How much time do you spend on system or strategy?

[trader-john]

Hello, guys.

Tell me how much time do you spend on System, Strategy or EA before you stop or go live? I personally give chance all automated systems, I install them on my VPS and live for a long time. Sometimes I check the results, when there is something interesting I take a closer look.

I don't have much patience for strategies and if doesn't work for a month I'm done with it. It has to be profitable for 1-2 months before I put it on my live accont and will trade with small lot size for next 1-2 months.

Cheers.

John:clover:

 

[NVP]

also depends on the market..............

do you dump or bring a system to market based on last weeks events ?

or exclude the data ?

personally if a system continues to deliver consistent proits n line with previous weeks on last weeks action - then its probably ready to roll :smart:

N

 

[shadowninja]

The time is irrelevant. It's the number of trades that counts. You test a strategy for 2 months that gives you 3 signals, that's proof of nothing. You test another one for 2 weeks that gives you 200 signals...

 

[trader-john]

The time is irrelevant. It's the number of trades that counts. You test a strategy for 2 months that gives you 3 signals, that's proof of nothing. You test another one for 2 weeks that gives you 200 signals...

It's really good point. But 200 signals only scalping strategy can provide. Or I'm wrong. :whistling

 

[meanreversion]

The original point about discarding systems is an interesting one. I find that rather than try to find an unbelievable system, it's better to have a few good ones on the go and then have some criterion for ending a system. In his monumental thread in the journals section, Travis outlines how he rejects a system which exceeds 50% of its backtested drawdown. Seems fair enough.