|
|
10-23-2020, 10:29 AM
|
#91
|
Registered User
Join Date: Feb 2003
Posts: 2,105
|
I convinced myself to attempt the 7th grade math and think I have it right.
If a is a number between 0 and 1 and you make a combination line of
a*hp+(1-a)*pp where pp is the normalized public line and hp is home-made
b is (1+cushion pct) so betting when a*hp + (1-a)*pp > b *pp is the same as
when hp > (b+a-1)/a * pp.
For example, a half and half line a=1,
.5*( hp+pp)> 1.5* pp when hp>2*pp
|
|
|
10-23-2020, 10:39 AM
|
#92
|
Registered User
Join Date: Jan 2010
Posts: 2,176
|
Quote:
Originally Posted by BarchCapper
Wouldn't average field size play a huge factor in this? Using unrealistic field size numbers just for examples' sake, if it was 67% when field size averaged 10, if it goes up to even 90% when field size averages 4, it's a much less powerful factor for weighting, right?
|
Huge. I think someone posted a chart here a good while back, that showed how the public did on different field sizes. But yes it's huge.
__________________
One of the downsides of the Internet is that it allows like-minded people to form communities, and sometimes those communities are stupid.
|
|
|
10-23-2020, 11:42 AM
|
#93
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,912
|
Quote:
Originally Posted by FakeNameChanged
Didn't you say that your research shows that the top three betting choices has risen to 73% when it was at 66-67% twenty or thirty years ago. That's a 9% to 11% improvement on the public selection, whether it's from handicapping, or money flow on picking the winners. Why would anyone drop their weighting of something that is a reliable factor?
|
Because the goal is not to agree with the public but to find the spots when the public is wrong.
If we used your approach to the letter we'd simply give the horses the precise probabilities indicated by the tote board.
Where would that leave us?
|
|
|
10-23-2020, 11:57 AM
|
#94
|
Registered User
Join Date: Dec 2018
Posts: 246
|
Maybe it's time to think like a turtle and accept the markets is moving in a direction and not fight it. A little tougher to make money at the racetrack vs in the stockmarket where many who followed Richard Dennis and the turtles had made millions following this strat. But it is the foundation of your Renegade Handicapper product isn't it?
|
|
|
10-23-2020, 12:24 PM
|
#95
|
Registered User
Join Date: Feb 2009
Location: NY
Posts: 18,971
|
Quote:
Originally Posted by MPRanger
|
My favorite lines from this editorial because of how TRUE these factual statements are:
Quote:
Well, Bill Benter figured out, if you can’t get insider knowledge you could at least incorporate it into your model. Wait, but didn’t we just say that Benter had no access to insider knowledge? How can he then incorporate it in his model?
|
Quote:
Therefore, Benter argues, only a combination of a model’s estimate with the one of the market provides a true measure for its quality.
|
Quote:
After all, Bill Benter has written it in 1995, more than 20 years ago. There has certainly been a myriad of much more sophisticated and advanced models written in the meantime, which could easily beat Bill Benter‘s model on the racetrack.
|
|
|
|
10-23-2020, 12:36 PM
|
#96
|
what an easy game.
Join Date: Dec 2002
Posts: 43,096
|
I'll state the obvious. If i want to compare win probability of say the top public choice in different time periods, say monthly, I would normalize the monthly top choice win probability and compare that sum to the actual number of winners. Of course, the best analysis, is to compare these two sums, expected wins to actual wins by incremental odds. I would seek to minimize the increment for the largest
sample size . I always want to know the ratio of actual wins/expected wins . Repeat this for 2nd and 3rd choice.
__________________
Peace on earth, good will to all
GOD BLESS AMERICA
" I pass with relief from the tossing sea of cause and theory to the firm ground of result and fact"
Winston Churchill
|
|
|
10-23-2020, 01:40 PM
|
#97
|
clean money
Join Date: Sep 2006
Location: Maryland
Posts: 23,559
|
public odds modeling (known vs. unknown)
also 'obvious' category -
'Unknowns' (first time starters without Super(debut) Trainers, certain layoff situations, etc...) - 'unknown' that is cold 'dead' or hot 'live' on the board may correlate to performance, sometimes more than pedigrree, workout, trainer stat,, etc... models models that we have available...
correlated
also, occasionally see more established (' known') horses whose odds on the board stand in significant distance from their projected performance models. - maybe this is exactly the opportunity which we've been waiting for! and overlay, or an underlay... Time to Wager!!
- and/or maybe this horse has a 'narrative' (a fan favorite, an obvious trip, etc...) that is being overbet...
- and/or maybe this horse has some inside informatiion that is pertinent to today's performance, while not evident in past performance
When the odds 'don't make sense' you want to do everything you can to be able to distinguish overlays from inside-info. Addiing some models, and/or blending today's projected odds into the model are ways to overcome this.
__________________
Preparation. Discipline. Patience. Decisiveness.
|
|
|
10-23-2020, 02:31 PM
|
#98
|
Registered User
Join Date: Jan 2010
Posts: 2,176
|
Quote:
Originally Posted by Dave Schwartz
Because the goal is not to agree with the public but to find the spots when the public is wrong.
If we used your approach to the letter we'd simply give the horses the precise probabilities indicated by the tote board.
Where would that leave us?
|
I think you're missing the point. You don't know which ones are the top three, especially #1, until after the race is over, or halfway around the track. I defy anyone to give me the #1 odds horse at post time, and get a win bet down on it in 20 straight races. The top three, NFW.
Trust me, I understand what overlays and underlays are. I think Benter was wise in using the public along with his other 9 hcp factors. The public has an outstanding record of consistently picking the exact order of how often horses finish, 1 thru 10 positions.
I thought it was interesting in Benter's article, that he also used the picks of 47 newpapers pickers or touts(Only in Hong Kong I suppose) and their R2 value in selecting the winner almost duplicated his Fundamental Hcp model: .1016 to .1014. That makes perfect sense, since the majority of those newpaper picks come from hcapping the PP's.
I wasn't proposing betting the top three,(even though we currently don't know), but how can we use that info to improve our models. Top four, and it must be closer to 90%, granted not that hard in a 5 horse race, but 10 or 12, that's a lot of toss outs.
At no time did I say follow Wm. Scott's method, "to the letter" as you say. I was just suggesting focusing on the top three or four using whatever model you use, pace, money flow, hieroglyphics, it doesn't matter to me. When I attempted his method 3-1/2 decades ago, you couldn't tell who the top three were going to be. I was suggesting a model to get us closer to that as a starting point.
__________________
One of the downsides of the Internet is that it allows like-minded people to form communities, and sometimes those communities are stupid.
|
|
|
10-23-2020, 02:32 PM
|
#99
|
Registered User
Join Date: Mar 2005
Location: Queens, NY
Posts: 20,613
|
Quote:
Originally Posted by Dave Schwartz
Because the goal is not to agree with the public but to find the spots when the public is wrong.
If we used your approach to the letter we'd simply give the horses the precise probabilities indicated by the tote board.
Where would that leave us?
|
Right.
That's why I've been saying winning is not so much about making an odds line for every horses as it is about finding situations where the public tends to screw up.
At first Benter probably identified a handful of inefficiencies, but his odds lines weren't inclusive enough to be playing all the overlays he "thought" he was finding. Adding public money was a breakthrough for him because it made his overall odds lines more inclusive and accurate, but the key to winning was actually located in the inefficiencies he found to start.
I'd WAY rather find a few inefficiencies than worry about having a better quality odds line for every horse in every race. Find me some trainer pattern that wins at 25% at average price of $10.00 and I'll take care of winning without an odds line. Find me 10 things like that and I'll keep myself quite busy even though I won't be able to make profitable odds lines in many races.
__________________
"Unlearning is the highest form of learning"
Last edited by classhandicapper; 10-23-2020 at 02:42 PM.
|
|
|
10-23-2020, 04:15 PM
|
#100
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,912
|
Quote:
Originally Posted by FakeNameChanged
I think you're missing the point. You don't know which ones are the top three, especially #1, until after the race is over, or halfway around the track. I defy anyone to give me the #1 odds horse at post time, and get a win bet down on it in 20 straight races. The top three, NFW.
|
I think I have a very clear grasp of the point.
I've got zero interest in predicting who the top 3 public choices are. IMHO, that is a waste of time because - as you (basically) said - it's a moving target.
That's why I concentrate on what the odds SHOULD be.
If anything, the difference between what they should be and what they ARE would indicate the smart money.
But, still the same problem: How to accomplish that when that info isn't available until after the race is over.
For the record, our SHOULD BE predictions are "good." About 84% of the time the top horse goes off as either fav or 2nd choice.
Of course, a small percentage of the time a horse predicted to be under 3/1 goes off at 12/1 or above. Those horses are worthy of automatic bets.
|
|
|
10-23-2020, 04:18 PM
|
#101
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,912
|
Quote:
I'd WAY rather find a few inefficiencies than worry about having a better quality odds line for every horse in every race. Find me some trainer pattern that wins at 25% at average price of $10.00 and I'll take care of winning without an odds line. Find me 10 things like that and I'll keep myself quite busy even though I won't be able to make profitable odds lines in many races.
|
That approach will not work.
If it did, one could just use a number like BPP that produces a win rate around 30% in the top rank. Using your theory, a horse at 2/1 is break even and 4/1 should be a killer.
The problem is that the 30% hit rate goes higher or lower based upon the odds.
Same thing with trainer stats.
Or any other factor.
|
|
|
10-23-2020, 04:56 PM
|
#102
|
Registered User
Join Date: Apr 2012
Posts: 1,115
|
Are races where your selection to bet is one of the top 2 or 3 betting choices profitable long term for anyone?
|
|
|
10-23-2020, 05:38 PM
|
#103
|
Registered User
Join Date: Nov 2003
Posts: 4,668
|
Quote:
Originally Posted by sjk
Mark,
Anyone who makes a line and looks for overlays is using the public odds in making the comparison.
Most would have a requirement that there be a cushion between the value presented by the public odds and the expected return based on a personally derived line, such as looking for opportunities where the overlaid value is 100% greater than the oddsline expected return. (I am using a requirement along those line)
Making a composite odds line really does nothing more than adjust the required cushion percentage. You can manipulate the inequalities and turn a requirement about a composite line into a requirement on the oddsline at a higher cushion percent.
SK
|
TX very much for the response. It's well known that I'm an intuitive handicapper who constructs no handicapping line, but instead goes by feel in assessing value-which, of course, traces directly to closing odds.
So, you are saying that "smoothing" means blending the actual odds with the computer-churned line before comparing that product with the raw, real-time odds to assess value?
|
|
|
10-23-2020, 07:19 PM
|
#104
|
Registered User
Join Date: Jan 2010
Posts: 2,176
|
Quote:
Originally Posted by Dave Schwartz
I think I have a very clear grasp of the point.
I've got zero interest in predicting who the top 3 public choices are. IMHO, that is a waste of time because - as you (basically) said - it's a moving target.
That's why I concentrate on what the odds SHOULD be.
If anything, the difference between what they should be and what they ARE would indicate the smart money.
But, still the same problem: How to accomplish that when that info isn't available until after the race is over.
For the record, our SHOULD BE predictions are "good." About 84% of the time the top horse goes off as either fav or 2nd choice.
Of course, a small percentage of the time a horse predicted to be under 3/1 goes off at 12/1 or above. Those horses are worthy of automatic bets.
|
Okay, my bad. I thought this thread was about how Benter's model of betting overlays was a loser until he incorporated the tote board into his model. On my original post, I said maybe there's a way to predict the tote before post time. It sounds like your method arrives at that 84% of the time(fav or 2nd choice). Then congratulations, you've cracked the code.
__________________
One of the downsides of the Internet is that it allows like-minded people to form communities, and sometimes those communities are stupid.
|
|
|
10-23-2020, 08:03 PM
|
#105
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,912
|
Quote:
Originally Posted by FakeNameChanged
Okay, my bad. I thought this thread was about how Benter's model of betting overlays was a loser until he incorporated the tote board into his model. On my original post, I said maybe there's a way to predict the tote before post time. It sounds like your method arrives at that 84% of the time(fav or 2nd choice). Then congratulations, you've cracked the code.
|
LOL - Not exactly cracked the code but it is proving to be solidly profitable.
The real question I need to get answered is, "Will it stand the test of time?"
What we've been finding is that our approach to the world has been needing an overhaul every January.
I'm hoping that the things I learned from the Covid Era will carry over. Really started looking strong a few weeks ago.
Sadly, just in time for the Winter Doldrums.
|
|
|
|
|
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
|
|
|
|
|