ANYONE USING POWER POINTS TO CAP?

[JustMissed]

I copied this from the bris.com library about the use of their prime power ratings:
QUOTE:
"Win up to 55% of the Time!
So, just how good is Prime Power? Take a gander at these results: based on a lengthy study of tens of thousands of races, Prime Power hits 31% top winners! That's sounds great, but when the study is broken down further and we focus strictly on dirt races, the results get even more interesting:

Top Prime Power horses by three full points (3.0+) or better score 39% of the time!

Top Prime Power horses by six full points (6.0+) or better score 46% of the time!

Top Prime Power horses by ten full points (10.0+) or better score 55% of the time!

If the results of this study don't get you excited, you may need a reality check. Nowhere else will find any rating that even comes close to this! Prime Power pinpoints key horses that win a high percentage of the time. We're talking big percentages here - much higher than the success rate of favorites." END QUOTE

I am wondering if anyone uses power ratings(from any source) to select the contenders and how far down the rankings do you go? I usually just throw horses out till I get down to 1/2 of the field less 1.

I am thinking for Saturday at Tampa of using the TSN power ratings to get my contenders, then using race shape, track bias, TSN race and class rating and speed rating to get my picks.

I may even dutch the final two horses, if close in power rating. The last time I placed win bets on my two best horses(Calder) I actually did pretty good.

Any suggestions will be appreciated.

JustMissed

:)

[BillW]

justMissed,

The #'s are fairly accurate (i.e. the pct. they advertise). Unfortunately they point to a lot of odds-on/low priced winners. :mad:

Occasionally they point to a nugget, but I typically use them as a secondary indicator. I believe there have been threads here before on the subject. Maybe a search will turn something up.

Your 1/2 field - 1 guideline is a good one, a piece of data I have handy on 2814 6f dry, dirt sprints show an IV of 0.82 for "first half" of field. Top 3 Bris #'s are: 2.66, 1.60, 1.08.

BTW a +10 in my data shows 29 wins out of 62 starters, but this is out of a sample of 2814 races.

As usual, go slow and validate any use for your self and have fun.

Happy holidays,

Bill

[JustMissed]

Great info Bill.

I just finished capping the 1st for Tampa tommorrow, 12/27. A cheap5K MC, 5.5 dirt. I went through the pp's in about 8 minutes and came up with Bellamanga #9, a P1 & Kippertip #7, an FTS.

As it turned out the 9 horse is the top rated Prime Power horse at 92.5 and the #7 FTS Sire win % with 1st-Time starters is 20% with a Sire AWD of 5.9f. Derek Bell, the jock, shows 10% wins with N/A horses which I assume are mostly FTS and lightly raced horses.

Both horses are 8-1 MLO. If they are not bet down too bad and look ok in the paddock and warm up- I'll probably dutch both to win.

My son and I are using two of Jerry Stokes's longshot angles and one of Jim Lehane's, so it only takes a few seconds to scan the pp's for these and we are hoping the Power Ratings will give us more time to find the winner/key horses.

Thanks again,

JustMissed

[formula_2002]

Just Missed, I study the bris_power ratings and find that although the win % are high there are no profits to be made...

The bris_power raings win inaccordance with the horses final post time odds.

On average it correlates directly to the final odds.
But no profits can be made when that happens....

[Handle]

EquiSim tracks Prime Power in its profiles. At Tampa, from what I have that has Prime Power in it:

Dominant means 5 points or more better than the rest
Type/Races/starts/win%/ROI
Overall:
Dom /116 / 116/ 28.45/-22.41
Best/457/458/ 28.17/ +6.86

All things considered, that's pretty good - but uncommon at most tracks.
Other stats include super sprints, long routes, etc., here are a few:

Sprint:
Dom /32/32/25/-12.5
Best /146/146/23.97/-26.25

Route:
Dom /20/20/-20.5
Best / 85/85/22/35/-26.35

Turf Route:
Dom /35/35/31.43/-20.29
Best /74/74/31.08/9.19


-Nathan

[hdcper]

I use Nathan's program Equisim and did some tracking of the prime power number too. I reviewed 1656 races and looked at the top prime power horse through the 14th rated prime power horse.

Below of some of the results:

Top prime power
1654 horses
573 wins
34.64%
-4.53 ROI
2.77 IV(based on avg field size of 8 or 34.64/12.5)

Second prime power
1658 horses
315 wins
19.00%
-21.85 ROI
1.52 IV

Third prime power
1703 horses(apparently ties in rating exist)
235 wins
13.80%
-17.79 ROI
1.10 IV

Fourth prime power
1740 horses
183 wins
10.52%
-23.62 ROI
.84 IV

Thus top four prime power win 1306 of the 1656 races or roughly 79% of the races (strong consideration for exotic plays to say the least). Also top pick strong for a single factor alone, with slight loss overall. Improvement maybe obtainable by additional factors to separate this group further.

Hdcper

[formula_2002]

Hdcaper and Nathan. If you dont mind, do 1 further calculation for each sector presented.

calculate the ratio for ,

Sum # of wins/ sum # expected wins
(# expected wins = 1/(odds+1))

That's the number of actual wins divided by the number of expected wins.

I would expect the ratio will fall between .75 min to .90 max.

Thanks
Joe M

[JustMissed]

Originally posted by hdcper
I use Nathan's program Equisim and did some tracking of the prime power number too. I reviewed 1656 races and looked at the top prime power horse through the 14th rated prime power horse.

Thus top four prime power win 1306 of the 1656 races or roughly 79% of the races (strong consideration for exotic plays to say the least). Also top pick strong for a single factor alone, with slight loss overall. Improvement maybe obtainable by additional factors to separate this group further.

Hdcper

GREAT INFO FELLOWS. I THINK WE ARE ON TO SOMETHING.

79% win out of top four is a hell of a starting point. I'm thinking to focus on these four to come up with one or two win possibilities or a KEY horse for the exacta and trifecta. If I only played value for the win(one bet or two horse dutch) and then only played a $20 minimum expected payoff for the exacta and only played the trifecta if total odds were greater than 12, I could probably overcome the low ROI.

The only other thing I would add is to look at the throw out horses for longshot angles.

I have always been impressed by Mr. Ainslie's approach to compreshensive handicappiing where he says to first consider qualifying a contender as to:

1. Distance
2. Form
3. Class
4. Footing/Track bias

Then you can look at other factors(angles) to identify the potential winner.

Anyway, thanks again for your effort. Your analytical capibilities are very impressive.

JustMissed

:)

[formula_2002]

Just Missed, from an analytical pont of view, if you cannot make a profit in the win pool, you cannot make a profit in the exacta pool.

Also, if you cannot make a flat bet profit in the win pool you cannot make a dutching profit.

[GameTheory]

Originally posted by formula_2002
Just Missed, from an analytical pont of view, if you cannot make a profit in the win pool, you cannot make a profit in the exacta pool.


Why not?

I would say if you can ALMOST make a profit in the win pool (say a 3-4% loss), then you could make a profit in the exacta pool betting the same horses as your keys. Exacta pool is less efficient -- easier to beat.

[formula_2002]

GT, Based upon thousands of "live" exacta tote board studies, I find that the exacta pool is very efficient.

Ther are no crumbs to be had there.

[JustMissed]

Originally posted by formula_2002
Just Missed, from an analytical pont of view, if you cannot make a profit in the win pool, you cannot make a profit in the exacta pool.

Also, if you cannot make a flat bet profit in the win pool you cannot make a dutching profit.

I believe you did not understand my post. What I was saying is that if a sample shows 573 winners out of 1643 races, 34.64% with a negative -4.53% ROI, you need only need to be more selective in your betting to improve the ROI.

If you can maintain the approximate win percentage but only bet on races with longer odds, you will without question make more money. This is why the PROS, I believe, play fewer races than hacks like me. They know when to keep their hands out of their pocket.

JustMissed
:)

[Tom]

I think the PP number has speed as a component of it. It would be interesting to see how the two numbers compare-does PP outperform speed? If not, they are just diluting the speed number.
Does anyone know exacctly who goes into PP?

[Rick]

Tom,

I once did a regression study on the Prime Power number. I got a very good correlation with race rating, class rating, speed rating and odds in the last two races. If anyone is thinking about using this in a method though I would advise against it. I don't think that -4% will hold up since my testing showed something more like -15% at Southern California tracks. I guess it would be OK for contender selection if you like that kind of thing, but so would using the top four morning line choices. I'm sure it influences odds now, so it might also be useful for predicting who the favorite will be.

[Rick]

formula,

I'm not surprised. Some are a little better than others, but you definitely couldn't use any single factor to win. And power ratings like Prime Power combine too many intercorrelated variables in order to boost their win %. Unfortunately, doing that also hurts the ROI.

By the way, I looked at your site and couldn't find anything except an excel spreadsheet download file. Is that all there is?

[KyRacer]

Rick,

Have you ever done any studies with Prime Power and recency. I've seen a top Prime Power horse that hadn't raced in 8 months and had no workouts, so those factors don't seem part of the equation. Would be interesting to know if those factors or Jockey and trainer factors migth improve it.

KyRacer

[Rick]

KyRacer,

I didn't check recency, jockey, or trainer. The correlation was high without using those. I tried to use the most important correlated factors that didn't negatively affect the ROI in a power rating for a while. Three factors got something like -6%. It didn't seem too worthwhile since I found combinations of two factors that did better than that and later found several three-factor models that were much better.

[hdcper]

Rick,

Would you be willing to share some of your results?

If not, I certainly understand.

Hdcper

[Rick]

hdcper,

If it's about Prime Power, yes if I can still find it. If it's about anything else, no.

[hdcper]

Rick,

I completely understand, if you find what you would share you could email me at hdcper@hotmail.com.

Just identify the post with something like PrimePower since most of my email goes to junk mail so I can avoid the spam.

Thanks,

Bill