Your paper testing might not have a sufficient sample size if your strategy strays a significant distance from a 50/50 strike rate. With 50/50 a few 100 might give you a guide but one that is nearer 1/99 either way would need 1000s.
Have you tried this on several years of historical data? It's often the case that very high/low strike rate strategies can exhibit periodicity that spans years rather than months.
I think you said your strengths were programming and maths? Use that! It's a massive positive, and attack this with some hardcore analysis on a decent dataset. The free stuff is a start and goes back 10years but priority #1 and #2 for you is to start capturing data, ideally something fairly niche, and get 100 strategies tested in a few months not just 1 or 2.
How to define 'its working'
-
ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
.... and when you've proved something by backtesting, you'll find that going live with them is a whole other world of pain. 
£18k profit demonstrates some identified value and success. However, 23 subsequent losses cannot really be attributed to statistical possibility. There is a problem with method, execution or data. July losses confirm this.
Unless you can fix the flaw(s) your model is unreliable. Trading is a different concept and using it to exit losing bets is not a viable solution. Any significant variance between paper trading and actual results inidcates a subconcious bias, probably aversion or reaction to losses.
You might find that extending your dutching selections in some markets, to return zero profit, would be a step in the right direction.
Unless you can fix the flaw(s) your model is unreliable. Trading is a different concept and using it to exit losing bets is not a viable solution. Any significant variance between paper trading and actual results inidcates a subconcious bias, probably aversion or reaction to losses.
You might find that extending your dutching selections in some markets, to return zero profit, would be a step in the right direction.
Didn't realize those were betting results, you would probably need to start from scratch when learning to trade, it's a whole different game. Nobody sane would tell you to quit your job for a betting strategy. You can always revisit your betting strategy at a later time, if it doesn't become obsolete by then, you could potentially automate it with Bet Angel or at least use it to churn more commission to mitigate some of the premium charge.
You said you're programer, if so, then soon you'll want to automate your trading (if not already), and then you're gonna be hit with premium charges. So it's maybe reasonable to sacrifice good looking flat graphs for a money you can save by generating more commission by betting.
So, read about premium charges and plan your business accordingly.
So, read about premium charges and plan your business accordingly.
It's a backing strategy on longshots mainly, so Premium Charges probably won't be a problem anytime soon.
What he needs first is to gain some confidence in his method and edge.
Xeres, you said you have a background in math and programming. There are a couple of methods that could help you with analysing your results. Try a Monte Carlo simulation of your results for example. Variance can be huge when backing longshots.
As Linus said, trading out could help to smooth your profits, but you don't want to give away too much of your edge just to smooth your results. Edges are hard to find and hedging can be costly.
I would start with a more rigorous analysis of edge and variance of your system.
What he needs first is to gain some confidence in his method and edge.
Xeres, you said you have a background in math and programming. There are a couple of methods that could help you with analysing your results. Try a Monte Carlo simulation of your results for example. Variance can be huge when backing longshots.
As Linus said, trading out could help to smooth your profits, but you don't want to give away too much of your edge just to smooth your results. Edges are hard to find and hedging can be costly.
I would start with a more rigorous analysis of edge and variance of your system.
Thanks all, some good comments.
I should probably point out, my current strategy is back bets only, it's not limited to any odds ranges, if there is value and it fits the criteria then it going in the mix. I calculate a relative probability for every horse in every race, then assess the value at that moment in time, this mainly involves getting rid of short priced runners although not always. I then run a secondary process to filter that pool down so I have roughly a bet in each race give or take.
The process is fully automated from start to finish, the 'paper trading' uses the Betfair API to pull all the market info etc. so it's as close to reality as possible.
What I find is the winner will swing from my top rated horses to the next best rated, I've tried loads of ways to figure out why this is, tried lots of optimisation algorithms and run thousands of simulations but it just seems thats how it is - hence why I thought ok I will dutch the top 2, plus it keeps the losing streaks down - but the margins are smaller and the volume is basically doubled. I'm not equalising the profits or stakes with dutching either, I use a slightly modified version of kelly that I came up, it's significantly more profitable than just staking £100 or staking to make £100 regardless.
It's very frustrating to say the least!!
I am trying to figure out if I can use what I have and modify it for a trading strategy but my lack of trading experience is a bit of an issue right now.
I should probably point out, my current strategy is back bets only, it's not limited to any odds ranges, if there is value and it fits the criteria then it going in the mix. I calculate a relative probability for every horse in every race, then assess the value at that moment in time, this mainly involves getting rid of short priced runners although not always. I then run a secondary process to filter that pool down so I have roughly a bet in each race give or take.
The process is fully automated from start to finish, the 'paper trading' uses the Betfair API to pull all the market info etc. so it's as close to reality as possible.
What I find is the winner will swing from my top rated horses to the next best rated, I've tried loads of ways to figure out why this is, tried lots of optimisation algorithms and run thousands of simulations but it just seems thats how it is - hence why I thought ok I will dutch the top 2, plus it keeps the losing streaks down - but the margins are smaller and the volume is basically doubled. I'm not equalising the profits or stakes with dutching either, I use a slightly modified version of kelly that I came up, it's significantly more profitable than just staking £100 or staking to make £100 regardless.
It's very frustrating to say the least!!
I am trying to figure out if I can use what I have and modify it for a trading strategy but my lack of trading experience is a bit of an issue right now.
Yes this is very true, I was careful in my modelling, I was creating each day as if it was the actual day, doing the predictions, then moving to the next day and so on but there were still anomilies. Mainly with non-runners, I knew who they were but I didn't know what time in the day they were declared, this impacted several variables in my model, as well as the fact I'm not doing intra-day modelling either. It's been live now for over 14k races in the UK and Ire but I still feel like I could do with 2 full years so I can get a better read on any seasonal impacts.ShaunWhite wrote: ↑Thu Jun 13, 2019 11:50 pm.... and when you've proved something by backtesting, you'll find that going live with them is a whole other world of pain.
It's good fun and keeps me busy in the evenings!
..Just dashing out but thought Id post a link to a presentation by Noah Silverman
https://www.youtube.com/watch?v=5gW0PO7g6pY
In his summing, he mentions a 'portfolio' ie spreading his bets across multiple combinations. IMO this is good advice
Anyway must dash
(Xeres, Good luck with this; I cant think of a better combination skill set of Maths/programming!
https://www.youtube.com/watch?v=5gW0PO7g6pY
In his summing, he mentions a 'portfolio' ie spreading his bets across multiple combinations. IMO this is good advice
Anyway must dash
(Xeres, Good luck with this; I cant think of a better combination skill set of Maths/programming!
This is the video what set me on my journey, I got talking to a guy a work who made a ton of cash in the early days of Betfair, basically exploiting some loophole in the system. He offered to help me put some strategies in place so off I went. That was the first video I saw on it and went on to develop a model using conditional logistic regression.PeterLe wrote: ↑Fri Jun 14, 2019 2:29 pm..Just dashing out but thought Id post a link to a presentation by Noah Silverman
https://www.youtube.com/watch?v=5gW0PO7g6pY
In his summing, he mentions a 'portfolio' ie spreading his bets across multiple combinations. IMO this is good advice
Anyway must dash
(Xeres, Good luck with this; I cant think of a better combination skill set of Maths/programming!
The model is specified was more technical than mine, i didn't use any regularisation techniques, or at least I couldn't get them to work as well as just doing the clogit model.
I have since recoded in neural networks because it handles the interaction terms much better, as he sayd trying to code them all up would mean thousands of variables. I use R for all my coding so I am limited to what it can do and I can only afford a modest sized VPS so again I am constrained in those ways.
Yes optimising over many types of bets is something I did think of, I usually pick 4/5 races a day where I get 1st and 2nd place for example. It's just a question of time I guess, I often thought about seeking help but literally nobody I know has any interest in this field. Nobody want to spend their time coding up models in the evening - why not???
Betting is such a dirty word though, as soon as I mention it to anyone they think I have a gambling problem, tradin on the other hand seems to spark more interest.
Yes, from my experience, of course. I'm not good at defining ideal closing positions so I prefer to leave bets unclosed. Modified most of my bots from trading to pure betting and now they generate more commissions and more profit, can't be hurt by outages and by chasing possibility to close my position in illiquid markets. But this's not fit for a scalpers, of course.
So, if you came from gambling background, have a solid and consistent edge and can automate it - there is no point in trading. Just make bets, collect your winnings and be happy.
Yeah? Then just say you have a superduper bot and it earns you a few tons per month, and then enjoy the reaction of those incompetent amateurs.!Betting is such a dirty word though, as soon as I mention it to anyone they think I have a gambling problem, tradin on the other hand seems to spark more interest.
Last edited by napshnap on Fri Jun 14, 2019 3:39 pm, edited 1 time in total.