|
|
04-29-2018, 12:40 PM
|
#16
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,910
|
Quote:
Originally Posted by mikesal57
Question ...Why would you believe that your in "competition" with the other guys..??
Maybe I'm misinterpreting the R stuff....but
Isn't the math based on the factors you created?
Lets use your 4000 factors ( I know you cant use em all at once )
or some other amount....
Logistic regression will give you a correlation of those factors to the winning horses..you take out the ones that dont model winning horses ...and then refine those that are left .....to a point that you can be used and weighed like the Fibonacci approach...but a better predictability...
Mike
|
While there are certainly some factors that are better than others, ultimately, at the 50+ factor level just about everyone is going to have the same data. What separates one person/group from the other is the way they choose to analyze the data.
My opinion -- and it is just that (yours may be different) -- is that if I do what the whales are doing, then I am going to be in direct competition with them.
That won't likely have a good end result for me because: 1. they're better at it
2. they've got more horsepower (in terms of manpower as well as education)
3. they've been doing it successfully for a long time.
So, I choose to do something different. I try to play to the whales' weakness. That would be small, highly specific models. It is simply not what they do.
Just my opinion.
Dave
|
|
|
04-29-2018, 01:07 PM
|
#17
|
Veteran
Join Date: Sep 2003
Location: NEW YORK CITY
Posts: 3,670
|
What do the "Whales" make up of the whole...2-3-4 % ????
And that "weakness" they have....do they have a weakness?
So I say , they are there and you cant do anything about it....
Just beat the other 95% , either you grind it out with low odds or wait for the 2 of 20 hits on double digits ...
Mike
|
|
|
04-29-2018, 02:34 PM
|
#18
|
Veteran
Join Date: Sep 2003
Location: NEW YORK CITY
Posts: 3,670
|
Quote:
Originally Posted by pandy
It's nice that people have tested the Bris Prime Power, but it would be interesting to see how it does using odds ranges and using different rankings. For instance, playing horses that went off at 4-1 or better than had the top ranking, or playing the highest odds horse ranked in the top four Bris Prime Power. Either of these should produce higher ROI's than betting the top ranked pick.
|
JCapper stats for 2017 shows the Prime rank..#1....had a .84 ....for horses paying over $10
All other ranks are too low to even print...yuck
|
|
|
04-29-2018, 07:17 PM
|
#19
|
crusty old guy
Join Date: Aug 2003
Location: Snarkytown USA
Posts: 3,918
|
I think that the gaps between ratings also needs to be tested -- ranks just aren't enough imo. I would think that a race with the top 4 Prime Power ratings of 150, 139, 133, 132 would play differently than one with 139, 138, 136, 134. Understanding the significance of gap size is probably important for all factors. In the program that I am currently working on it seems to be important. I could be wrong though.
Last edited by headhawg; 04-29-2018 at 07:22 PM.
|
|
|
04-29-2018, 07:53 PM
|
#20
|
Veteran
Join Date: Dec 2003
Posts: 7,139
|
Quote:
Originally Posted by GMB@BP
I am only interested in the contenders, I am much better than 31% picking a winner out of 4 or 5 horses.
|
I doubt that. Because my data agrees with the article's assertion that the top 5 Bris PP win approximately 87% of the races.
What it doesn't state is that the average payoff of those winners is $10.
So if you can pick the winner 31% from the top 5 PP horses (which you poo poo as a low ability) you would make a steady 50% ROI which is unheard of.
Because in 10 races, with a 31% hit rate, you would receive $30 on average and your investment would be $20.
You would basically be the best handicapper on the planet.
|
|
|
04-29-2018, 09:25 PM
|
#21
|
Registered User
Join Date: Feb 2003
Location: Dark Side of the Moon
Posts: 5,870
|
Quote:
Originally Posted by Light
I doubt that. Because my data agrees with the article's assertion that the top 5 Bris PP win approximately 87% of the races.
What it doesn't state is that the average payoff of those winners is $10.
So if you can pick the winner 31% from the top 5 PP horses (which you poo poo as a low ability) you would make a steady 50% ROI which is unheard of.
Because in 10 races, with a 31% hit rate, you would receive $30 on average and your investment would be $20.
You would basically be the best handicapper on the planet.
|
I dont believe the 10 dollar mutual, I have tracked about 100 races and average payoff is 6 and change.
And remember I said 2 horses in a 6 horse field, 3 max if within 2 prime points, 4 in 8 and only 5 when greater than 10, and no mdn races.
I think the biggest prices was 8/1, once. its a steady stream of odds on to 7/2....again, using tighter parameters than 5 horses in a 7 horse field thus catching more fliers is much much less likely.
Last edited by GMB@BP; 04-29-2018 at 09:27 PM.
|
|
|
04-30-2018, 07:57 AM
|
#22
|
Veteran
Join Date: Sep 2003
Location: NEW YORK CITY
Posts: 3,670
|
Quote:
Originally Posted by headhawg
I think that the gaps between ratings also needs to be tested -- ranks just aren't enough imo. I would think that a race with the top 4 Prime Power ratings of 150, 139, 133, 132 would play differently than one with 139, 138, 136, 134. Understanding the significance of gap size is probably important for all factors. In the program that I am currently working on it seems to be important. I could be wrong though.
|
For you Brian and my friend Derek....
GAPS.....nothing much showing here..this is from all of 2017
Code:
By: SQL-F30 Gap
>=Min < Max P/L Bet Roi Wins Plays Pct Impact
--------------------------------------------------------------------------------------
-999.0000 -15.0000 -76276.50 200538.00 0.6196 4964 100269 .0495 0.3762
-15.0000 -10.0000 -29339.60 122476.00 0.7604 5093 61238 .0832 0.6320
-10.0000 -5.0000 -21547.60 123814.00 0.8260 8231 61907 .1330 1.0104
-5.0000 0.0000 -15616.10 90512.00 0.8275 8845 45256 .1954 1.4853
0.0000 5.0000 -8584.70 55474.00 0.8452 7497 27737 .2703 2.0541
5.0000 10.0000 -3744.50 24872.00 0.8494 4550 12436 .3659 2.7805
10.0000 15.0000 -900.80 6676.00 0.8651 1528 3338 .4578 3.4788
15.0000 20.0000 -130.00 1350.00 0.9037 375 675 .5556 4.2220
20.0000 25.0000 -37.70 292.00 0.8709 71 146 .4863 3.6957
25.0000 30.0000 -38.60 116.00 0.6672 21 58 .3621 2.7516
30.0000 35.0000 -10.10 44.00 0.7705 8 22 .3636 2.7635
35.0000 40.0000 -2.40 54.00 0.9556 11 27 .4074 3.0961
40.0000 45.0000 -1.80 44.00 0.9591 5 22 .2273 1.7272
45.0000 50.0000 -32.00 32.00 0.0000 0 16 .0000 0.0000
50.0000 55.0000 -23.20 32.00 0.2750 2 16 .1250 0.9499
55.0000 60.0000 -6.70 28.00 0.7607 5 14 .3571 2.7141
60.0000 65.0000 -1.20 26.00 0.9538 4 13 .3077 2.3383
65.0000 70.0000 -16.20 24.00 0.3250 2 12 .1667 1.2666
70.0000 75.0000 -8.00 8.00 0.0000 0 4 .0000 0.0000
75.0000 9999.0000 -3.00 20.00 0.8500 3 10 .3000 2.2799
|
|
|
04-30-2018, 08:13 AM
|
#23
|
crusty old guy
Join Date: Aug 2003
Location: Snarkytown USA
Posts: 3,918
|
Are those gaps between the highest and next highest PP? Between highest and 5th highest? Highest and lowest? I don't speak Jcapper -- how can you have negative gaps? The -999 is probably all data, but I am not sure of the other lines with negative ranges.
|
|
|
04-30-2018, 08:15 AM
|
#24
|
Veteran
Join Date: Sep 2003
Location: NEW YORK CITY
Posts: 3,670
|
Quote:
Originally Posted by headhawg
Are those gaps between the highest and next highest PP? Between highest and 5th highest? Highest and lowest? I don't speak Jcapper -- how can you have negative gaps? The -999 is probably all data, but I am not sure of the other lines with negative ranges.
|
This is for RANK #1 only....gap from 2nd place horse...
JCapper cant do other ranks...
|
|
|
04-30-2018, 02:42 PM
|
#25
|
Veteran
Join Date: Dec 2003
Posts: 7,139
|
Quote:
Originally Posted by GMB@BP
I dont believe the 10 dollar mutual, I have tracked about 100 races and average payoff is 6 and change.
|
I have tracked thousands of races, not 100. The top 5 PP horses usually correlate to the top 5 public favorites. And the average payoff of the top 5 public winners is certainly not $6+. I believe $6 + is the average payoff of the top 2 PP horses or the top 2 public favorites.
|
|
|
04-30-2018, 07:44 PM
|
#26
|
Registered User
Join Date: Feb 2003
Location: Dark Side of the Moon
Posts: 5,870
|
Quote:
Originally Posted by Light
I have tracked thousands of races, not 100. The top 5 PP horses usually correlate to the top 5 public favorites. And the average payoff of the top 5 public winners is certainly not $6+. I believe $6 + is the average payoff of the top 2 PP horses or the top 2 public favorites.
|
my communication skilz are poor
can you run the query like I said, I actually would really like to know, my sample size is really small
2 top pp in a 5 or 6 horse field = win % and avg mutual
3 top pp in 7 horse field = win % and avg mutual
4 top pp in 8 = win % and avg mutual
5 in fields > than 9 = win % and avg mutual
non maiden races only
|
|
|
05-01-2018, 02:46 PM
|
#27
|
Veteran
Join Date: Dec 2003
Posts: 7,139
|
GMB@BP
I cant answer your question because my data does not have a field for number of horses in a race. That is because all my data has races of 8 horses or more, so field size would not become an issue. But like you, all my samples are non maiden races, no turf, no 2yo's or races for 3yo's only. No races shorter than 5.5f or longer than 1.1/8. No off track conditions either.
But if it helps, there was a regular poster who posted here about a year ago about the top 2 PP horses showing that they won 50% of the time. He had done a 15,000 race sample study. What he made me understand was that hit rate and average payoff were not affected by field size. I did not believe it at first but I did come to that same conclusion myself after looking into it.
His reason for posting was to see if we could find when one of the top 2 PP horses would win. If you knew that you could turn a profit because betting them blindly each race was a loss. But that the $64K question in horse racing.
|
|
|
05-02-2018, 04:17 PM
|
#28
|
Registered User
Join Date: Feb 2003
Location: Dark Side of the Moon
Posts: 5,870
|
Thanks for getting back to me.
|
|
|
05-08-2018, 11:26 AM
|
#29
|
Veteran
Join Date: Sep 2003
Location: NEW YORK CITY
Posts: 3,670
|
Quote:
Originally Posted by GMB@BP
my communication skilz are poor
can you run the query like I said, I actually would really like to know, my sample size is really small
2 top pp in a 5 or 6 horse field = win % and avg mutual
3 top pp in 7 horse field = win % and avg mutual
4 top pp in 8 = win % and avg mutual
5 in fields > than 9 = win % and avg mutual
non maiden races only
|
Most of the above are like this a % pt or 2
F 27 is Prime
Code:
SQL: SELECT * FROM STARTERHISTORY
WHERE [DATE] >= #01-01-2017#
AND [DATE] <= #12-31-2017#
AND RANKF27 <= 5
AND FIELDSIZE >= 9
Data Summary Win Place Show
-----------------------------------------------------
Mutuel Totals 107916.20 109612.00 111398.10
Bet -134252.00 -134252.00 -134252.00
-----------------------------------------------------
P/L -26335.80 -24640.00 -22853.90
Wins 10187 19619 28317
Plays 67126 67126 67126
PCT .1518 .2923 .4218
ROI 0.8038 0.8165 0.8298
Avg Mut 10.59 5.59 3.93
|
|
|
05-08-2018, 09:27 PM
|
#30
|
Registered User
Join Date: Jan 2017
Posts: 39
|
Mike, to get pp do you have to have the BRIS data in JCAPPER or can you get it in the HDW data?
tia
|
|
|
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
|
|