Thank you. What is/are the main differences between the two and is there a way to trick the bot to behave more like practice mode vs live?
Bot chooses different horses in practice mode vs live
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
As I said, timing, PIQ and your effect can't be perfectly simulated so although it's damn close it can't be exactly the same as live.
'time to appear in the market' means just that. A bet submitted in-running is subject to a 1s delay before its either matched or appears as unmatched and other people see it. Sub-min bets take 2s. But aside from that it's not unual for there to be 100s of bets placed/matched a second so live and any sort of simulation will always have differences. Practice mode isn't Test mode.
- jamesedwards
- Posts: 2309
- Joined: Wed Nov 21, 2018 6:16 pm
In my experience Practise Mode is incredibly useful to check an automation behaves and fires as planned. It's saved me thousands over time for sure, by highlighting some really big mistakes .
But profitability in Practise Mode may be very different to Real Mode for several reasons. A couple of the biggies are:
1) as soon as you become any part of the market you are dragging the efficiency away from your selection. The higher your stake, the more the drag. Don't underestimate stake. I've had several automations over thousands of markets that return reasonably at £2 but drop off a cliff just by doubling stakes to £4. A tiny difference in % return from stake increase can add up substantially in the long run.
2) Practise Mode will assume an infinite value available to match at the best available price. This means you will always get fully matched at the best possible price when in Real Mode this is, of course, not the case.
I've found best practise is to use Practise Mode only for testing behaviour, then move onto £1 stakes for testing profitability. If results are favourable then increase stakes very slowly.
But profitability in Practise Mode may be very different to Real Mode for several reasons. A couple of the biggies are:
1) as soon as you become any part of the market you are dragging the efficiency away from your selection. The higher your stake, the more the drag. Don't underestimate stake. I've had several automations over thousands of markets that return reasonably at £2 but drop off a cliff just by doubling stakes to £4. A tiny difference in % return from stake increase can add up substantially in the long run.
2) Practise Mode will assume an infinite value available to match at the best available price. This means you will always get fully matched at the best possible price when in Real Mode this is, of course, not the case.
I've found best practise is to use Practise Mode only for testing behaviour, then move onto £1 stakes for testing profitability. If results are favourable then increase stakes very slowly.