Quote from Buy1Sell2:
If a trader feels that the trade they are in warrants having part of it taken off, then it really warrants taking off the whole trade. Winners are where the money is. They must be allowed to ride. Thanks for the question.
B1S2,
The markets are not that accomodating where one can simply state that scaling out decreases performance....And these are the numbers talking..
I would agree that one should let winners ride,but you must define ride..For an EOD trader with a longer perspective,does ride mean 1 ATR?? /10ATR..20%?? 70%??
Run thousands of backtests and of course there will always be some "hypothetical" scale out point that maximises return. Is there a flat plateau in the optimisation graph which would indicate with some level of confidence that scaling out may be the correct path. And if that level is 50% or greater,would you consider that "letting it ride"??
Also,when you backtest,do you run long and short positions at the same time to validate scaling out/NOT scaling out?Obviously if one doesnt,depending upon what type of market one is testing on will skew results dramatically...
Last but not least,are we speaking strictly in terms of Net Profit when deciding upon scaling out or not?Are we ignoring volatility of returns or the smoothness of ones equity curve?
B1,I have seen no statistical validity to scale or not..The markets are too efficient to have one style be more profitable than the other...In the end,all it comes down to the other points you brought up as well as the method that one can emotionally live with