Pre-off automation challenge

[jamesedwards]

I've been trading full-time on in-running horse racing for several years but never really touched pre-off trading.

During the recent weather-related downtime, I thought I'd challenge myself to create a profitable Greyhound automation using only;
> info shared on this forum
> a good knowledge of Guardian
> a working knowledge of Betfair market forces

After two weeks of testing things are looking promising, and I go live tomorrow...

[Dallas]

Good luck with it

[Kai]

Don't limit your challenges, challenge your limits

Good luck have fun

[Kai]

After two weeks of testing things are looking promising

Greyhound traders right now

[Anbell]

I've been trading full-time on in-running horse racing for several years but never really touched pre-off trading.

During the recent weather-related downtime, I thought I'd challenge myself to create a profitable Greyhound automation using only;
> info shared on this forum
> a good knowledge of Guardian
> a working knowledge of Betfair market forces

After two weeks of testing things are looking promising, and I go live tomorrow...

Godspeed

[sIGNAL]

Sounds good. Good luck. I've been automating a greyhound and IR horse bot the past few weeks. Greyhound bot is very tight, needs relaxing abit I feel., minimal losses but winnings are very low.

IR horse bot is doing quite well, I'm just experimenting with how to minimise losses. That feels like the hardest part to crack as there are so many options so trying to find the balance.

So I'm really interested in how you go, keep us posted

[jamesedwards]

18 hours and 207 races in, and off to a good start.

Only minimum stakes so far. Going to spend plenty of time tweaking before I see if there's any scale possible.

Capture27.png

[Naffman]

Are you trading or betting?

Good luck with it James

[jamesedwards]

Are you trading or betting?

Good luck with it James

Thanks.

Right now I'm limiting volatility (and giving back a little bit of expected value) by greening up just before the off. If it works longer term then I plan eventually to remove the greening as the swings will help contribute commission towards PC.

[Naffman]

Are you trading or betting?

Good luck with it James

Thanks.

Right now I'm limiting volatility (and giving back a little bit of expected value) by greening up just before the off. If it works longer term then I plan eventually to remove the greening as the swings will help contribute commission towards PC.

Nice one, yeah anything to help reduce PC will be worth it, even if you have to lose a little too (but preferably not)

[jamesedwards]

Capture28.png

[Roobetter]

what happened?

[Anbell]

Capture28.png

We all have hundreds of versions of this story in our history. Chin up.

[jamesedwards]

3,500 races in and definitely looks to be returning a positive EV.

The size of the swings are very interesting, with much longer winning and losing runs than I would have expected. Will have a deep dive into these at some point as further gain possible if patterns can be identified and losing runs reduced.

The big disappointment is the apparent complete lack of scalability. Just a small increase in stakes from £1 to £1.50 for 500 races completely tanked any profit. The gearing ratio of ↑stake to ↓ROI% in this game never fails to astound me.

Capture38.png

[jamesedwards]

Some interesting info started getting pulled into another thread so bringing it back into here...


Have spent zero time on this over Christmas, just left it running in the background and it continues to make steady profit at £1 stakes.

Picture4.png

As per other thread, I thought I'd try increasing stakes from £1.00 to £1.50 again this afternoon just to double check the impact of increased stake and total lack of scalability. Sure enough, the swing from profit to loss was immediate, giving back 10% of a month's profit in just 3 hours.

Picture6.PNG

I thought it might be interesting to share a little about where the negative swing is coming from by comparing results of £1 stake to £1.50 stake trades. It seems the key disparity comes from a much greater increase in average loss compared to average win. Presumably this is because my -EV offers are hoovered up by the market while my +EV offers become harder to completely fill.

Picture7.png