers as Bill V showed could be used the same way

Dorianmode 01-29-2012 02:03 PM

Oops...the trifecta box net...was not as stated. I better learn how to add and subrtact. $2.00 trifecta box : $369.20- $24.00 = $325.00 net ...Not. Should be $345.20 net.

Dorianmode 01-29-2012 02:36 PM

Also a "procedural" addendum to my "method" note....


For every race I do, I keep a 3x5 index card.
On it, I keep my decision process, for that race, so I can review the bad outcomes.

(Someone said once that the important races today are not the ones you win, but the ones you loose..to learn from)

...anyway...I keep the cards to put in a few piles...

(actually you can do this process electronically, in a spread sheet, very easily also),...and I do that too, but

a. eliminated the winner, from my contenders
b. eliminated the winner , from my top 4
c. eliminated the winner from my top 3
e. eliminated the place from my place selections
f. if doing the trifecta, eliminated the show horse
g. race was not amenable to my mental powers/capacity...or subject to "evaluation" at all

(Now since I selected the race for evaluation from a group of possibles, the "g" had better be a damn small pile.)


But as you get better, the "a" pile will become smaller and smaller, then the "b", then the "c"..you can track your progress.


When I get to them all going into the "h" pile...I guess I will have the game beat.

shoeless 01-29-2012 04:58 PM

I have to say you really do a complete job of cappin the races

Bill Lyster 01-30-2012 03:45 PM

I think you meant to say Race 7 from SA on this date, because 3-4-7 was the finish order in race 8?

Quote:

Originally Posted by Dorianmode (Post 77565)
Another, somewhat "quirky" race. #8 Santa Anita Saturday, 1/28/12.
6.5 F $40K Claimer.

8 is out, (30/1), 6 can't possibly win. 3 and 2 are out by NP. Leaves 1,4,5,7. I fired up my RDSS 1.0 to more carefully look at the pace of the race. 4 looks very dangerous, but runs too fast too early. Went back to NP, took out the 4, and the three left need a very careful look. TE looks OK for all. Then I noticed that the 7 was going off at 7/1, and was thinking of doing a one horse bet because it was #1 in RDSS AND #1 in LRP/DCL. But, since my second choice, the 5 was #2 in LPR/DCL was also going off at 7/1, it made a nice dutch, so I went with that.

I did have time to look at the exotics, and they were very nice ... and ta da...it was a nice way to end the day.

7 won.. $15.60
1 placed, 5 showed.

$2.00 exacta box : $95.40-$12.00 = $83.40 net
$2.00 trifecta box : $369.20- $24.00 = $325.00 net


Bill Lyster 01-30-2012 04:08 PM

2 Attachment(s)
"Originally Posted by Dorianmode
Another, somewhat "quirky" race. #8 Santa Anita Saturday, 1/28/12.
6.5 F $40K Claimer.

8 is out, (30/1), 6 can't possibly win. 3 and 2 are out by NP. Leaves 1,4,5,7. I fired up my RDSS 1.0 to more carefully look at the pace of the race. 4 looks very dangerous, but runs too fast too early. Went back to NP, took out the 4, and the three left need a very careful look. TE looks OK for all. Then I noticed that the 7 was going off at 7/1, and was thinking of doing a one horse bet because it was #1 in RDSS AND #1 in LRP/DCL. But, since my second choice, the 5 was #2 in LPR/DCL was also going off at 7/1, it made a nice dutch, so I went with that.

I did have time to look at the exotics, and they were very nice ... and ta da...it was a nice way to end the day.

7 won.. $15.60
1 placed, 5 showed.

$2.00 exacta box : $95.40-$12.00 = $83.40 net
$2.00 trifecta box : $369.20- $24.00 = $325.00 net"

Okay, I am officially confused (and perhaps very anal as you will see).

The 8 in race 8 is 20/1 ML and went off at over 30/1, but the race in question really looks like race #7 (at least as to the results). But the 8 in race 7 was ML at 8/1 and went off a shade under 14/1, and one could argue that the 4-2-8 are too close to call for a clear toss out (trackmaster NP version). I see that the 6 should not be considered.

One question needs to be asked here is what version of NP do you use or do you use and cross check both versions? The first version being the Trackmaster mode, the second being the Adjusted Speed Rating mode? There often are differences. Which do you use?

Dorianmode 01-30-2012 04:10 PM

Thanks for the the correction Bill
 
<TABLE class=raceHeader><TBODY><TR><!-- Column Widths: currently totaling 648, which is equal to the main table width --><TD style="WIDTH: 81px">You are absolutely correct.
Actually I'm not in a position to check right now, but it looks like I'm using the Trackmaster version.



http://www1.drf.com/images/entriesResults/Race-7.jpg



</TD><TD class=postOffTime></TD><TD class=surfaceDescription>Surface: Dirt </TD><TD class=getPPs>GET PPS: This Racecard | Next Card: 01/29 </TD></TR><TR><TD colSpan=4><HR style="HEIGHT: 1px" class=divider></TD></TR><TR><TD class=distanceDescription colSpan=4></TD></TR><TR><TD class=distanceDescription colSpan=4>6 1/2 Furlongs | 4 Year Olds And Up | ALLOWANCE OPTIONAL CLAIMING ( $40,000 ) | Open </TD></TR></TBODY></TABLE>
<!-- Runner info --><TABLE class=runner cellSpacing=0 cellPadding=0><!-- Header --><THEAD><TR class=header><!-- Column Widths: currently totaling 648, which is equal to the main table width --><TH style="WIDTH: 20px" id=headerProgramNumber># </TH><TH style="WIDTH: 192px">HORSE </TH><TH style="WIDTH: 192px" class=coloredBackground>JOCKEY </TH><TH style="WIDTH: 72px">WIN </TH><TH style="WIDTH: 72px" class=coloredBackground>PLACE </TH><TH style="WIDTH: 72px">SHOW </TH></TR></THEAD><!-- Loop over each runner --><TBODY><TR class=runner><!-- Choose the saddle cloth text color --><TD style="BACKGROUND-IMAGE: url(images/entriesResults/7.jpg); COLOR: #000000" id=programNumber>7 </TD><TD class=horseName>Sahara Sky </TD><TD class=coloredBackground>C. Nakatani <TD>$15.60 </TD><TD class=coloredBackground>$7.00 </TD><TD>$4.00 </TD></TR><TR class=runner><!-- Choose the saddle cloth text color --><TD style="BACKGROUND-IMAGE: url(images/entriesResults/1.jpg); COLOR: #ffffff" id=programNumber>1 </TD><TD class=horseName>Scofield Barracks </TD><TD class=coloredBackground>M. Smith <TD></TD><TD class=coloredBackground>$7.40 </TD><TD>$3.80 </TD></TR><TR class=runner><!-- Choose the saddle cloth text color --><TD style="BACKGROUND-IMAGE: url(images/entriesResults/5.jpg); COLOR: #ffffff" id=programNumber>5 </TD><TD class=horseName>Runflatout </TD><TD class=coloredBackground>J. Rosario <TD></TD><TD class=coloredBackground></TD><TD>$2.80 </TD></TR></TBODY></TABLE>
<!-- Fractions -->Times in 5ths: :21 :43<SUP>1</SUP> 1:08<SUP>1</SUP> 1:14<SUP>2</SUP>
Times in 100ths: :21.11 :43.33 1:08.28 1:14.50
<!-- Winning Info --><!-- Using runner[0] because they are order in the stored procedure by official position -->Winning Trainer: Hollendorfer Jerry | Owner: Hollendorfer, Jerry and Sweetwater Stable

<!-- Scratch Info --><!-- Loop over each scratch --><!-- Claim info --><!-- Loop over each claim --><!-- Display -->Claimed: Jim's No Preacher by Asmussen Steven M. (Robison)

<!-- Also ran info --><!-- Loop over each also ran --><!-- Display --><TABLE id=alsoran class=others><TBODY><TR><TD>Unplaced horses listed in order of finish. </TD></TR><TR><TD>Also ran: Privilaged, Arraignment, Jim's No Preacher, Thinking of Girls and Gator Warrior </TD></TR></TBODY></TABLE><!-- Payoff info --><!-- Loop over each payoff --><!-- c:set var="tempPayoffString" value="$2 ${payoff.wagerName} (${payoff.winningNumbers}) ${payoff.numberOfRights} Correct Paid: ${payoffAmountCurrency} ${payoff.wagerName} pool ${totalPoolAmountCurrency}"/--><!-- c:set var="tempPayoffString" value="$2 ${payoff.wagerName} (${payoff.winningNumbers}) ${payoff.numberOfRights} Correct Paid: ${payoffAmountCurrency} ${payoff.wagerName} pool ${totalPoolAmountCurrency}"/--><!-- c:set var="tempPayoffString" value="$2 ${payoff.wagerName} (${payoff.winningNumbers}) ${payoff.numberOfRights} Correct Paid: ${payoffAmountCurrency} ${payoff.wagerName} pool ${totalPoolAmountCurrency}"/--><!-- c:set var="tempPayoffString" value="$2 ${payoff.wagerName} (${payoff.winningNumbers}) ${payoff.numberOfRights} Correct Paid: ${payoffAmountCurrency} ${payoff.wagerName} pool ${totalPoolAmountCurrency}"/--><!-- c:set var="tempPayoffString" value="$2 ${payoff.wagerName} (${payoff.winningNumbers}) ${payoff.numberOfRights} Correct Paid: ${payoffAmountCurrency} ${payoff.wagerName} pool ${totalPoolAmountCurrency}"/--><!-- Display --><TABLE id=payoff class=others><TBODY><TR><TD>$1 Exacta (7-1) Paid: $47.70, $1 Trifecta (7-1-5) Paid: $184.60, $1 Superfecta (7-1-5-8) Paid: $1,501.10, $2 Daily Double (5-7) Paid: $117.60 Daily Double Pool $29,916.00 , $1 Pick 3 (5-5-7) 3 Correct Paid: $272.50 Pick 3 Pool $63,411.00 </TD></TR></TBODY></TABLE>

cdax 01-30-2012 06:58 PM

Great thread
 
I'm really enjoying reading these posts and how DM is making his eliminations. I just may need to cut and paste some posts and start practicing this strategy.

Keep up the good work.

Chris

shoeless 01-30-2012 07:21 PM

Im hoping others will chime in with how they go about eliminating non contenders, DM does have some good ideas and thanks for sharing.

Bill Lyster 01-30-2012 07:22 PM

PRECAUTIONARY NOTE WHEN USING New Pace INFO
 
2 Attachment(s)
The New Pace screen in RDSS2 comes with two options.

When you click the Gear icon, it will show you two choices - Trackmaster SR and Adjusted SR.

If, on the last race you worked, you chose Adjusted SR, when you open another race and "It looks like its set to Adjusted SR" but sometimes its not. (The TM SR info will show even though the little button says Adj SR.)

So when you get to the next race and hit the "gear" icon, toggle back and forth a time or two until you can see the difference between TM and Adj SR info. You will see that it is quite different from one mode to another.

Bill Lyster 01-30-2012 07:58 PM

TM vs Adj SR results
 
Last summer when I did a 500+ race sample of NP Ted asked me to also test Adj SR and Total Energy using the same principles of bracketing the 90% range of actual speed ratings (SRs).

In my sample NP and Adj SR had the same number of winners (but different winners) but the ROI expressed as money won in excess of money bet was 18% for NP and 21% for Adj SR. This was based on $1 win bets on the 4 horses designated as either E1, E2, L1 or L2. Neither system provided positive exacta results, both losing about $0.50 to $.60 cents on the dollar wagered.

shoeless 01-30-2012 08:28 PM

Bill,

Thanks for the info




As what I was attempting to do was use Pizzolla's idea of a Fulcrum to weed
out non contenders.I was pretty good at doing so but my biggest problem
was trying to figure out on who to bet.

Dorianmode 01-30-2012 08:46 PM

Who to bet on....???
 
Me too. I can get the winner 97-99 % of the time in the last 3. But that 's no help if you leave out the winner during the last step. THAT is my big problem. AM working on a decision model for eliminating the last horse, OR if things don't stack up right, passing the race, (or requiring a higher payoff for certain situations), or in certain situations not plaing the win pool, but only the exotics, depending on the pay-off tables. I know I HAVE to also start looking at pick 6's and pick 4's and daily doubles.

Thanks for the data on the two speed rating systems. Seems I have, either an older, or different RDSS 2.0..as I do see the gear icon, but not the choice of rating systems.

Ted Craven 01-30-2012 10:53 PM

Quote:

Originally Posted by Dorianmode (Post 77656)
Thanks for the data on the two speed rating systems. Seems I have, either an older, or different RDSS 2.0..as I do see the gear icon, but not the choice of rating systems.

You do have the most recent RDSS2. Just click on the Config Icon (the 'gear') and the NP Configuration settings will pop-up for you to choose among SR options. Everywhere in the program where you see the 'gear' icon (i.e. a Configuration Icon) clicking on it brings up some kind of relevant dialog box for modifying the settings.

Ted

BJennet 01-31-2012 12:30 AM

Your method seems random
 
Quote:

Originally Posted by Dorianmode (Post 77536)
Hi Mike, .....um.....not exactly. The standard deviations, (NP) of the 3, the 6, and the 7 tell me that it is 95% probable that they will run WITHIN their NP range. It is highly unlikely, (only 5%...statistically speaking), that one of them will, all of a sudden, "jump" out of that range in that race, and even if they did, it wouldn't be good enough to win). I don't really use NP for "handicapping", only for elimination-efficiency. The remaining 4 I look at, in their RDSS screens. Does that make sense ?

Hi DM,

I have to admit that most of what you've written is this thread doesn't seem to hold together. First, the 95% confidence interval is a fairly useless way to measure the value of a horse's performances, if you're only looking at speed figs, since the range is so wide that it does nothing to separate the performances of the contenders, which is all that really matters. Also, you seem not to understand Dave's reasoning in weighting low variance (aka 'good form') as heavily as he has - especially in cheap races, such as this one, where it can be as valuable as a speed rating in evaluating win probability. Personally, I think Dave has weighted it too heavily, but the low-variance factor can produce high-price winners, precisely because the crowd wieights speed figs so heavily.

Your elimination method basically seems to involve throwing out all the non-NP horses and then checking the remainder vs. RDSS. Why you or anyone else would do this mystifies me, since NP is a much weaker handicapping factor, with an average hit-rate of .18. The top two choices hit RDSS (either TE or V/DC) hit 60% of winners, whereas the top four of NP (which offers no rank-order) hit only 74%. To put it another way 97% of NP winners are already contained within RDSS. So, if you are going to go to the trouble of handicapping with RDSS, why bother to 'confirm' it with NP? To apply your elimination standard, RDSS eliminates only 15% of winners, whereas NP eliminates 26%.

Cheers,

B Jennet

Dorianmode 01-31-2012 01:30 AM

1/30 Good evening, Mr. Jennet
 
Thanks for posting this. If you... one of the leading lights here, have these questions, it makes me think I have done a very poor job explaining myself.

"the 95% confidence interval is a fairly useless way to measure the value of a horse's performances, if you're only looking at speed figs"

Correct. I don't just look at speed figures. One dimensional handicapping is not for me. I use NP to make SOME easy eliminations. The range may be wide, (and often is), but statistically, a horse with a significantly different range than another, requires that BOTH, (ie the probability is multiplied).. have to run outside their (95% SD) range, (if something "different" is going to happen). I only use it to help identify the non-contenders. I DO understand Mr. Schwatrz's reasoning, it does not appeal to me, entirely, using it the way he does. The reason I suggested there needs to be the calculation of a Confidence Interval, and a Neyman Construction, is to complete the probability picture, (as I said in one of my first posts. I care MORE about my win percentage, and frequent wins, than I do about high priced, less frequent wins.

Your elimination method basically seems to involve throwing out all the non-NP horses and then checking the remainder vs. RDSS.

No. I do not even look at whether a horse is a "NP" horse. I only look at the ranges, and the total range picture, to get a feel for the competitive nature of the entire field. Many times NP has as one or more of it's horses, an animal that could not possibly win the race. I would NEVER bet a dime on many of them.

"Why you or anyone else would do this mystifies me, since NP is a much weaker handicapping factor, with an average hit-rate of .18. The top two choices hitRDSS (either TE or V/DC) hit 60% of winners, whereas the top four of NP (which offers no rank-order) hit only 74%."

Exactly. That's why I NEVER use NP to pick winners. After I have made the eliminations, I ONLY use RDSS. The fewer contenders I have, (3 or 4 at the most), the clearer the RDSS picture is.

"To put it another way 97% of NP winners are already contained within RDSS. So, if you are going to go to the trouble of handicapping with RDSS, why bother to 'confirm' it with NP?"

Exactly backwards. I do not use NP to confirm anything. I use RDSS to pick winners, AFTER I have looked at a number of things to pick the the contenders, and eliminate the others, (including NP). I do not use it the way Mr. Schwartz suggests, for the reasons I have written about above. It's statistical validity, as far as it goes, makes sense, and I'm sure he will improve it. For now, all I know is, it help me eliminate non-contenders, quickly and consistently. :)

BJennet 01-31-2012 03:33 AM

Still doesn't make sense
 
Quote:

Originally Posted by Dorianmode (Post 77659)
Thanks for posting this. If you... one of the leading lights here, have these questions, it makes me think I have done a very poor job explaining myself.

"the 95% confidence interval is a fairly useless way to measure the value of a horse's performances, if you're only looking at speed figs"

Correct. I don't just look at speed figures. One dimensional handicapping is not for me. I use NP to make SOME easy eliminations. The range may be wide, (and often is), but statistically, a horse with a significantly different range than another, requires that BOTH, (ie the probability is multiplied).. have to run outside their (95% SD) range, (if something "different" is going to happen). I only use it to help identify the non-contenders. I DO understand Mr. Schwatrz's reasoning, it does not appeal to me, entirely, using it the way he does. The reason I suggested there needs to be the calculation of a Confidence Interval, and a Neyman Construction, is to complete the probability picture, (as I said in one of my first posts. I care MORE about my win percentage, and frequent wins, than I do about high priced, less frequent wins.

Your elimination method basically seems to involve throwing out all the non-NP horses and then checking the remainder vs. RDSS.

No. I do not even look at whether a horse is a "NP" horse. I only look at the ranges, and the total range picture, to get a feel for the competitive nature of the entire field. Many times NP has as one or more of it's horses, an animal that could not possibly win the race. I would NEVER bet a dime on many of them.

"Why you or anyone else would do this mystifies me, since NP is a much weaker handicapping factor, with an average hit-rate of .18. The top two choices hitRDSS (either TE or V/DC) hit 60% of winners, whereas the top four of NP (which offers no rank-order) hit only 74%."

Exactly. That's why I NEVER use NP to pick winners. After I have made the eliminations, I ONLY use RDSS. The fewer contenders I have, (3 or 4 at the most), the clearer the RDSS picture is.

"To put it another way 97% of NP winners are already contained within RDSS. So, if you are going to go to the trouble of handicapping with RDSS, why bother to 'confirm' it with NP?"

Exactly backwards. I do not use NP to confirm anything. I use RDSS to pick winners, AFTER I have looked at a number of things to pick the the contenders, and eliminate the others, (including NP). I do not use it the way Mr. Schwartz suggests, for the reasons I have written about above. It's statistical validity, as far as it goes, makes sense, and I'm sure he will improve it. For now, all I know is, it help me eliminate non-contenders, quickly and consistently. :)

DM,

What you say here seems to contradict a number of your handicapping examples in this thread. You open many of the with "a,b,c,d, horses out by NP elimination". So you begin by eliminating non-NP horses, and then cut out one or two more by traditional handicapping criteria (odds too high, bad PP's, long layoff), and then check the NP contenders vs. the RDSS horses. My point is that, if you've already handicapped the race with RDSS, you already have 97% of NP winners plus an additional 10% of winners that NP would have elminated. There's no contest between the two. And because it is rank-ordered, RDSS is much stronger against a broad range of races. So why not just use RDSS?

In addition, you make no response to the value of Dave's weighting of variance (form, form cycle) as a factor, even though you say you understand what he's doing. Also, if you hve the background to do Neyman pseudo-experiments to provide rank-ordering for NP, I think most people here would find it helpful.

Cheers,

B Jennet

Dorianmode 01-31-2012 10:17 AM

1/31
 
"Out by NP" does NOT mean they are "non-NP" horses. I have extensively discussed they are out by their standart deviation "range" .They are eliminated by "poor" ranges", (their standard deviations).

"and then check the NP contenders vs. the RDSS horses"

No. I have already explained, that is not how I use it.

"My point is that if you've already handicapped the race with RDSS"

No, I already stated I do NOT use RDSS until the contenders are eliminated.

"So why not just use RDSS?"

Because my centender selection process, (which is why/how I use NP at all), is more efficient.

"the value of Dave's weighting of variance (form, form cycle) as a factor"

I HAVE discussed my problem with HIS problem with "recency" above, and other problems. When I posted my "method", I said, it was "only for me. If you don't like it, dont use it. I was merely suggesting there may be some usefullness to NP, in looking at it another way than Schwartz intended.

gl45 01-31-2012 02:39 PM

DM,
can you explain in detail what is the standard deviation "RANGE". Perhaps an example
would help me. I'm not very educate in statistic.
Th

rmath 01-31-2012 04:25 PM

Dorianmode
 
Did you get a chance to check you stats on the races where the #1 DCL and LPR were not the same horse?
Just curious.

shoeless 01-31-2012 05:26 PM

B Jennet,

Would you share with us your method of weeding out the non contenders?

shoeless 01-31-2012 08:27 PM

To put my 2 cents in here I would like to commend DM for sharing his way
of contender selection.As he said it works for him and he is not twisting
anybodys arm to use it.

Dorianmode 01-31-2012 09:25 PM

gl45
 
Greetings,
I will try to put together a very brief (but detailed) explanation of what standard deviations are. I will think about it, what to include, what to skip. Meanwhile, there is a nice little set of free tutorials on statistics and useful tools at : http://stattrek.com/online-tutorials/tutorial-list.aspx Most of it is "too much information", and I wish the lessons were numbered.
I'm busy for a few days. But I will work on it.

BTW, does anyone know if there is a study somewhere of which/who's speed ratings are the most reliable, and why ?

BJennet 02-01-2012 01:06 AM

Clarity on NP
 
Quote:

Originally Posted by Dorianmode (Post 77664)
"Out by NP" does NOT mean they are "non-NP" horses. I have extensively discussed they are out by their standart deviation "range" .They are eliminated by "poor" ranges", (their standard deviations).

"and then check the NP contenders vs. the RDSS horses"

No. I have already explained, that is not how I use it.

"My point is that if you've already handicapped the race with RDSS"

No, I already stated I do NOT use RDSS until the contenders are eliminated.

"So why not just use RDSS?"

Because my centender selection process, (which is why/how I use NP at all), is more efficient.

"the value of Dave's weighting of variance (form, form cycle) as a factor"

I HAVE discussed my problem with HIS problem with "recency" above, and other problems. When I posted my "method", I said, it was "only for me. If you don't like it, dont use it. I was merely suggesting there may be some usefullness to NP, in looking at it another way than Schwartz intended.

Hi DM,

I think I'm beginning to get a clearer idea of what you're saying now, but you don't make it easy. What you say about the way you eliminate obvious throwouts and use RDSS is clear, and basically similar to what most handicappers would do. What is not really clear is the criteria you're using to eliminate NP horses. Let's see if I understand what you're doing.

Basically, you're ignoring Dave's designation of NP 'contenders' and using the speed figure range that is part of NP, but applying it to the whole field. As we have noted before, Dave slightly penalizes the horse with high variance (wide range, aka, bad form), and slightly rewards the horse with lower variance (good form). So, if I understand, you only use NP to evaluate form.
If this is not right, maybe you can explain how you use 'range' to make NP eliminations.

BTW, as I've said to others, I'm not asking these questions to criticize you or change your handicapping, but to find out if there's anything in what you do that could be of benefit to others. This site is intended as a community of players trying to help each other to play the game better.

Cheers,

B Jennet

BJennet 02-01-2012 01:18 AM

Standard Sartin
 
Quote:

Originally Posted by shoeless (Post 77676)
B Jennet,

Would you share with us your method of weeding out the non contenders?

Hi Jeff,

My contender selection is basically the same as my paceline selection. As I've written about a few time, I use the lines of the horses with the highest TEs within the last 100 days. This also tends to fit into the standard Sartin measure of 'best of last three'. I do make two basic eliminations beforehand - 'herd horses' - those that often hit the board, but rarely, if ever win, and 'need to lead' horses, that also rarely win (although I will sometimes use the latter in a speed-friendly context). The great advantage of this is that it's fast and completely mechanical - almost no judgment involved. It's worked very well for me.

BTW, I think you'll also do well using V/DC, if you prefer it. Rich M. has produced a relatively mechanical method for contender selection, if you search on his posts.

Cheers,

B Jennet

Dorianmode 02-01-2012 02:37 AM

1/31
 
I think it might be helpful for anyone interested, or who watched it some time ago, to review Schwartz's video #5, from the introductory set of "NewPace-Unlocking the Mystery of Pace". In it he describes his "revelation" about a "Confidence Interval" and the book he read which gave him the initial idea for his system. He talks about the range.

"Basically, you're ignoring Dave's designation of NP 'contenders' and using the speed figure range that is part of NP, but applying it to the whole field"

Yes

"As we have noted before, Dave slightly penalizes the horse with high variance (wide range, aka, bad form), and slightly rewards the horse with lower variance (good form)"

Perhaps. A horse with poor, (relative), raw speed ratings, will obviously have a poor NP "range". A wide range does not necessarily indicate bad form, especially if you are using many speed ratings, and the last few happen to be good, (and consistent).

"So, if I understand, you only use NP to evaluate form."

No. I only consider form, when I get to looking at horses in RDSS.

"If this is not right, maybe you can explain how you use 'range' to make NP eliminations."

Which horse is going to win this race ?

shoeless 02-01-2012 06:37 AM

B Jennet,

Thanks and for your advice as well

shoeless 02-01-2012 11:06 PM

B Jennet,

If not too much trouble would not mind seeing you post an example of
how you go about doing it.

Bill Lyster 02-02-2012 12:01 AM

Dorianmode
 
5 Attachment(s)
Using NP along the way. The first two shots are the entire field using first the TM SR and then the Adj SR. The 3rd shot is all the entrants with my line choices. At this point I eliminated the 6-3 and 8. The 6 and 3 are obvious throw outs on either SR panel but the 8 takes some review.

This race is atypical in my view, because in either view the 1-5 and 7 are standouts, and they did finish 7-1-5 with nice payouts. In most of the races that I have worked since this post and that is more than 100, once you view the Adj SR array, it is very hard to consistently eliminate the correct horses when you have made the eliminations like I have made.

So my question is for races that are more closely contested than this race, what criteria are you using to further eliminate horses so that you betting choices are either the top three or four?

Thanks,

Bill


All times are GMT -4. The time now is 08:54 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2026, vBulletin Solutions Inc.