Power Ratings

[aaron]

What are your experiences in the use of power ratings?Are any of the power ratings available particularly superior to others?Does anybody have any thoughts on this?

[BMeadow]

As someone who sells power ratings (Master Win Ratings), I think you need to ask several questions about such ratings:

What factors go into the ratings?
What factors do not go into the ratings?
How are these factors determined? (e.g., there are many ways to compile trainer ratings, and this is just one factor)
What is the importance of each factor?
Who does the compiling--human or computer?
Are the ratings adjusted for today's situation?

Any service selling power ratings should be able to give you a clear explanation of exactly what goes into such ratings, and what does not, available free to interested parties (my own MWR instructional package is at my website www.trpublishing.com).

No rating system is perfect because each one is different, covering different items, compiled differently, and is to be used in different ways. You have to figure how to integrate the particular product into your own handicapping.

[socalsportsbook]

Without trying to sound like a "schill" I can vouch for the Master Win Ratings. I use them and they are great.

As with any such ratings they will often point to the logical ML favorite BUT you will also find a lot of highly-rated horse going off at good odds (4-1 to 8-1). For the exotic players they are a great tool.

When I come across a logical horse in the MWR and he gets away from 4-1 to 8-1--as a friend of mine likes to say, "push your f'n car title through the window".

Barry's numbers are not selections but combined with your handicapping skill will usually point you in the right direction.

[GR1@HTR]

Barry,

I think you have a solid methodology behind your ratings since part of the calculation includes video replay, pace, workouts, class drops and all the important variables. Are the Master Win Ratings were calculated by you or perhaps another skilled employee?

[BMeadow]

I'm the skilled employee. In a short while I'll be watching and taping the Hollywood races on TVG, and watching the Bay Meadows races on the Internet, making notes as I go.

People sometimes ask if I plan to do other-than-California races. I tell them that at least for me, doing two tracks (occasionally three when Stockton, Ferndale, Sacramento, and Fresno are in action) is already a full-time job since you have to learn not only the horses but the trainers, jockeys, and the track.

[sligg]

Barry Meadows: (place/show overlays)

Glad to see you are on the PA boards. I just finished reading this chapter from your book Money Secrets and I'm wondering what your current feeling is about this method of betting using the online toteboard supplied by Brisbet. They have a screen that gives the money pool in percentages for the win/place/show positions for each horse.

For example at Belmont in the eighth race at zero minutes to post
these were the numbers:

win 44% (4/5)
place 37%
show 24%

Note the small % in the show slot.

The favorite came in second paying:

$2.80 place
$3.20 show

I'm sure this would have been a Dr. Z overlay bet.

The big problem with the online toteboards is they are inconsistent. The minutes to post don't necessarally coincide with the track toteboards-so that there are many times I can't place the online bet at zero minutes because at the track the race is already off.

[GameTheory]

The toteboard at www.racingchannel.com is much more reliable (and up-to-the-minute) then the Bris / Supertote one....

[David McKenzie]

//The toteboard at www.racingchannel.com is much more reliable (and up-to-the-minute) then the Bris / Supertote one....//

It works faster than the Bris or the DRF tote for getting live odds. The only drawback would be their inconsistency in posting results quickly, but that's a minor beef.

Tip: Sometimes the Racing Channel tote wil get stuck. Press "F5" or "Ctrl+R" to refresh the page when that happens (on a PC).

[sligg]

I agree that the Racing Channel tote is better than Brisbet and I use it to bet.

I have two problems with Racing Channel:

When I place a bet the bet screen pops up telling you what you bet and if it was accepted. I have tried several ways but I never seem to be able to get back to the track/pool screen. I click on the back button, the track/pool screen appears but I can't load a track.

The other problem is I keep getting shut out. I have to bet as close to post time as I can and sometimes I can bet at zero minutes to post and other times my bet is rejected.

Youbet does not have a good betting site at all for my purposes.

Brisbet has several features that are better than the others.

1. You can list five tracks at a time.
2. You can view the betting pools in the money mode or the percentage mode which is what I use looking for ineffiencies in the place/show pools.

If you read my previous post I give an example of this at Belmont yesterday.

I know no one on the PA boards is much interested in this place/show overlay. Some of you have already discarded Dr. Z's book on this subject, namely Dick Schmidt and a few others. This appeals to me because I'm tired of the rigors of full handicapping. I realize my talent limits and patience levels (and age is a factor). Looking for these overlays is less stressful and mostly mechanical in operation.

Since December 2001 to April 2002 I have shown a profit from a low of 10.9% to a high of 19.8%. I only bet $5.00 per race to place or show so I'm not looking to get "rich" but it's fun and it keeps my mind agile and I certainly won't lose any money.

Thus far in May there were 38 bets, 32 placed (84.2%) and 37 showed (97.4%). The place profit is 7.2% and show 10.5%.

Right now I don't have a smooth betting operation going because of the inconsistent online toteboards and I'm doing a lot of Excel spread sheet filtering trying to find the best mix. I have learned several things. You must stay away from cheap tracks and only bet the premier tracks. I am looking to bet on the class of horses that have proven consistency and the maximum post odds must be at even money preferably less. It will be interesting to see what happens because I plan on giving this a full year test.

I will post the results, modifications and all.

There was one stretch in February where I won 29 consecutive bets-which was exciting. Using the percentage pools on Brisbet based on an idea from Barry Meadow's book Money Secrets I don't need to use a hand held computer or the software found on two web sites which are expensive. I don't understand why these programs cost so much.

[BillW]

Originally posted by sligg
I agree that the Racing Channel tote is better than Brisbet and I use it to bet.

I have two problems with Racing Channel:

When I place a bet the bet screen pops up telling you what you bet and if it was accepted. I have tried several ways but I never seem to be able to get back to the track/pool screen. I click on the back button, the track/pool screen appears but I can't load a track.




Sal,

It sounds like you are trying to do everything in one browser window. Typically I have a window open for each tracks tote I am following (via the Racing Channel) and a window open for Brisbet. That way interacting with Brisbet is independent of the tote displays (and I can have as many tracks displayed as I wish via Racing Channel).

The windows with the tote displays can be shrunk down to maybe 3" X 3" for minimum interference with each other.

Take a look at this approach. Hope it helps.

Bill

[BMeadow]

I also monitor the place/show percentages for low-odds horses that interest me, and try to bet at the last minute if a horse has below the place/show percentages I describe in Money Secrets At The Racetrack.

Rather than use the Bris percentages (which are usually not as up-to-date as the odds on the Racing Channel), I simply use a $5 calculator to figure these percentages, often as the horses are steps from entering the starting gate.

However, at some major tracks, more than 40% of the money comes in after the race already goes off--and with so many people doing this, many times a supposed overlay has turned itself into an underlay in the final reckoning.

Having done this daily for years (both on-site and in recent years via the Internet toteboards), my experience is that this is basically a break-even strategy, of use mainly to people getting rebates. Without rebates, I wouldn't bother with it other than as a just-for-fun strategy.

Of course, I'm not hitting 97% of my show bets, either.

[sligg]

Bill,

Can you explain how you setup the screens? I tried to shrink the phonebet race screen but I wasn't able to size it using the cursor.

[BillW]

Originally posted by sligg
Bill,

Can you explain how you setup the screens? I tried to shrink the phonebet race screen but I wasn't able to size it using the cursor.

Sal,

I was refering to the racing channel tote windows and Brisbet. Possibly phonebet prohibits resizing in their code. I can't get to them to check it out as they only allow microsoft browsers. (guess my money is not good enough for them :) ) .

If standard dragging of the window's edge with the cursor doesn't work, it probably cannot be done with that site. Possibly starting a thread in the software section will stir up an answer from a Microsoft user.

Good luck,

Bill

[sligg]

Barry Meadow:

I'm sceptical of my results thus far. As I said in my previous post, I sometimes can't get my late bets in but I still record the bet in my spreadsheet. There is no way to overcome the late money after post-so it appears that my results are a fluke and maybe fudged by using final odds which can't be bet.

But there was always late money even in the time of Dr. Ziembas book, so I'm inclined to think he also used final odds in his research-which would make his research flawed.

Oh well, at least I'm having fun and not losing money but then again I'm not making money.

When using the Racing Channel tote, how many horses are you calculating the percentages for and are they horses you handicapped?

My results are based on the post time favorite at even money or less and that is why the place and show win percentages are so high and naturally the payouts so small.

Rebates would help, but living in NYC, and not being a high roller I don't think this is an option for me. I don't know if Atlantic City gives rebates but it's a 2.5 hour trip each way-so that is out of the question. There is an online betting site www.betehorse.com (not sure of the address) that gives rebates but they don't carry all the tracks and I think it's an offshore operation.

[so.cal.fan]

"However, at some major tracks, more than 40% of the money comes in after the race already goes off--and with so many people doing this, many times a supposed overlay has turned itself into an underlay in the final reckoning. "

Interesting, Barry:
We checked this out briefly (two races) at Santa Anita, and it was about 25%, but that was on the early races. 40% is a big chunck of the pool! No wonder the odds change.

[Dave Schwartz]

SCF,

We also did a short study at about 6 major tracks and found that typically from 35-55% appears on the tote after 0 minutes to post.

Some of that is because the web pages are actually as much as 2 1/2 minutes behind! Some tracks are worse than others because they (apparently) do not get a simulcast refresh as often.

As I understand it, basically it is a function of bigger tracks having better hardware/software/connections.

Had an interesting experience one day in comparing a tote board I was watching with someone else who was watching from another part of the country. We were showing different pool information for the same race!

I believe that the bottom line is that there is no such thing as "final odds" any more. They are just too volatile.


Regards,
Dave Schwartz

[so.cal.fan]

That is very interesting, Dave.
I had no idea it was that much.
So.......the tote board means zilch, at least to me it does.
How can anyone set odds limits or watch "action" or anything else?
I bet a horse at Fairplex last year, 9/2 going into the gate, 5/2 when the flag went up and 9/5 after the race!
At the time, I was trying to confine my bets to 3/1 and up, but how can one do that anymore?

[ranchwest]

Some of the video feeds are also far behind. I saw one the other day that appeared to be about 4 minutes behind.

[aaron]

It is impossible in this age of racing to know your final odds.I primarily bet the NY circuit and I've seen odds change from 9/5 to 1/1 after the race has gone off.I've bet a horse at 8-1 and when it won I couldn't believe it had gone off at 5-1.On the other hand I've had horses drift up and win at bigger odds,especially when a short priced horse is beaten.As for tote board watching I don't really know what the answer is.I still find it helpful to pay attention to the board even though I don't know what the final odds will be.

[andicap]

Longer priced horses tend to fluctuate upwards closer to post-time as "steam" horses get hammered. Or lukewarm chalk gets bet down very late when the public thinks it has an overlay.
I don't see too many 8-1 shots bet down to 6-1 after the last flash.

[aaron]

Yesterday's 1st race at Belmont, the winner went from 16-1 to 12-1 after the race went off,but you are right its usually the short priced horses who get hit.It was probably Earnie Dahlman money yesterday in the 1st

[Dave Schwartz]

I believe I have related this story before, but because of its relevance to this thread I will share it again.

This was about 4 years ago and I was with a client at the Reno Hilton's racebook. We we interested in a particular horse (at Belmont) that was staying between 35-1 and 39-1. I had vowed to not bet him if he went above 39-1.

We watched the tote for the last 5 minutes, noting every flash. Just before the first horse entered the gate, I made my bet. The TVs showed the odds of the horses during the race and as they headed for home my horse took the lead. I watched the odds go over 40-1 and cashed an $84 winner.

Later that evening I was talking to one of our users who was playing that race at Louisiana Downs. He had the same rules as me (i.e. no play above 39-1) and did not have the horse because the horse never went UNDER 40-1 in the last 10 minutes!

So, I guess where you stand matters. (Perhaps you can get better odds if you raise one foot off the ground.) <G>


Regards,
Dave Schwartz

[anotherdave]

Originally posted by Dave Schwartz
We we interested in a particular horse (at Belmont) that was staying between 35-1 and 39-1. I had vowed to not bet him if he went above 39-1.


Interesting story, but I don't get why you would accept 39-1, but not 40-1??

AD

[andicap]

http://www.drf.com/news/article/37537.html

relevant to this discussion

[Dave Schwartz]

AD,

The simple answer is, "The system's definition said 'don't play horses at 40-1 and above' so I did not play it."

If you want to read more about the concept behind the strategy try this link:

http://www.horsestreet.com/freestuff/articles/key2mint/index.html

Dave

[tanda]

Interesting article, Dave.

I have been testing a program I have designed. Most of my recent testing is based on playing long-shot contenders or the highest odds contenders without "splitting hairs" and trying to pick the winner. My initial results are that there is profit from that perspective. My non-contenders have a terrible R.O.I and I.V. Also, when testing all long-shot winners, those who are contenders win at a much higher I.V. than non-contenders. Also, I agree with the 40-1 idea. My contenders who win as longshots are in the 10-1 to 25-1 range with most at 15-1 or so. I am tinkering with a 25-1 or 30-1 odds cut-off.

Your results seem to verify mine and vice versa.

[keilan]

Great article Dave, wish someone would have shown me this when I first started. My betting strategies for WIN horses is very much the same as your article outlined, I then bet the two horses from my list of contenders $20.00 to win. Many days if I only make this bet I would show a profit between $200.00 to $300.00. Decent days wages for many.

But oh no that's not nearly good enough I have to play the exactors and pick 4's, and more days then I care to admit to I go home even or down.

The allure of catching the big ticket or grinding it out every day. Greed has not always been my friend. This is something I struggle with.

[Dave Schwartz]

Over time and with research we have determined that 40-1 is not optimal for all (or any?) field sizes. We scaled it downward as the fields got smaller.

BTW, one might logically ask, "If this approach is so good, where is our yacht?" We determined that in the long run we would never be able to bet enough money on these horses to make significant profit relative to the work involved. (And it is very work-intensive.) The pools just won't support it.


Dave

[tanda]

Dave,

Why is it work-extensive? I presume that you use HSH to help with the data processing?

Also, if it generates several playable races per card, then by playing multiple card (all those on a given day), you should be able to churn a decent size bankroll with medium sized bets due to the high volume of bets.

[Dave Schwartz]

Tanda,

Not True. It is actually quite intensive. First, you MUST handicap the race. Second, you must order the horses. Third, and this is critical... you must watch the tote very carefully, as it often comes down to a question of betting the highest odds horse (usually a loser) or throw him out and bet the low odds horse. Try it... you will see what I mean.

But the real issue is that the profit is being fueld by longshots. The average mutuel is in the $40 with some ranging up to twice that. You cannot wager very much on this level of horse. In fact, remarkably little.

Further tests have shown the the advantage seems to run in the 4-6% range now, not the 15-19% we saw when we tested it several years ago.

Nevertheless, everytime I take this out and test a full day, it seems to win. Not much, but it wins.

Before anyone gets crazy and runs out to mortgage the horse, remember that it is still dependent upon a good "contender" strategy.


Dave

[tanda]

Dave,

I was assuming an automated betting process as well. I use Automate with self-designed macros to place bets and watch the tote automatically. This drastically reduces the work of placing the bets.

Also, your lower R.O.I. does significantly reduce the percentage of bankroll which can be wagered.

My program "handicaps" for me, so a lot of work is saved. Of course, it is a direct expression of my handicapping philosophy and I tinker with it almost daily, so I do not consider it a "black box". If I processed data, then handicapped myself, there would be added time. However, once it processes the data, it produces the same ordering and contenders that I would if it did it myself. Can't you program HSH to do that for you?

[Dave Schwartz]

Tanda,

One can program anything to do anything, given the resources (i.e. time).

This is just not a high priority since I have no interest in playing this approach.


Dave

[MikeH]

Some statistics from my database for the recently-concluded Santa Anita meet (fast track only in my numbers.):

There were 757 horses, or 13% of the total,that went off above $39-1.

4 of those won (0.5% of the winners.) (That's one-half of ONE percent...)

The highest-price winner was $62.

[Dave Schwartz]

Mike,

40-1 shots can't pay $62.


Dave

[MikeH]

Dave-

All my prices are to $1. That's $61.90, or $125.80 returned for a $2 tix...

[bedford35]

Question for Barry Meadow:

You mention earlier in this thread that you watch the Bay Meadows races on the internet. Is the streaming picture quality good enough for trip handicapping? I wouldn't have thought it was. Do you also use downloadable replays from Compurace?

Unrelated question: Are you still planning to write a book with a tentative title "The Skeptical Handicapper" or has that been shelved? Any other books in progress? Thanks in advance.

bedford35

[rrbauer]

re: Streaming Video Quality - Bay Meadows

If you have a broadband network connection (DSL, Cable, T1, etc.) then the internet feeds from the Magna tracks are very close to what you would see on a simulcast-site TV. Good enough to pickup trip info and trouble. The exception to that is that sometime there is a lot of network congestion and on those occasions the stream gets slowed down and gets the herky-jerky's.

[alyingthief]

i recall keeper hill, whose first major race--before she went on to win the kentucky oaks, etc., etc.--paid me 114 bucks, with a 452 dollar exacta, and a 1100 to 1 tri that was so simple it was painful. i remember makeroomforzoom, a horse so outstanding on the face of it i passed out from disbelief when i collected 87, 40+, and 24 in three out of four races.

i hardly ever bet long priced horses like these, but when i do, i DO. a system like you mention would get its neck wrung at hollywood summer meet in those races most of us would consider "playable", because the chalk is just too hard to beat. i recall one of their pick sixes, with a guaranteed purse. every race was won by one of the top two public choices, except for a maiden, ridden by, you guessed it, stevens--no, his brother!!! come all the way from TP!! hilarious. and you know what? they had two such guaranteed purses that summer, and in the second, it was the same thing, except the bomber was ridden by, you guessed it!! Gary Stevens!!!

the system might work on SA winter, or Fairgrounds, though.

[Barremill]

MikeH,
using the same criteria 40-1/+ i get:

34,290 horses
1827 40-1/+(.053%)
26 WIN((.014)
$110 AVERAGE MUTUAL
-.211 ROI

barremil

[kenwoodall]

Sligg- keep stats on the success % and payoffs if you had bet to place if 6 or less are running. It helps my ROI. About 60% of my show bets are favorites if not over 30% of the show pool. I average 1 place or show bet every 2 races.

[formula_2002]

aaron and all

I generated an excel table indicating the results by ranking order for over 100,000 bris power rated horses.

The ratings correlate very nicely with the win%.

Anyone wishing a copy of the file may e-mail me..
There is never a charge for anything.

But I would appreciate any comments you may have.
Just place the comments here.

Joe M