• Hi Guest, The forum will be moving hosts on 26 July and as such will be closed from Midday until the move has completed.
    As we will be with new hosts it may take a while before DNS get updated so it could take while before you can get back on the forum.
    I think it will take at least 4 hours but could easily be 48!
    Ark Royal
  • There seems to be a problem with some alerts not being emailed to members. I have told the hosts and they are investigating.
  • Sorry for the ongoing issues that you may have been experiencing whilst using the forum lately

    It really is frustrating when the forum slows down or Server Error 500 pops up.

    Apparently the hosts acknowledge there is a problem.
    Thank you for using our services and sorry for the experienced delay!
    Unfortunately, these errors are due to a higher server load. Our senior department knows about the issue and they are working towards a permanent resolution of the issue, however, I'd advise you to consider using our new cPanel cloud solutions: https://www.tsohost.com/web-hosting


    I will have to investigate what the differences are with what We have know compared to the alternative service they want us to migrate to.
    Keep safe.
    AR

Luke's Ratings/Starting Price Predictions

LukeyBoy

Yearling
Good Morning,
Just a sheet for a little idea/test of mine.If we say that a Horse's true chance is at BFSP surely there must be some logic in trying to estimate it?
The idea is to try and use it and develop it to some sort of trading strategy or something or just as a good old fashion backing strategy.
The data has been collected purely from HRB to get the estimates.Obviously in the case of non-runners the ratings will need to be normalised back to a 100% book.
I'm still gathering more data and refining it but this is my first attempt.
 

Attachments

Last edited:

LukeyBoy

Yearling
Interesting. How often do you intend to update the predicted price throughout the day?
Hi mate.The sheet is only based on what I can download from HRB in the morning (not sure if forecast data or bookies odds are used).
Obviously the closer I can update to the off the stronger the correlation.For the moment it's purely take what's available to me in the morning and see how close(or not!) it's been.
Still plenty of ideas and learning to do and pretty sure it can be improved.
The sheet is easy to use with Gruss/Betfair so anyone using that can easily compare.
Only problem with the sheet is the non-runner thing and normalising back to 100 might throw everything.
 

LukeyBoy

Yearling
Todays attempt.Extra column for another Price Pred this is based on the live data from when testing started (Yesterday!) as time goes on that will get more accurate as I have more data from the forecast price.
 

Attachments

LukeyBoy

Yearling
I've had a play with tissue prices myself. I'm also working on something based off HRB data myself and once I think its up to sharing I'll post. Best of luck.
Funnily enough I had seen your Tissue prices thread before and went to try and find it when I posted mine but couldn't.
Impressive work mate love to see what you come up with in the future.
 

LukeyBoy

Yearling
Today's sheet.I'll post something up at 11-midday (If I remember) and use the BF prices available at the time instead of the Forecast prices and see the difference.
-----------------------
Added sheet that adjust things the best I could at 11am and updated the Pred BFSP.I'll add the actual BFSP's when I have them and compare.
 

Attachments

Last edited:

ArkRoyal

Administrator
Funnily enough I had seen your Tissue prices thread before and went to try and find it when I posted mine but couldn't.
This one?

 

LukeyBoy

Yearling
This one?

Yeah that's the boy found it earlier via his profile thanks none the less ArkRoyal ArkRoyal
 

LukeyBoy

Yearling
Todays 19/08/2020 Tried to iron a out a massive flaw as in the model is tested on SP data and I only ever have forecast data so I've gone through the system builder on HRB as tried to sort out some link between forecast and SP and apply it to the sheet daily.
 

Attachments

AustinDillon75

Yearling
LukeyBoy LukeyBoy I've used HRB to take the last 1,000 ten runner AW handicaps, and I find that the SP compared to BFSPs are broadly these. In short an even money shot would have the BFSP of 2.12.

1597843280243.png

I'm with the idea of trying to gauge BFSP as its close to a 100% book so my take on the 4:20 at York (admittedly not AW but principle the same) is this:

1597843695283.png

One of the second, third and fourth gets backed in late and that whole book then goes out of the window doesn't it?
Can we guess what horses are going to possibly get the late money that affects the BFSP?
 

LukeyBoy

Yearling
4.20 York%PriceValue
Dancin Inthestreet
0.302371​
3.30720056​
Dancin Inthestreet
Pink Sands
0.107194​
9.328910913​
Keep Busy
0.148078​
6.753183771​
Keep Busy
Sampers Seven
0.132752​
7.532820551​
Sampers Seven
Lady In France
0.03355​
29.80655522​
Glamorous Anna
0.075915​
13.17265698​
Glamorous Anna
Dandys Beano
0.049802​
20.07943518​
Free Love
0.095007​
10.52556063​
Free Love
Our Little Pony
0.027107​
36.89101974​
Mighty Spirit
0.028225​
35.4299143​
 

LukeyBoy

Yearling
=(0.9704*(SP_Odds%)+ (-0.01428)
If you're looking for a general fit for SP-->BFSP then the above formula is OK.I've got a better version or 3 but that one is sweet and simple.
(SP_Odds) is 1/(odds+1)
 

LukeyBoy

Yearling
I've cut it down to one Rating now.I used the sheet yesterday to concentrate on Top Rated runners and ones where my win % was greater then Betfair (perceived value) and my results were good.
So I think the way I'm going forward is the sheet is more of a Rating used to highlight overlays and top rated horses.I'll post up the BFSP's and results from todays sheet tonight and see how they got on.
Based on the model it says if horse has X,Y,Z ratings then this is kind of the Win% chance it should have based on past ratings and BFSP.
 

Attachments

LukeyBoy

Yearling
Been away for a week but continued to collect results and things.
Someone rather kindly provided me Press forecast data for over a years worth of races so If tried to build that in the model and (on paper) it's improved some of the accuracy. So here are ratings for today.
These are my first attempts and building ratings with Neural Networks don't fully understand them..but I can get an output so that's all that I'm worried about!
 

Attachments

Top