How big of an edge?

[burdo77]

I have been working on a data set for uk horse racing based on naps and lays. Under certain criteria when laying I have a 9% edge based on 500 races after I take out commission.

Is that good or should I try to improve my edge?

In addition does anyone know what edge some laying services have, if any?

[spreadbetting]

Dunno how you're calculating your % edge but if that's on turnover 9% is a massive edge. Plenty of my 'edges' are under 1% but give me a decent enough living. 500 races is a small sample but you've got to start somewhere. If it's a theoretical 9% might be worth you sticking small money into it as that usually shows up any flaws very quickly.

As for the laying services, they don't really need an edge whilst idiots are willing to pay for their tips.

[burdo77]

I put my data through some simulations using some software known as the staking machine which compares all known staking methods and takes into account commission.

Laying by liability produced the best profits.

I'm sure if I do a bit more work I can increase the edge.

[Akindwurd]

I wouldn't be worried about whether 9% is enough, I'd be worried that it's probably too good to be true.