regarding the stops issue, let's consider one simple, hypothetical system where one has a 1% trailing stop (the stop could be arbitrarily increased, but let's use 1% for simplicity). if 1%-move up/down probability is 50/50, then
a) 50% of the time: stopped out immediately at 1% loss
b) 25% of the time: stopped out at breakeven
c) 12.5% of the time: stopped out at 1% gain
d) 6.25% of the time: stopped out at 2% gain, etc.
if you started with $100, you are down to $50 immediately due to a). also, b) does nothing for you, so you must rely on c) onward (i.e., only 25% of trades) to make up the lost $50 and gain a profit, and also cover trading costs. as you progress down to e), f), etc. the profits of each trade increase, but their likelihood of occurrence decreases. it would take until l) (an event occuring only 0.012% of the time) before the system becomes profitable assuming no costs. add in the trading costs and spreads, and this seems like a losing system.
(obviously, a trader might hope for better than 50/50 directional odds in this system, by relying on a separate direction-predicting system.)
there is some pretty interesting discussion of "outliers" in the book "Fooled by Randomness"--you would probably like it if you haven't read it already. |
