Im sure allot of you guys use either discretionary stops slightly below obvious support, or volatility based stops.
Discretionary stops are clearly the best (if your good) but im trying to create a mechanical EOD system and discretionary doesnt fit with what im trying to do.
Also volatility stops tend to give you the lowest position size for the best trades, V shapes and such. I want to back test using constant position sizes and risk, and I have found some success using fixed percentage trailing stops, say for example 4% below your fill price.
Now clearly, a fixed stop (and risk) is only going to work in specific markets at specific times. If the system uses a 4% stop but the daily noise is often 4% or greater then you will get stopped out too much by noise.
So my question to you guys, how often are you stopped out by noise? (not a short pullback of a few periods, just a wide range within one period)
And what would you consider to be an acceptable percentage of trades to be stopped out immediately by noise?
I have created an indicator that calculates a moving percentage of the stops that would have been taken out by noise at certain price levels and stop factors, but I am still trying to work out how to read into the numbers.
Most people seem to use ATR but its for factoring in gaps between periods. As far as im concerned with this system, if the price gaps against me the next day, I am wrong and I want to be out anyway. My own indicator calculates the percentage the low is from the open, and if greater than my stop %, indicates a stop.