90%?? - page 2

 
Craig:
Update, being doing some out of sample testing tonight, I have a couple of EA's which have about 1.7 profit factor over a year, trying them on the prevous years data now has them making small profits, as opposed to totally bombing. I would take this as evidence that there is less curve fitting going on over the tick data, therefore my hope is that higher performing systems will do even better on out of sample data. I guess I answered my own question.

True, there is a lot of missing tick data versus what you see in 1M. However, how many EA's benefit with micro analysis of each 1M bar? Sure, some of them. But keep in mind that if your EA trades frequenty as in every few minutes, you could be viewed as scalping. Scalping isn't always permitted depending on your broker.

Not trying to argue with you. Just trying to better understand the pros and cons of the strategy tester.

 

The EA's are trying to catch trends at the 15min level, so not really scalping. Remember even at 1 hour you are only getting 60 ticks, I can give up a EA which exploits the tick interpolation on the H1 timeframe and makes billions . Little things count in large amounts.