Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board

Go Back   Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board > Off Topic > Off Topic - General


Reply
 
Thread Tools Rate Thread
Old 12-08-2016, 08:49 PM   #31
CincyHorseplayer
Registered User
 
CincyHorseplayer's Avatar
 
Join Date: Jan 2008
Location: Cincinnati,Ohio
Posts: 5,289
Where I have arrived at has absolutely nothing to do with escaping variance or not having a certain temperament for A or B. As your thinking changes and you evolve what you discover is that it is not all black and white and basic as it seems. For one in my first post one of the greatest realizations I have made is exactly embracing the reality of variance and battling it in the form of more bets as opposed to less bets. That it is through 100 or more bet cycles that it evens out and that approach would get you to where you are going. But most of the discovery had to do with what to do with new ratings and new analysis and analyzing it in a statistical manner. While at first I tried to create weighted models across the degrees of success with several rankings I realized that it distorted what a good bet was and that the strongest rating that was showing up in a track profile was what to play. The other thing I had to personally overcome is the desire to handicap the horses to be bet when they fit. It was a case where the strength was the statistic and it demanded to be bet period or you would handicap away profitability. It was hard to go against my own nature. It was hard to comprehend that betting ALL of the profile horses to win would yield a bigger return even if the overall ROI might go down with a more selective approach. It's all about more return right? Plus this would mean a paradigm shift to more research and modeling as opposed to constant handicapping. Find what is winning and bet it, period. It took a while to get through my head last winter at Gulfstream. From 12/9 to 2/15, 71% of the turf routes were won by the top 2 in my turf rating. Still playing a Pick Em approach and not tracking % of entries the avg mutual was over $11 with weekly ROI's ranging between 56% to 77%. It was steady for that period and blew up completely in late February. I hadn't adequately capitalized but it was an eye opening lesson.

Fast forward to October 2016. Starting to profile Santa Anita for the BC. It didn't take long to see that my secondary turf rating(T2) that usually made it's presence felt on short turf courses was winning races at a big clip and it was practically the antithesis of the Del Mar turf course which I played lightly late summer. I started betting by 10/6. The model held all of the short meet. The T2 rating won 89% of the races but there were a lot of contenders. They represented 30% of the entries but won 2.97% of their fair share at a dead even mutual of 10.00. ROI was 55%. Lesson learned for good. This was a new method of play that had previously not existed. With high hit rates and ROI it represented virtues that were the antithesis of the perceived modern game. This wasn't aggressive and swinging for the fences. It was taking what was there.

The same thing happened about 5 years ago when I realized how many wins I was squandering playing purely exotics. I made a distinct split between bet approaches. And using a 3 unit method(3 units win on sub 4-1)(2 units win, 1 unit place on 4-1 up). I realized that this also was a thing in itself. A self contained method that could stand on it's own. But to me it was just another method.

A similar thing happened when I stopped tracking contenders by what I call the fun with numbers method(1st choice 2nd choice 3rd choice etc). If you had a 1st choice as the ONLY contender in a race a 3rd choice winning at long odds was meaningless. So I tracked contenders simply by AB(strong contenders) and C+(marginal contenders) and let over 1,000 races pile up to start making conclusions. Which shattered my illusions of excess value handicapping.(I covered this is an entirely other thread). What this ultimately laid the basis for was not only proper ticket structure for sequential bets but gauging the potential value in any race for exotics. Tracking the odds ranges kept me out of low expectation returns that also included many longshots.

Add in developing specific uses for certain bets and I had 5-6 approaches to making money at this game all with different hit rates and profit potentials. They all have added up to a more well rounded game but being as they are largely independent of each other even in the same races it wasn't geared toward lowering risk at all. In fact that it requires a higher level of investment it is quite the opposite.

In the last 18 months I've learned a lot and still have much more to utilize. Most of my conclusions and examples are relatively new and the future still has to play out. But I feel at least mentally I am playing a different game. The only thing I have made easier is formatting for pace figures while adding twice as much work! LOL!

Last edited by CincyHorseplayer; 12-08-2016 at 08:55 PM.
CincyHorseplayer is offline   Reply With Quote Reply
Old 12-08-2016, 09:22 PM   #32
CincyHorseplayer
Registered User
 
CincyHorseplayer's Avatar
 
Join Date: Jan 2008
Location: Cincinnati,Ohio
Posts: 5,289
Utilizing the approach of more investment and higher volume is definitely a task to swallow and execute. Which for personal reasons is difficult, at least for another 4 years because of certain financial obligations. I am forced to be content with my modest earnings and being on a new wave of learning. And I am content to a great extent. Laying some good track for the future when navigating an often bleak landscape. The shift to turf as my primary surface has definitely altered the equation. But to the extent where I was pretty dismal on dirt in 2016. Something I had reversed in the 2nd half of 2015 and vowed to continue. By no means have I found the keys to the kingdom.

Plus there is the reality of what I call dead zones. A fair track on both surfaces can be this when there is no leaning bias and I am not talking track bias but a rating bias. Tampa is a place like this with it's IMO loose and sandy surface and quirky turf course it's hard to get a handle on. Good bets require different skills and are hard earned. A place like northern and eastern tracks in winter that really are dictated by shifting biases where guys like EMD4me thrive because of what he does. Middle of meets with shifting biases either rating or actual. Bad weather. A "dead arm" period for horses who have been running. Beginning and ends of meets. Sinking your teeth into anything is short lived and we are always on the chase. There is always work to do and adversity to face. One thing I need to do is up my computer skills because I need to do even more research and need to do it more efficiently. Turf courses IMO have to be all hand done as they are quirky to interpret with so many layoffs. All else I need more time efficiency. I haven't learned how to do that yet. Delta and Traynor you guys have given me some good starts for that. All this is an on going process and I love talking about it. Great game.

Last edited by CincyHorseplayer; 12-08-2016 at 09:27 PM.
CincyHorseplayer is offline   Reply With Quote Reply
Old 12-08-2016, 10:25 PM   #33
Robert Fischer
clean money
 
Robert Fischer's Avatar
 
Join Date: Sep 2006
Location: Maryland
Posts: 23,512
randomness exists in everything probabilistic (excuse me tonight, my brain's battery is on 'safe mode') , I'm assuming as long as that probability is <1 or <100%.


this includes horse racing

also includes simplified examples of horse racing using known, controlled probabilities

A coin flip, simulated in my head
I'll bet on 'HEADS'. 50% win chance.
1,000 coin flips and I can see as many as 10 consecutive tails loss outcomes.

A true 2/1 favorite. 33% , 1,000 races = as many as 17 consecutive losses...

those are just the simple examples i'm capable of describing.
I'm sure a better mathematician could demonstrate a simulation with a worst-probable outcome scenario.


So yes, randomness exists.

We can fight randomness with high percentage wagers, and lower percent of bankroll wagers

will not completely eliminate, but we can produce a 'manageable' environment with an very low probability of ruin due to randomness.

^all that stuff is talking about known probabilities and by extension, players capable of knowing their probability range estimates
__________________
Preparation. Discipline. Patience. Decisiveness.
Robert Fischer is offline   Reply With Quote Reply
Old 12-08-2016, 10:42 PM   #34
traynor
Registered User
 
traynor's Avatar
 
Join Date: Jan 2005
Posts: 6,626
Quote:
Originally Posted by CincyHorseplayer
Where I have arrived at has absolutely nothing to do with escaping variance or not having a certain temperament for A or B. As your thinking changes and you evolve what you discover is that it is not all black and white and basic as it seems. For one in my first post one of the greatest realizations I have made is exactly embracing the reality of variance and battling it in the form of more bets as opposed to less bets. That it is through 100 or more bet cycles that it evens out and that approach would get you to where you are going. But most of the discovery had to do with what to do with new ratings and new analysis and analyzing it in a statistical manner. While at first I tried to create weighted models across the degrees of success with several rankings I realized that it distorted what a good bet was and that the strongest rating that was showing up in a track profile was what to play. The other thing I had to personally overcome is the desire to handicap the horses to be bet when they fit. It was a case where the strength was the statistic and it demanded to be bet period or you would handicap away profitability. It was hard to go against my own nature. It was hard to comprehend that betting ALL of the profile horses to win would yield a bigger return even if the overall ROI might go down with a more selective approach. It's all about more return right? Plus this would mean a paradigm shift to more research and modeling as opposed to constant handicapping. Find what is winning and bet it, period. It took a while to get through my head last winter at Gulfstream. From 12/9 to 2/15, 71% of the turf routes were won by the top 2 in my turf rating. Still playing a Pick Em approach and not tracking % of entries the avg mutual was over $11 with weekly ROI's ranging between 56% to 77%. It was steady for that period and blew up completely in late February. I hadn't adequately capitalized but it was an eye opening lesson.

Fast forward to October 2016. Starting to profile Santa Anita for the BC. It didn't take long to see that my secondary turf rating(T2) that usually made it's presence felt on short turf courses was winning races at a big clip and it was practically the antithesis of the Del Mar turf course which I played lightly late summer. I started betting by 10/6. The model held all of the short meet. The T2 rating won 89% of the races but there were a lot of contenders. They represented 30% of the entries but won 2.97% of their fair share at a dead even mutual of 10.00. ROI was 55%. Lesson learned for good. This was a new method of play that had previously not existed. With high hit rates and ROI it represented virtues that were the antithesis of the perceived modern game. This wasn't aggressive and swinging for the fences. It was taking what was there.

The same thing happened about 5 years ago when I realized how many wins I was squandering playing purely exotics. I made a distinct split between bet approaches. And using a 3 unit method(3 units win on sub 4-1)(2 units win, 1 unit place on 4-1 up). I realized that this also was a thing in itself. A self contained method that could stand on it's own. But to me it was just another method.

A similar thing happened when I stopped tracking contenders by what I call the fun with numbers method(1st choice 2nd choice 3rd choice etc). If you had a 1st choice as the ONLY contender in a race a 3rd choice winning at long odds was meaningless. So I tracked contenders simply by AB(strong contenders) and C+(marginal contenders) and let over 1,000 races pile up to start making conclusions. Which shattered my illusions of excess value handicapping.(I covered this is an entirely other thread). What this ultimately laid the basis for was not only proper ticket structure for sequential bets but gauging the potential value in any race for exotics. Tracking the odds ranges kept me out of low expectation returns that also included many longshots.

Add in developing specific uses for certain bets and I had 5-6 approaches to making money at this game all with different hit rates and profit potentials. They all have added up to a more well rounded game but being as they are largely independent of each other even in the same races it wasn't geared toward lowering risk at all. In fact that it requires a higher level of investment it is quite the opposite.

In the last 18 months I've learned a lot and still have much more to utilize. Most of my conclusions and examples are relatively new and the future still has to play out. But I feel at least mentally I am playing a different game. The only thing I have made easier is formatting for pace figures while adding twice as much work! LOL!
VERY much agree.
traynor is offline   Reply With Quote Reply
Old 12-09-2016, 09:40 AM   #35
Robert Fischer
clean money
 
Robert Fischer's Avatar
 
Join Date: Sep 2006
Location: Maryland
Posts: 23,512
Quote:
Originally Posted by Robert Fischer
I'm sure a better mathematician could demonstrate a simulation with a worst-probable outcome scenario.

I would be highly entertained by a program which you could enter inputs for things like hit%(or even a range) and odds, and bankroll size, and then receive a report for a number of things such as:
  • Kelly Criterion
  • risk of ruin (a probability, perhaps given with times, eg 100trials, 1000trials, 10000trials)
  • worst-case probable scenario (fastest way to ruin within probability limits)
  • best-case probable scenario
  • a 'dice roll' button to simulate either 1 race or an entered number of races at once
  • pretty colors


I don't know if you guys ever played 'Oregon Trail' in school?
__________________
Preparation. Discipline. Patience. Decisiveness.
Robert Fischer is offline   Reply With Quote Reply
Old 12-09-2016, 11:46 AM   #36
traynor
Registered User
 
traynor's Avatar
 
Join Date: Jan 2005
Posts: 6,626
Quote:
Originally Posted by Robert Fischer
I would be highly entertained by a program which you could enter inputs for things like hit%(or even a range) and odds, and bankroll size, and then receive a report for a number of things such as:
  • Kelly Criterion
  • risk of ruin (a probability, perhaps given with times, eg 100trials, 1000trials, 10000trials)
  • worst-case probable scenario (fastest way to ruin within probability limits)
  • best-case probable scenario
  • a 'dice roll' button to simulate either 1 race or an entered number of races at once
  • pretty colors


I don't know if you guys ever played 'Oregon Trail' in school?

Assumptions (especially unstated assumptions) can be misleading. There are few areas in which this is more evident than in "race outcome prediction."
traynor is offline   Reply With Quote Reply
Old 12-09-2016, 11:56 AM   #37
Robert Fischer
clean money
 
Robert Fischer's Avatar
 
Join Date: Sep 2006
Location: Maryland
Posts: 23,512
that's 'deep'
__________________
Preparation. Discipline. Patience. Decisiveness.
Robert Fischer is offline   Reply With Quote Reply
Reply




Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

» Advertisement
Powered by vBadvanced CMPS v3.2.3

All times are GMT -4. The time now is 05:47 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright 1999 - 2023 -- PaceAdvantage.Com -- All Rights Reserved
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program
designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.