Hi all. I currently trade quite a few tennis markets a day where I lay the server when they are a break up and green up on a break back or at the end of a set. I have various rules that differ in their staking amounts based on my perceived liquidity of the respective markets. This is so that when I green up I can do so without having to take too much money at low odds. So for a 1st round challenger event I might risk x in liability but an ATP final I might risk 10x. The issue I have is that a lot of the time my estimates are off and I either don't get matched quickly or I miss out on some liquidity I wasn't anticipating. To play it safe I've taken to going with the lowest likely liquidity but this means I'm missing out when there's more available.
So... I'd like to try and build a dynamic staking system that looks at how much liquidity there is on the back side over a period of time and if a player goes up by a break then I lay a stake in proportion to the average I've been measuring. I've built an automation rule that measures the average liquidity across 3 time points with each one 5 seconds apart. Rule works fine but as expected liquidity bounces around quite a bit and I wouldn't want to trust it in its current form. Before I get stuck into trialling a bunch of scenarios, e.g. more points, higher or lower time intervals, I thought I'd ask around and see if anyone else did anything like this and had some tips OR someone does something different that might be easier/more reliable?
Dynamic staking automation advice needed
I guess it would just depend on your strategy and the staking level that suits you.
Apologies for bugging you Dallas... do you have any idea on ratios? Thinking about this I got something like:
Traded volume over last 30 seconds = x then can lay a stake of 33% of x
Theory here is that if the same trading volume holds firm then I would have to wait circa 10 seconds to get matched. I realise there are an incredible amount of variables that will throw this assumption off course but I figured it would be a place to start.
I thought about extending the time window for traded volume measurement but things change so much I decided against it.
How do you manage variable staking in your strategies?
Traded volume over last 30 seconds = x then can lay a stake of 33% of x
Theory here is that if the same trading volume holds firm then I would have to wait circa 10 seconds to get matched. I realise there are an incredible amount of variables that will throw this assumption off course but I figured it would be a place to start.
I thought about extending the time window for traded volume measurement but things change so much I decided against it.
How do you manage variable staking in your strategies?