Just out of curiosity, has anyone created an automation set thinking "yeah this is definitely going to work the way I want it to" but then the exact opposite has happened?!
I'm trialling out a new set of rules in automation and it is being consistent but not in the way I was expecting - it is always returning a loss
If I can ensure it is a consistent loser, I can flip it so it becomes a winner but I was just wondering if anyone has come across this before?
Yup! I've definitely had a share of times when an automation has made more profits than expected due to a bug (and vice versa). I've also had situations where I've realised that I've completely misunderstood why my automation was making money (so sort of misunderstanding where the edge was coming from).
Although on first thought it might seem intuitive that you can just flip a consistently losing strategy around to make a winning one, it doesn't work because of the bid / offer (back / lay) spread. Or similarly if you were to be matched during the losing system, you wouldn't necessarily be able to be matched on the other side (for the flipped winning system).
One example that'll be a very consistent loser: Repeatedly back at the back price, then lay to close at the lay price immediately after!
That's not to say that flipping won't ever work. If anything, understanding why something is consistently losing might give you some new ideas, if not quite as simple as just flipping the losing system. So in the simple example above, it's clear that you're losing money consistently to the spread, so rather than take a price, you offer a back at the lay price, then close with a lay at the back price. But of course, you can't guarantee that you'll get matched on either end which is where the problem comes. So homework for the newbies - Think about how you can improve your chances of getting matched or predict the direction of the movement.
xitian wrote:
That's not to say that flipping won't ever work. If anything, understanding why something is consistently losing might give you some new ideas, if not quite as simple as just flipping the losing system. So in the simple example above, it's clear that you're losing money consistently to the spread, so rather than take a price, you offer a back at the lay price, then close with a lay at the back price. But of course, you can't guarantee that you'll get matched on either end which is where the problem comes. So homework for the newbies - Think about how you can improve your chances of getting matched or predict the direction of the movement.
Thank you very much for the advice, most appreciated and duly noted. Homework will hopefully be handed in just in time for a profitable Cheltenham!
