I'm running an automated value betting strategy on the greyhounds and am currently beating the BSP 65% of the time.
It is so far in profit, but I don't know if this calculation alone tells me it's a long term profitable strategy or whether I need to drill down further into, for instance, how much I'm beating the BSP by versus how much the BSP is beating me 35% of the time.
I'm wondering if anyone here would have any more info on how I would arrive at an efficient value calculation? Or is it as simple as saying that random betting would probably beat the BSP 50% of the time, so getting to 65% should be profitable in the long term?
Thanks
Value betting calculation (Greyhounds)
BSP is a good benchmark as a starting point, especially for the dogs. Beating BSP 65% of the time is a very strong start, especially if you have a lot of races behind you.TJSN1 wrote: ↑Fri Feb 02, 2024 10:20 amI'm running an automated value betting strategy on the greyhounds and am currently beating the BSP 65% of the time.
It is so far in profit, but I don't know if this calculation alone tells me it's a long term profitable strategy or whether I need to drill down further into, for instance, how much I'm beating the BSP by versus how much the BSP is beating me 35% of the time.
I'm wondering if anyone here would have any more info on how I would arrive at an efficient value calculation? Or is it as simple as saying that random betting would probably beat the BSP 50% of the time, so getting to 65% should be profitable in the long term?
Thanks
You have to take your wins and losses together tho. If you beat BSP by 1% when you beat it, but are 20% below bsp when you lose then you'll run into trouble. You PNL should tell the story (if you have had enough bets)
Thanks. Yes I guess the best way is letting time tell, before I consider upping stakes. I'm still under 1,000 bets so far so will let it run until I have a few thousand results to look back on. But BSP is still being beaten just over 65% of the time which I think is promising.Anbell wrote: ↑Sat Feb 03, 2024 6:15 amBSP is a good benchmark as a starting point, especially for the dogs. Beating BSP 65% of the time is a very strong start, especially if you have a lot of races behind you.TJSN1 wrote: ↑Fri Feb 02, 2024 10:20 amI'm running an automated value betting strategy on the greyhounds and am currently beating the BSP 65% of the time.
It is so far in profit, but I don't know if this calculation alone tells me it's a long term profitable strategy or whether I need to drill down further into, for instance, how much I'm beating the BSP by versus how much the BSP is beating me 35% of the time.
I'm wondering if anyone here would have any more info on how I would arrive at an efficient value calculation? Or is it as simple as saying that random betting would probably beat the BSP 50% of the time, so getting to 65% should be profitable in the long term?
Thanks
You have to take your wins and losses together tho. If you beat BSP by 1% when you beat it, but are 20% below bsp when you lose then you'll run into trouble. You PNL should tell the story (if you have had enough bets)
-
- Posts: 1338
- Joined: Fri Nov 20, 2015 9:38 am
Have a think about how to segment your data, back vs lays being an obvious starting point to see if any patterns...
Good luck..
Good luck..
Update (in case anyone's interested, if not this is purely for my benefit ): I continued to beat the BSP on the majority of occasions but the profits didn't really materialise as I hoped so I implemented the same strategy but just added a greenup rule, 20 seconds before the off.
Since then I've made 50pts profit on just over 500 markets (it's been running for a month 24 hours a day but on the vast majority of markets I don't get a bet matched). However, if I'd left this as a value betting strategy I would have made more than twice the amount of profit.
I'm leaving the greenup rule for now as I'm happy with the consistent profits (profitable almost every day) rather than the wild variance of value betting. However, I may change my tune further down the line.
Since then I've made 50pts profit on just over 500 markets (it's been running for a month 24 hours a day but on the vast majority of markets I don't get a bet matched). However, if I'd left this as a value betting strategy I would have made more than twice the amount of profit.
I'm leaving the greenup rule for now as I'm happy with the consistent profits (profitable almost every day) rather than the wild variance of value betting. However, I may change my tune further down the line.
Good to hear! Have you considered running the original system in parallel for small stakes?TJSN1 wrote: ↑Thu Mar 07, 2024 11:34 amUpdate (in case anyone's interested, if not this is purely for my benefit ): I continued to beat the BSP on the majority of occasions but the profits didn't really materialise as I hoped so I implemented the same strategy but just added a greenup rule, 20 seconds before the off.
Since then I've made 50pts profit on just over 500 markets (it's been running for a month 24 hours a day but on the vast majority of markets I don't get a bet matched). However, if I'd left this as a value betting strategy I would have made more than twice the amount of profit.
I'm leaving the greenup rule for now as I'm happy with the consistent profits (profitable almost every day) rather than the wild variance of value betting. However, I may change my tune further down the line.
- jamesedwards
- Posts: 3153
- Joined: Wed Nov 21, 2018 6:16 pm
Greening is like giving up a small piece of your pie to ensure you have pie today. Important for those without much pie, but illogical once you have a good fridge-full. In the long term you're just giving away pie.
jamesedwards wrote: ↑Fri Mar 08, 2024 1:39 pmGreening is like giving up a small piece of your pie to ensure you have pie today. Important for those without much pie, but illogical once you have a good fridge-full. In the long term you're just giving away pie.
This is why liquidity suffers, nobody exits anymore!
Interesting - how is your system going now over 6 months later?TJSN1 wrote: ↑Thu Mar 07, 2024 11:34 amUpdate (in case anyone's interested, if not this is purely for my benefit ): I continued to beat the BSP on the majority of occasions but the profits didn't really materialise as I hoped so I implemented the same strategy but just added a greenup rule, 20 seconds before the off.
Since then I've made 50pts profit on just over 500 markets (it's been running for a month 24 hours a day but on the vast majority of markets I don't get a bet matched). However, if I'd left this as a value betting strategy I would have made more than twice the amount of profit.
I'm leaving the greenup rule for now as I'm happy with the consistent profits (profitable almost every day) rather than the wild variance of value betting. However, I may change my tune further down the line.
I too have found a system with a similar circumstance to yours. I was told to get a P-value for my system which can help to tell you how likely your results are based on luck versus skill. I had no idea how to do that but followed some simple instructions from Grok (AI) and my results was 94.9% confident it was not luck. Which is a good start!
It's only based on 5 months of racing data though and average odds of bets taken is around 20(!) so I think i'm going to need a LOT more data to be sure. But the profit is showing considerably well. I could remove my top 5 highest odds winners, and still be in a nice profit.
I removed the green up rule soon after I posted this and left it as a value betting strategy. It's been profitable in every month since, except September when I got over confident and increased my stakes by way too much, encountered a downswing and nearly blew my entire bank (which isn't that much tbf)B-Rex wrote: ↑Fri Oct 25, 2024 3:40 pmInteresting - how is your system going now over 6 months later?TJSN1 wrote: ↑Thu Mar 07, 2024 11:34 amUpdate (in case anyone's interested, if not this is purely for my benefit ): I continued to beat the BSP on the majority of occasions but the profits didn't really materialise as I hoped so I implemented the same strategy but just added a greenup rule, 20 seconds before the off.
Since then I've made 50pts profit on just over 500 markets (it's been running for a month 24 hours a day but on the vast majority of markets I don't get a bet matched). However, if I'd left this as a value betting strategy I would have made more than twice the amount of profit.
I'm leaving the greenup rule for now as I'm happy with the consistent profits (profitable almost every day) rather than the wild variance of value betting. However, I may change my tune further down the line.
I too have found a system with a similar circumstance to yours. I was told to get a P-value for my system which can help to tell you how likely your results are based on luck versus skill. I had no idea how to do that but followed some simple instructions from Grok (AI) and my results was 94.9% confident it was not luck. Which is a good start!
It's only based on 5 months of racing data though and average odds of bets taken is around 20(!) so I think i'm going to need a LOT more data to be sure. But the profit is showing considerably well. I could remove my top 5 highest odds winners, and still be in a nice profit.
This month I lowered my stakes back to where I started and it's back in profit. I have learned my lesson and will scale up again gradually.
To gauge value, I store the green all profit value at the start time and export it to a CSV. The sum of those values increases over time which to me indicates I am getting value bets.
Interestingly, I implemented a very similar strategy using Bet Angel for Betdaq, although the ruleset is more rudimentary I am finding the profits are more consistent. No idea on the value I'm getting but I get a lot of bets matched in illiquid markets like Irish greyhounds.