Smartgambler
Pro-Punter

Go Back   OZmium Sports Betting and Horse Racing Forums > Public Forums > Horse Race Betting Systems
User Name
Password
Register FAQ Search Today's Posts Mark all topics as read

To advertise on these
forums, e-mail us.

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 21st March 2018, 03:34 PM
evajb001 evajb001 is offline
Member
 
Join Date: Oct 2009
Posts: 1,278
Default Rating Adjustment Help or Ideas Needed

Hi all,

I'm playing around with a ratings system/method that basically looks at a horses last 5 starts and any other races that are at a similar distance AND similar tack conditions to todays race. It takes the starting price of the horse in that race and adjusts for field size and class of race, for example:

Horse 1 last start raced in a 8 horse field, starting price was $5, race had $30,000 prizemoney.

Today's race is a 10 horse field with $50,000 prizemoney. So that $5 starting price would need to be adjusted upwards because the horse is now in a higher class race with a higher level of runners.

I've used historical data and done the math on that side of things as my starting point which I'm fine with. I then weight those new prices based on days since that start as generally recent form/prices speak greater volumes as to a horses chances today. This gives an initial price point / rating based on the assumption that in general a horses starting price is relatively accurate in terms of its chances of winning that given race and thus can somewhat be dragged forward to give a starting point of what a horses chances will be today.

Now obviously so far I've simply used the input of starting price and adjusted for prizemoney (class) and field size. What i'm posting here for is to ask everyone in general what elements they believe are major factors of changing a horses starting price from today's race vs their last race if all else is held equal.

So for instance a horse is running today at the same venue, same field size, same race conditions, same everything EXCEPT its running at a race 200m longer then last start - is this a significant factor for influencing that horses price now?

The obvious ones I can think of are weight change, distance change and the actual performance of a horse in that given race (i.e. won, lost by 1L or lost by 10L), but are there any others people can suggest or think of? I'm open to any and all suggestions as long as people have a solid belief that it has a reasonable influence on the price of a horse if all else were held equal. So for instance something like being in barrier 2 last start to barrier 12 this start may have a marginal influence on the strike rate of a horse but you are rewarded with a marginally higher price as well so its potentially not a large enough influence to include in the rating/price adjustment.

I hope that all makes some form of sense. Its just a little fun project I'm working on to form a rating given I just like messing around in excel and looking at the figures. As I said I'm open to all suggestions or if you have any questions then fire away. Once i've tested it and fiddled with it a bit I might try and post the ratings for a 1 or 2 week period and see how it goes.

Cheers

Josh
Reply With Quote
  #2  
Old 21st March 2018, 05:12 PM
Try Try Again Try Try Again is offline
Member
 
Join Date: Nov 2007
Posts: 1,425
Smile

Hi Josh,

The one that I have always struggled with is the interference that occurs in each and every race and how to numerically penalise or reward horses affected. If a horse with a SP $5 was significantly affected in the run and finished 8L behind the winner how will you treat it?

How will you handle the Trainer and Jockey and how horses are overbet because they are trained by D.Weir, for example, or ridden by H.Bowman?

Good luck with your endeavours - not an easy task as there are so many variables that influence a race.

Try Try Again
Reply With Quote
  #3  
Old 21st March 2018, 10:15 PM
UselessBettor UselessBettor is offline
Member
 
Join Date: Sep 2011
Posts: 3,589
Default

Josh,

Avoid my advice at your own peril

I have been down this road and I can save you many hours of hard work. Ask your self these simple questions :

What are you trying to predict and why are you predicting it?

Are you trying to predict the winner ?
Are you trying to predict the % chance of winning ?
Are you trying to predict if a horse will run to its optimum peak?
Are you trying to predict if a horse will run below par ?
Are you trying to predict the true odds for a horse ?
Are you trying to predict the final odds for a horse ?
Are you trying to predict the chances in a race?

Without knowing what you are trying to predict you can't accurately measure if your rating is working or even good. You have explained a method to come up with a number but you haven't really explained what that number is suppose to be and how you would use it.

Next thing... to save you thousands upon thousands of hours... The best prediction method in the market of a horses chances of winning are the final odds. If you can accurately predict the final odds you have a great ratings method and will be be a millionaire in no time. But its hard to do early and much easier to do late.

Here is what I find are the best methods to use:
1. Take the odds as your about to place a bet.
2. Find a reliable way to make them more accurate.
3. Profit.

The advice might be simple but doing it is also very simple when you have data to check against. Simplest example I can give you is inside barriers are over bet and outside barriers are under bet. Do this adjustment to the final odds and you will move slightly closer to break even. Do this for a number of variables and your on your way to a very profitable ratings method.

I wish you luck in your current ratings method and if you wish to go down that path just look up impact values for horse racing. At least that will give you a good starting point.
Reply With Quote
  #4  
Old 24th March 2018, 08:57 AM
demodocus demodocus is offline
Member
 
Join Date: Apr 2010
Posts: 187
Default

Josh,
You're trying to predict an outcome in a chaotic system.
The SP PRICE simply reflects the collective opinion of punters which is reflected in the weight of money.
The sad thing is .... there is NO Golden Key or Bookies Nightmare or Lost Gold Mine.
What all hands and the cook have to do is find out how to eliminate from winning the horses that can't possibly win (but they do).

Stay away from US races ..... concentrate on UK races and you'll find that price will help.
Reply With Quote
  #5  
Old 24th March 2018, 09:35 AM
Lord Greystoke Lord Greystoke is offline
Member
 
Join Date: Mar 2012
Posts: 3,199
Default

Quote:
Originally Posted by demodocus
What all hands and the cook have to do is find out how to eliminate from winning the horses that can't possibly win (but they do).



The starting point I believe although its taken me what feels like an eternity to get this point! No-where near as difficult as a lottery imho if you take onboard that there are various factors which contribute to the 'impossibles' that can still get up and win and that every now and then we can pinch a glimpse of some of these. One word comes to mind, more often than not.

Trainer

LG
__________________
Many punters on here can make money just looking at the prices of the horses without having a clue on the form analysis - Vortech

Last edited by Lord Greystoke : 24th March 2018 at 09:39 AM.
Reply With Quote
  #6  
Old 27th March 2018, 10:57 AM
Puntz Puntz is offline
Member
 
Join Date: Jan 2012
Posts: 556
Default

UB, interesting set of questions.
I think, or I'm sure some apply and some don't depending on strategy.

Quote:
What are you trying to predict and why are you predicting it?

Are you trying to predict the winner ?
Are you trying to predict the % chance of winning ?
Are you trying to predict if a horse will run to its optimum peak?
Are you trying to predict if a horse will run below par ?
Are you trying to predict the true odds for a horse ?
Are you trying to predict the final odds for a horse ?
Are you trying to predict the chances in a race?


Are you trying to predict the winner ?
After testing, automation in chronological time order of every race in Australia every day, AFTER results ( Interim for testing purposes, Paying is ideal, but less races are tested due to a linear strategy, meaning one does not do the next race until the previous bet race's results have come through),
You eventually get winners, but also a loosing streak, and for some quirk of nature, loosing streaks seem longer than winning streaks,
why ?


Are you trying to predict the % chance of winning ?
No, because I know I will get a win.

Are you trying to predict if a horse will run to its optimum peak?
It helps to know, sometimes prices and form may offer an edge, sometimes just walk away for the day, these are animals...they have their moments.

Are you trying to predict if a horse will run below par ?
Same as above answer


Are you trying to predict the final odds for a horse ?
No, one can only assume it will be within 20 percent less take-out if pari-mutual markets. Fixed odds are what they are, Fixed, so ya know what ya gonna get.

Are you trying to predict the chances in a race?
No, because a sound money management system factors win/loss because of the inevitable win.
However, you try to eliminate much of, A: hype, and
B: calculate a method of Value suitable or accepted by the "buyer", punter.

For example, if a horse is showing 3.50, less a variable percent, in this example 20% less than 3.50, (2.80 if that is your accepted lowest), then you have the choice to buy/bet, of no less than 2.80,( win or place ).
Because chances are in pari-mutual @ 3.50 at time of bet, 2.80 is probably what you get, maybe less, hopefully more.
Reply With Quote
  #7  
Old 27th March 2018, 06:19 PM
UselessBettor UselessBettor is offline
Member
 
Join Date: Sep 2011
Posts: 3,589
Default

Quote:
Originally Posted by Puntz
UB, interesting set of questions.
I think, or I'm sure some apply and some don't depending on strategy.


Yes you are right. Those questions were aimed at josh on what he actually was using the rating for. Ratings don't have to rate the bet chance and they can focus on other things (ie fitness, likely odds, etc).
Reply With Quote
  #8  
Old 28th March 2018, 07:43 AM
kiwi kiwi is offline
Member
 
Join Date: Jan 1970
Location: new zealand
Posts: 1,103
Question

I read some time ago American professionals are more interested in previous speed figures, the pace of today's race and any possible previously revealed track bias.
They are more interested in how a previous race was run, rather than
than who took part.
Somebody may have read Allan Potts books and be able to explain further.
__________________
At the feast of ego everyone leaves hungry
Reply With Quote
  #9  
Old 28th March 2018, 08:10 AM
demodocus demodocus is offline
Member
 
Join Date: Apr 2010
Posts: 187
Default

Quote:
Originally Posted by Lord Greystoke
The starting point I believe although its taken me what feels like an eternity to get this point! No-where near as difficult as a lottery imho if you take onboard that there are various factors which contribute to the 'impossibles' that can still get up and win and that every now and then we can pinch a glimpse of some of these.


LG, the more I look the more I find races with BIG pools are the most accurate in terms of prices. Looking back over my auto-lay punting for the last month or so I find I haven't had a loss in the UK, 4 hits in AUS, and 9 in USA. All in the same 'price band' but the pool size matches the risk. I've now stopped fiddling with the USA ..... too unreliable, and results match pool size.
Reply With Quote
  #10  
Old 28th March 2018, 08:23 PM
Lord Greystoke Lord Greystoke is offline
Member
 
Join Date: Mar 2012
Posts: 3,199
Default

Quote:
Originally Posted by demodocus
LG, the more I look the more I find races with BIG pools are the most accurate in terms of prices. Looking back over my auto-lay punting for the last month or so I find I haven't had a loss in the UK, 4 hits in AUS, and 9 in USA. All in the same 'price band' but the pool size matches the risk. I've now stopped fiddling with the USA ..... too unreliable, and results match pool size.


More accurate = less risk of unpredictability = higher returns. I like the sound of that. When you say BIG, how much bigger are UK pools v Aus, Demodocus?

Cheers LG
__________________
Many punters on here can make money just looking at the prices of the horses without having a clue on the form analysis - Vortech
Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

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

vB code is On
Smilies are On
[IMG] code is On
HTML code is On
Forum Jump



All times are GMT +10. The time now is 05:17 AM.


Powered by: vBulletin Version 3.0.3
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
2008 OZmium Pty. Ltd. All rights reserved . ACN 091184655