Automating Competitive Markets

A place to discuss anything.
Post Reply
User avatar
alexmr2
Posts: 768
Joined: Wed Sep 26, 2018 12:32 am

Do bots usually perform better in markets that are less likely to have successful manual traders participating?

E.g. is it harder to make a profitable bot in UK preoff horse racing than other markets that serious manual traders aren't going to bother with, or does the extra liquidity even out the bots' chance in either market?

Considering PW still trades preoff UK racing manually I assume humans still have the edge there. It would be interesting to estimate how much % of a month's manual results would be achievable by a bot trying to replicate his manual trading.

I understand a lot comes down to perceptual judgement on what is going on at the track and in the market at the time + humans can optimise the execution better. I suppose it would also depend on the effort spent maintaining a bot which could translate to time better spent manual trading anyway.

Basically can you successfully automate competitive markets or is it better to focus on EV spots that manual traders are unlikely to cover?
Anbell
Posts: 2395
Joined: Fri Apr 05, 2019 2:31 am

alexmr2 wrote:
Tue May 31, 2022 7:51 am
Basically can you successfully automate competitive markets?
yes
User avatar
Euler
Posts: 26436
Joined: Wed Nov 10, 2010 1:39 pm

I trade manually, just because I can. But I run 5/6 pieces of automation on the same markets as well at the same time.

Some of my automation only works on certain markets and others on the completely opposite. But they just evolved differently over many years.

All my automation started with a manual process that I refined over time. Automating the entry, the exit and then the trade management until it could run on it's own.

My suggestion is to write alerts in the automation to certain things and when you are confident that they are working, use them to trigger actual trades.
Anbell
Posts: 2395
Joined: Fri Apr 05, 2019 2:31 am

Euler wrote:
Tue May 31, 2022 9:08 am
My suggestion is to write alerts in the automation to certain things and when you are confident that they are working, use them to trigger actual trades.
This is good advice, of course, but you can also do it the other way: Use auto to place bets and then work out which are profitable
User avatar
alexmr2
Posts: 768
Joined: Wed Sep 26, 2018 12:32 am

Euler wrote:
Tue May 31, 2022 9:08 am
I trade manually, just because I can. But I run 5/6 pieces of automation on the same markets as well at the same time.

Some of my automation only works on certain markets and others on the completely opposite. But they just evolved differently over many years.

All my automation started with a manual process that I refined over time. Automating the entry, the exit and then the trade management until it could run on it's own.

My suggestion is to write alerts in the automation to certain things and when you are confident that they are working, use them to trigger actual trades.
Thanks for the response and suggestions
User avatar
alexmr2
Posts: 768
Joined: Wed Sep 26, 2018 12:32 am

Anbell wrote:
Tue May 31, 2022 8:26 am
alexmr2 wrote:
Tue May 31, 2022 7:51 am
Basically can you successfully automate competitive markets?
yes
99% of my time trading has been in preoff racing so good to know it's worth continuing with it in auto or semi-auto
Anbell
Posts: 2395
Joined: Fri Apr 05, 2019 2:31 am

alexmr2 wrote:
Tue May 31, 2022 10:34 am
Anbell wrote:
Tue May 31, 2022 8:26 am
alexmr2 wrote:
Tue May 31, 2022 7:51 am
Basically can you successfully automate competitive markets?
yes
99% of my time trading has been in preoff racing so good to know it's worth continuing with it in auto or semi-auto
I know that youve spent a lot of time trying to get profitable. The issue isnt preoff or racing or manual or auto. The issue is whether or not you have an edge.
User avatar
ShaunWhite
Posts: 10498
Joined: Sat Sep 03, 2016 3:42 am

Automation is harder than manual trading Imo because you have to do without that unquantifiable but oh so significant subjectivity. But as I've said till its boring, just use the right tool for the right job. Why automate what requires subjectivity when there's a whole world of things you can do that manual traders can't, and therefore don't necessitate subjectivity.

And don't be fooled into thinking it will remove psychological issues, everyday is a should I/shouldn't I. Then there's the disappointment of false dawns on projects that took months, long term good things fading, the need to constantly have new ideas to develop to replace dying ones. Daily analysis and headscratching. You'll soon want to be back doing 4hrs a day and getting the chance to be a hero in the next ten minutes.
User avatar
Derek27
Posts: 25159
Joined: Wed Aug 30, 2017 11:44 am

Automation is like a single malt whiskey. You need to blend several whiskeys together for the perfect taste. One whiskey won't produce consistent results. :)
Post Reply

Return to “General discussion”