Value Betting model

[sniffer66]

There was a post recently about using ML modelling and selling the resultant model onto a syndicate. That went quiet but it got me interested in revisiting some of my old ML models and data. I've recently treated myself to an Nvidia 5060 ti GPU with 16gb VRAM and that can crunch through years of data in minutes using CUDA compared to my old CPU, taking days

I pulled out and updated my 12 years of pre off and post race data and trained a few models looking at value betting close to the off. I'm using a mixture of history data per runner and grabbing quite a bit more pre off price data until 10s from the start. That then gets passed to the model and model probability is compared to BF Implied probability. It then backs Top n runners based on my back testing (Max of 2 per race) where the model disagrees by x% - with a few filters for max\min price, runner count etc

Its not looking at individual horse data as such but modelling patterns, win rates, going , tracks and giving each runner a win probability based on the models learning. I'm using about 20 features\data points to achieve that.

I've done all the usual Train\Test data splits and tested again on Out Of Sample data so I know it's not backfitted

Put it live about 10 days ago and results are promising so far. For now its a £1 min stake live bet until I have some confidence. On Win , a 42% ROI is obviously ridiculously good so I expect a reversion back to around my back tested ROI soon. Place is ticking away, but given prices, it's not expected to be spectacular.

Drawdown on the backtest is acceptable so I'm happy to let it run in live for a while.

Will post updates as time goes on

output.png

Final value ≈ £93.90

ROI ≈ 42.3%

output (1).png

Final value ≈ £11.45

ROI ≈ 3.6%

[Euler]

The only thing I'd say to you, based upon my data and how my models are performing, is that favourites are winning slightly more often than they should be at the moment. You may want to take account of that if you're betting on favourites.

Or if you're betting on outsiders that may actually do the opposite and depress your returns, but it's worth making a note of this current situation.

[sniffer66]

The only thing I'd say to you, based upon my data and how my models are performing, is that favourites are winning slightly more often than they should be at the moment. You may want to take account of that if you're betting on favourites.

Or if you're betting on outsiders that may actually do the opposite and depress your returns, but it's worth making a note of this current situation.

Looking at the price spread its actually a mix. It's purely based on the model probability vs BF Implied and taking the largest gaps between the two, exploiting possible mis-pricing. Hopefully it keeps going as well as it's started

[Euler]

Fingers crossed!

[tico]

Hopefully it keeps going as well as it's started

It wont !! I almost guarantee it.Automation and value seeking don't work siimply because AI (and more primitive models that came earlier) cannot be subjective. It is the subjective element that dictates what value is .
There is a major movement towards automation ever since the first PC spun it's floppies and DOS was born.(to the detriment of us all)
None of it works .The problems has always been that the ability (and it is a profoundly impressive ability) for computers to gather and present data ,is somehow equated with it's ability to predict what really is a randome outcome .A computer will hazard a guess at value ,but it will be acting on a human definition ,and in my experience ,there are more human definitions of what value is in sport than there are stars in the Galaxy.It is possibly indefinable.
So the base variable has to be flawed .But it want stop glorious runs of winning ,but by the same token less-glorious runs of losing.
Computer technology (and this applies even more to AI) is a remarkable aid to winner finding ,but it wont find you any winners ,yopu have to find those yourself.
Regards
Tico

[sniffer66]

Hopefully it keeps going as well as it's started

It wont !! I almost guarantee it.Automation and value seeking don't work siimply because AI (and more primitive models that came earlier) cannot be subjective. It is the subjective element that dictates what value is .
There is a major movement towards automation ever since the first PC spun it's floppies and DOS was born.(to the detriment of us all)
None of it works .The problems has always been that the ability (and it is a profoundly impressive ability) for computers to gather and present data ,is somehow equated with it's ability to predict what really is a randome outcome .A computer will hazard a guess at value ,but it will be acting on a human definition ,and in my experience ,there are more human definitions of what value is in sport than there are stars in the Galaxy.It is possibly indefinable.
So the base variable has to be flawed .But it want stop glorious runs of winning ,but by the same token less-glorious runs of losing.
Computer technology (and this applies even more to AI) is a remarkable aid to winner finding ,but it wont find you any winners ,yopu have to find those yourself.
Regards
Tico

Yeah, I do appreciate that, but I like playing with this stuff, and have done for years. I ran a pretty successful Poker bot for about a decade back in the 2000's - but the human element in that made it a little different. I'm not expecting anything life changing to come out of it but it will be interesting to track how it does for pennies against the back test.

At the end of the day if its has more than a 2% edge in picking the right price versus the market then its a winner. Again, will be interesting to watch

On a side note, I started in IT in the days of Windows 3.1 and DOS 6.21, so get you on that. Having said that MS paid for my house, career and decent pension lol

[Euler]

Hopefully it keeps going as well as it's started

It wont !! I almost guarantee it.Automation and value seeking don't work siimply because AI (and more primitive models that came earlier) cannot be subjective. It is the subjective element that dictates what value is .
There is a major movement towards automation ever since the first PC spun it's floppies and DOS was born.(to the detriment of us all)
None of it works .The problems has always been that the ability (and it is a profoundly impressive ability) for computers to gather and present data ,is somehow equated with it's ability to predict what really is a randome outcome .A computer will hazard a guess at value ,but it will be acting on a human definition ,and in my experience ,there are more human definitions of what value is in sport than there are stars in the Galaxy.It is possibly indefinable.
So the base variable has to be flawed .But it want stop glorious runs of winning ,but by the same token less-glorious runs of losing.
Computer technology (and this applies even more to AI) is a remarkable aid to winner finding ,but it wont find you any winners ,yopu have to find those yourself.
Regards
Tico

I think you're getting ahead of yourself there a bit.

[tico]

I think you're getting ahead of yourself there a bit.
Top

On the contrary ,I am simply pointing out that as of yet a computer program has never been successful in picking an endless supply of winners ,based on value(or anything else come to that).
I know we dont call them programs anymore ,but strategies!
My conclusions are based on my own research ,back in the old days with C and VB right through to the here are now with Python .Now I create databases that I could only dream of when all of this desktop lark took off.But the essentials are the same,the math is the same,and the odds have actually gotten worse.
Whole new areas of sports betting have emerged and the net has spawned endless white papers from academia on how to win by ressurecting Eistein.. that amount basically to nothing .Github will testify to that.And of course we must never forget that the bookies also have access to the same technology as us...a lot of people forget that and I would say without little hesitation that computer technology has definately irradicated pure value betting.It's still there ,but nothing like it used to be.
The upside is the power of the comp to produce stats and facts at the drop of a hat and arrange them in a myriad of ways...no more fingering through Superform Raceform or the Timeform Blackbook.
Automation is a natural expectation from modern technology ,but I will still say that for Trading ,laying or betting the automation concept is innefective.Moreso in trading than anything else.
The days of raking in the pennies throught trading on BF from botting about have gone the way of Poker bots.
Only one answer (horse racing) study the form,know the jocks ,know the trainers ,no every fall and rise of every racecourse in the country,know every bit of the form of every horse in every race.Know your databases and spend your life looking at re-runs...do that and you WILL win
And finally(horses) keep all serious betting for the flat....
Regards
Tico

[Euler]

I was invited out to dinner last night by a Bet Angel user that's been doing very well, and I know there are lots that are doing well.

Your stance is far too negative. Your role really is to look for opportunities, not to find problems, because it's very easy to find problems.

[tico]

I was invited out to dinner last night by a Bet Angel user that's been doing very well, and I know there are lots that are doing well.

Your stance is far too negative. Your role really is to look for opportunities, not to find problems, because it's very easy to find problems.

I dont dispute there are people who use BA and other similar software and profit from it .Part automation especially for dutching (where there is a lot of money to be made) is essential .But 100% full blown press the button and go on holiday come back to a fortune (or even 2%) automation to me is just a myth.One thing that does annoy me however is the idea that trading has a higher success rate that traditional betting or laying.The failure rate (of around 94%) is similar in all disciplines.
You quoted:
'Your role really is to look for opportunities, not to find problems, because it's very easy to find problems'....
Sorry but the whole thing about gambling is that almost everything is a problem ,there are few opportunities that exist without painstaking research and a lot of risk taking .It's not a walk in the park ,it never has been and it never will be.
You can only gain positivity ,once negative expectation has been tackled,and even then dont put your shirt on it.

[Euler]

I dont dispute there are people who use BA and other similar software and profit from it .Part automation especially for dutching (where there is a lot of money to be made) is essential .But 100% full blown press the button and go on holiday come back to a fortune (or even 2%) automation to me is just a myth.

It's 100% not a myth

This was my point: you've made some valid points, and I totally get that it's not for everybody and that lots of people struggle and you have to work hard to get an edge. There are loads of people out there now that are doing perfectly fine. A lot of them are using automation on Bet Angel.

Also, the manual trading is still working. I do that every week, especially on a Saturday.

This is my main point: if you start looking for things that could go wrong or problems, you end up down this rabbit hole of negativity. In fact, there's a lot of really good stuff going on out there, but lots of people never find it because they're not looking for it.

[Euler]

It's been a theme with my entire career.

When I started, people told me I was mad, crazy, stupid; it would never work. Five years in, people said I was just being lucky. Ten years in, people sort of said, "Well, you may have something, but it's not going to last." Fifteen years in, people stopped asking. Twenty years later, I'm still getting the same questions.

[sniffer66]

I think you're getting ahead of yourself there a bit.
Top

Only one answer (horse racing) study the form,know the jocks ,know the trainers ,no every fall and rise of every racecourse in the country,know every bit of the form of every horse in every race.Know your databases and spend your life looking at re-runs..

So what's the difference between a human taking all that data, and pushing the same data into a machine learning model that is trained to recognise the same patterns ?
I'm not talking AI either but a neural net or similar.

Interesting you bring up poker after I mentioned it. Back in the day I had 5 million harvested real hands in a Db. With those numbers I could tell you which hole cards were statistically ev+ from each position. Play enough hands from the correct position and you would win regardless, completely automated. Just shove all in
That's just maths and probability