Those 'real world variables' can really screw things up!
For me, trading environment could be a problem, sometimes I have an 11 hour time difference,
which makes concentration tough.
Also, I've never traded 20 contracts before, not sure if those orders would get filled without considerable slippage.

I hope I'm not under any delusions! I guess a successful backtest just proves that a system 'hasn't failed yet',
rather than proving the system 'actually works'.

I think I'll give it a go, at least up to 5 contracts to see how it goes.