6f and 7f lay markets query

[kenfrost]

Hi folks

Since October last year I have been running a very simple lay bot on the 6f and 7f in play uk/irish horse markets.

It runs on all those markets where there are 6 or more runners.

It is in profit, and out of all of the races it has laid since October only 5 have won.

However, of those 5 winning races, 3 of those wins were in May this year.

My questions are these:

-has there been some change in the markets/races during May that might have caused this bunching of wins?

- is this merely a statistical blip that will even out over time?

- is this an averaging down of my overall profitable performance that will in the end eat away all of my profits to zero/below zero if I continue to run the bot?

Thoughts/comments welcome.

Thanks

[Fugazi]

The flat season is well underway come May.

Hard to say without more detail on the strategy, though equally you don't want to give too much away. My gut feeling is if it is a very simple strategy then it is probably a result of variance. May be a losing system that just had a lucky start. But wouldn't like to discourage without more info really.

I posted a thread recently that back tested greyhounds and it found there was no profitable system if blindly backing / laying at BSP

[kenfrost]

The flat season is well underway come May.

Hard to say without more detail on the strategy, though equally you don't want to give too much away. My gut feeling is if it is a very simple strategy then it is probably a result of variance. May be a losing system that just had a lucky start. But wouldn't like to discourage without more info really.

I posted a thread recently that back tested greyhounds and it found there was no profitable system if blindly backing / laying at BSP

Thanks. I have to say my gut is telling me I have had a good run, and the 3 losses in May might well be the canary warning me to get out whilst ahead

[Fugazi]

The flat season is well underway come May.

Hard to say without more detail on the strategy, though equally you don't want to give too much away. My gut feeling is if it is a very simple strategy then it is probably a result of variance. May be a losing system that just had a lucky start. But wouldn't like to discourage without more info really.

I posted a thread recently that back tested greyhounds and it found there was no profitable system if blindly backing / laying at BSP

Thanks. I have to say my gut is telling me I have had a good run, and the 3 losses in May might well be the canary warning me to get out whilst ahead

Could be worth explaining the strategy to chat gpt and ask it if there are pitfalls. Chat gpt wont be great at answering that but it might just come up with something you hadn't considered

[kenfrost]

The flat season is well underway come May.

Hard to say without more detail on the strategy, though equally you don't want to give too much away. My gut feeling is if it is a very simple strategy then it is probably a result of variance. May be a losing system that just had a lucky start. But wouldn't like to discourage without more info really.

I posted a thread recently that back tested greyhounds and it found there was no profitable system if blindly backing / laying at BSP

Thanks. I have to say my gut is telling me I have had a good run, and the 3 losses in May might well be the canary warning me to get out whilst ahead

Could be worth explaining the strategy to chat gpt and ask it if there are pitfalls. Chat gpt wont be great at answering that but it might just come up with something you hadn't considered

Thanks, I asked various AI bots about my strategy. Their answers were consistent "the horse might win". Which of course is absolutely correct

[Fugazi]

Thanks. I have to say my gut is telling me I have had a good run, and the 3 losses in May might well be the canary warning me to get out whilst ahead

Could be worth explaining the strategy to chat gpt and ask it if there are pitfalls. Chat gpt wont be great at answering that but it might just come up with something you hadn't considered

Thanks, I asked various AI bots about my strategy. Their answers were consistent "the horse might win". Which of course is absolutely correct

Yeah can be rubbish with stuff like that. Just doesnt have enough user data on nuances of bet trading. I have to prompt it a lot and really explain when i give it data to look at. Storing data and getting it to analyse it is what you need to do to progress

[kenfrost]

Could be worth explaining the strategy to chat gpt and ask it if there are pitfalls. Chat gpt wont be great at answering that but it might just come up with something you hadn't considered

Thanks, I asked various AI bots about my strategy. Their answers were consistent "the horse might win". Which of course is absolutely correct

Yeah can be rubbish with stuff like that. Just doesnt have enough user data on nuances of bet trading. I have to prompt it a lot and really explain when i give it data to look at. Storing data and getting it to analyse it is what you need to do to progress

thanks

[Anbell]

Yeah can be rubbish with stuff like that. Just doesnt have enough user data on nuances of bet trading. I have to prompt it a lot and really explain when i give it data to look at. Storing data and getting it to analyse it is what you need to do to progress

I'm curious about how you ask for help in evaluating strategies? (I've only used it for coding help)

[Fugazi]

Yeah can be rubbish with stuff like that. Just doesnt have enough user data on nuances of bet trading. I have to prompt it a lot and really explain when i give it data to look at. Storing data and getting it to analyse it is what you need to do to progress

I'm curious about how you ask for help in evaluating strategies? (I've only used it for coding help)

I used to just say "analyse this, find a winning strategy" it would churn out some garbage like "pick all the horses that won" I then explain further.

What I have started doing now is asking it try multiple iterations of parameters. Eg in a "find the gap strategy" I may ask it : test this strategy for gaps of 1, 2 ,3 ,4 etc all the way to gaps of ten. Also test it for the time interval 5mins to 3 mins in 10 second steps.

It will send you lots of garbage scripts initially. Just have to be explain in a lot of detail what the files are you are sending and how trading works. If you get lazy with the prompts it will go in circles. Sometimes the best thing is to ask it to send you a small sample of example data and get it to show its workings. Then you can see what silly assumptions its making

Its important any data you send it the headings are clear. Youll spend an hour of it failing to sort your spreadsheet for you otherwise

Nb. Find the gap would be a bad starting point. It will assume it's getting matched everytime and making a tonne of profit

[jamesedwards]

Hi folks

Since October last year I have been running a very simple lay bot on the 6f and 7f in play uk/irish horse markets.

It is in profit, and out of all of the races it has laid since October only 5 have won.

However, of those 5 winning races, 3 of those wins were in May this year.

I'm guessing you're laying high price horses?

During the off-season for flat all races would have been on an all-weather surface.

The flat season starts in May and there will be lots more unknowns such as horses running for the first time of the year. Races will also be mostly on turf which I expect is a less consistent surface.

Blindly laying high price horses is unlikely to work in the long run. You need to pick and choose carefully.

[kenfrost]

Hi folks

Since October last year I have been running a very simple lay bot on the 6f and 7f in play uk/irish horse markets.

It is in profit, and out of all of the races it has laid since October only 5 have won.

However, of those 5 winning races, 3 of those wins were in May this year.

I'm guessing you're laying high price horses?

During the off-season for flat all races would have been on an all-weather surface.

The flat season starts in May and there will be lots more unknowns such as horses running for the first time of the year. Races will also be mostly on turf which I expect is a less consistent surface.

Blindly laying high price horses is unlikely to work in the long run. You need to pick and choose carefully.

Thanks, the change from off season may well account for the 3 losses in May.

Bit I thought the flat season started in March?

[jamesedwards]

Hi folks

Since October last year I have been running a very simple lay bot on the 6f and 7f in play uk/irish horse markets.

It is in profit, and out of all of the races it has laid since October only 5 have won.

However, of those 5 winning races, 3 of those wins were in May this year.

I'm guessing you're laying high price horses?

During the off-season for flat all races would have been on an all-weather surface.

The flat season starts in May and there will be lots more unknowns such as horses running for the first time of the year. Races will also be mostly on turf which I expect is a less consistent surface.

Blindly laying high price horses is unlikely to work in the long run. You need to pick and choose carefully.

Thanks, the change from off season may well account for the 3 losses in May.

Bit I thought the flat season started in March?

Correct, but May is when the turf flat really gets going.

[kenfrost]

I'm guessing you're laying high price horses?

During the off-season for flat all races would have been on an all-weather surface.

The flat season starts in May and there will be lots more unknowns such as horses running for the first time of the year. Races will also be mostly on turf which I expect is a less consistent surface.

Blindly laying high price horses is unlikely to work in the long run. You need to pick and choose carefully.

Thanks, the change from off season may well account for the 3 losses in May.

Bit I thought the flat season started in March?

Correct, but May is when the turf flat really gets going.

Thanks. I will put my system back on hold until October.

[Fugazi]

Thanks, the change from off season may well account for the 3 losses in May.

Bit I thought the flat season started in March?

Correct, but May is when the turf flat really gets going.

Thanks. I will put my system back on hold until October.

Honestly worth learning how to backtest with python, could save you 3 years of laying the outsiders

[kenfrost]

Correct, but May is when the turf flat really gets going.

Thanks. I will put my system back on hold until October.

Honestly worth learning how to backtest with python, could save you 3 years of laying the outsiders

Thanks

backtest with python..you've completely lost me ?