Smartgambler
Pro-Punter

Go Back   OZmium Sports Betting and Horse Racing Forums > Public Forums > General Topics
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 8th April 2015, 08:33 AM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Default How much is enough?

I was having a play round yesterday with R&S's Neurals , once a favourite topic on this forum.

On their Neural racecards (R&S) have a NR rating and a PR price for each runner. The NR is the neural score and the PR is related to the NR and converted to $. Now when I say related, its not a straight mathematical relationship as some tinkering goes on in the conversion algorithm. I tried to crack it but couldn't, so I plotted a line graph instead to try and understand the conversion.

We all know about overround and that's how totes and bookies make their profits, but how much profit is enough, as the punters don't help themselves by weighting their betting mindset on the first 3 favourites and in so doing help the totes and bookmakers to reap even more than a 15 - 20% margin! When R&S convert a NR rating to Price they must be aware of this extra over bias and seek to replicate it. They do a pretty good job of it but it's a mystery algorithm for me. I took two races from yesterday at random. If you look at Goulburn 6 firstly. There's a NR rating and that's converted to implied probability (NR%). Take horse #6, it's NR% is 15.08% but it's implied price (PR) is $4.40 converted to implied probability (PR%) is 23.81%. Why the big difference?

Well if you look at the graph of the red (NR%) and orange (PR%) you'll note that the 1st fave has a PR% much higher than its rating (NR%) this line falls rapidly as the price rank increases and after the 4th fave, the lines cross and then the ranks 5 onwards allegedly have a better chance of winning than their price would indicate. The blue line is the Fixed price % that the tote sets which is in agreement with the false representation of probability of the orange PR%. (I converted everything to 100% for demonstration purposes)

OK, for some this might be confusing, what's the bottom line? Firstly, it goes part of the way to explaining to yourself, when a $10 plus horse get's up, why it isn't really a surprise? Secondly, people over bet the first favourite heaps, lesser the 2nd, 3rd, 4th favourite's price is true and then they under bet the rest.

How to counter this:
Backers - Don't follow the sheep over the cliff, stick with your ratings or stats and you'll find overlays a plenty with ranks 5 onwards, just might have to wait awhile.

Layers - Your underlays are in the first 2 or 3 faves.

Now every race is different though and I'll refer you to Goulburn 7. In this race there was an odds on fave and punters were falling over themselves to back it way above it's probability. Even though it won, it was 3 times a better value lay bet than it was a Back bet.

One last thing, the rank order for F Price$ does not necessarily match that of the rank order I've shown for NR% and PR%. Therein lies the conundrum as sometimes ratings are correct and sometimes not. One thing, I will say though, is divergence between ratings and price is more profitable than consensus for one simple reason, they're overbet.
Attached Thumbnails
Click image for larger version

Name:  Goulburn67pic.jpg
Views: 657
Size:  92.6 KB  

Last edited by Rinconpaul : 8th April 2015 at 08:36 AM.
Reply With Quote
  #2  
Old 8th April 2015, 09:59 AM
evajb001 evajb001 is offline
Member
 
Join Date: Oct 2009
Posts: 462
Default

RCP,

Interesting stuff given I'm more of a ratings person and frame my own market. I'm sure I still don't have the right way of going about it but I usually just use it to essentially rank my ratings rather than for direct comparison to the available prices. Although I have found that if an overlay is available (between 1x to 3x my calculated price) that it is a LOT more profitable then underlays or larger overlays, based on betfair prices.

For the races you covered RCP the following are the prices i came up with using my pricing method.

R1
3.23
9.16
9.35
12.02
17.16
16.96
17.69
21.56
26.41
24.18
26.74
31.84
43.12

R2
4.60
5.00
7.32
9.39
9.53
14.58
17.16
21.04
25.91
46.70

You'll note that my way of pricing gives more chance to the longershots winning compared to the odds which is generally in line with what your suggesting. I really wish I had the time, patience and know-how to set up my ratings to only bet overlays and lay underlays etc as I think it would be a worthwhile exercise.

I might go back over my recorded data thus far and see what laying underlays and/or laying the top fave when its not one of my top raters is like.
Reply With Quote
  #3  
Old 8th April 2015, 10:30 AM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Default

Well Josh, it might be comforting to know that your ratings are pretty well line ball with R&S PR prices.
Attached Thumbnails
Click image for larger version

Name:  Goulburn67J.JPG
Views: 627
Size:  43.3 KB  
Reply With Quote
  #4  
Old 8th April 2015, 11:03 AM
evajb001 evajb001 is offline
Member
 
Join Date: Oct 2009
Posts: 462
Default

Is that a good thing or a bad thing you think RCP? haha.

For interests sake i've found the ranking of the 'Est' scores in the 'Rating' tab for R&S to be pretty solid for its top selection, could just be hitting a good patch though.

I'll post back and let you know how my ratings go at laying the 1st, 2nd and 3rd faves if they are underlays and/or not in my top selections give the favoritism bias your talking about.
Reply With Quote
  #5  
Old 8th April 2015, 11:15 AM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Default

Quote:
Originally Posted by evajb001
Is that a good thing or a bad thing you think RCP? haha.

For interests sake i've found the ranking of the 'Est' scores in the 'Rating' tab for R&S to be pretty solid for its top selection, could just be hitting a good patch though.

I'll post back and let you know how my ratings go at laying the 1st, 2nd and 3rd faves if they are underlays and/or not in my top selections give the favoritism bias your talking about.



I use them each day but only to Lay contrarian values between PR$ and Back $ for < $10 Back price faves. There is a threshold though i.e. > PR$ and also sub groups, within the < $10 price range, to match. It might only throw up one opportunity a day but it's a good Lay payer.

I'm trying to automate it but the neural scraper that Shaun made yonks ago, for some reason, only picks up the NR value, not the PR$ value. So it's a manual operation for now. I think there's some worthwhile exotic pickups to be had though (tri's & 1st 4's).
Reply With Quote
  #6  
Old 8th April 2015, 11:19 AM
evajb001 evajb001 is offline
Member
 
Join Date: Oct 2009
Posts: 462
Default

I could give you my pricing technique if you like and you can adapt it to see how close you can get it to the PR$ prices and possibly use that? Or have you tried other pricing techniques to see if they work too? That way you could automate it off the NR value and not need to manually enter.

Note: I could only manage to get the 'Rating' tab to pull into excel. The neurals and work sheet tabs dont seem to work

Last edited by evajb001 : 8th April 2015 at 11:23 AM.
Reply With Quote
  #7  
Old 8th April 2015, 11:27 AM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Default

Quote:
Originally Posted by evajb001
I could give you my pricing technique if you like and you can adapt it to see how close you can get it to the PR$ prices and possibly use that? Or have you tried other pricing techniques to see if they work too? That way you could automate it off the NR value and not need to manually enter


I'm pretty happy with the PR$, as I've already built up a database and filters to suit, so I'll pass on the rating offer, thanks very much.

R&S have a csv file that you can open manually, and that has the PR$ figure, but importing it automatically is my issue.
Cheers
Reply With Quote
  #8  
Old 8th April 2015, 12:33 PM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Default

Good example, my first race Gawler #1. Layed Harmonium Bells, $8.65 SP for a grand, PR $61. The winner was ranked 4th at $8.35 with a PR of $4.80. I don't backem' though.
Reply With Quote
  #9  
Old 8th April 2015, 03:05 PM
Shaun Shaun is offline
Member
 
Join Date: Jan 1970
Location: Western Australia
Posts: 3,456
Default

Quote:
Originally Posted by Rinconpaul
I'm pretty happy with the PR$, as I've already built up a database and filters to suit, so I'll pass on the rating offer, thanks very much.

R&S have a csv file that you can open manually, and that has the PR$ figure, but importing it automatically is my issue.
Cheers



Problem solved
__________________
One Drive

"If the corporates are treating you poorly , just go elsewhere."
"If they need you , they will soon find out."
"If you need them , you will soon find out."
--moeee
_______________________________________________
Reply With Quote
  #10  
Old 8th April 2015, 03:11 PM
Rinconpaul Rinconpaul is offline
Banned
 
Join Date: Feb 2013
Posts: 755
Thumbs up

Quote:
Originally Posted by Shaun
Problem solved


I've said it before, and I'll say it again, "Shaun is one of a kind! Not only is he brilliant and benevolent but with a heart of gold to match"

Cheers
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 07:53 PM.


Powered by: vBulletin Version 3.0.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
©2008 OZmium Pty. Ltd. All rights reserved . ACN 091184655