Quick few questions. Assuming I have a psychological preference for accuracy, would somebody help me out to establish a baseline for decision making here. I have a "early exit" strategy where 75% of the times that I exit early, it hits my target anyways. I have tried to think of the trade individually, but am coming up with several different solutions. I am trying to establish a baseline for r/r so that I can take the "early exit" only if it makes sense in terms of the tradeoffs. It seems like the expectancy equation should be used here but im not positive. Assuming 1% for stop loss and profit targets in terms of % as well, what i have thought of is... (75%)(TARGET1%-EARLY EXIT%-StopLoss%) - (25%)(Early Exit%-StopLoss%) >=0. Solving for Target 1, it should be 1.33(Stop Loss%+Early Exit%) which can be simplified to 1.33(1%+Early Exit%). This is bare minimum positive expectancy. But if I have a preference for accuracy I should probably demand 1.5(1%+Early Exit%). My question is, am i analyzing my data correctly? I have other similar stats and am changing a few things around to get some better expectancy so want to be sure im incorporating changes correctly into my system. Thank you.
I'm lost. Your r/r should be id say at least a 1:3. After you take some profits, move stops to b/e slightly above the cost of commission. You dont want to give back profits.
I dont really think of this in terms of "giving back profits". I'm letting the statistics speak here and trying to make a cutoff point where i should and shouldn't take the early exit strategy. I'm not trading the early exit strategy out of fear, the statistic is much more prevelant for another setup where the chances of it hitting the target are actually reduced to 60%. Another option is to use both strategies and scaleout at both the early exit and the original. But, sometimes, I do not have enough cars to do this... An application of this is... I have a setup where sometimes my target will be 1:5 risk to reward and there is only a 60% chance that i hit the target if I dont use my trail stop. (not to say that my accuracy is 60%, but that once the trail stop is activated, there is only a 60% chance of actually hitting the target before being stopped)
The way I laid out the equation, I am thinking in terms of "Marginal profit of holding on" and comparing that to "marginal loss of holding on". The way i see it, we can only get (target 1-early exit) EXTRA in profit, but we can give up (Early exit-stop) in terms of profits/account equity. Maybe that clarifies. I am not sure if this is the right way to think about the decision, but as with all things in my trading, I analyze things independently to KISS.
Maybe if I put average down in the title lol. By the way, just noticed theres a thread, take profits early or let it run to target? Well, I think the only answer to this is, applying both criteria to old trades, what fits your personality better in terms of accuracy and what is higher expectancy? I think there is a way to address that question if you are keeping up to date with the statistics on your trading.
Maybe this paper can help you: http://tinyurl.com/39tw342 It shows the relation of R/R to Win rate. This paper shows how you derive the expectancy: http://tinyurl.com/2c7kng2
If Profitability = PF / (PF + Ratio of Wins/Losses) Substituting Marginal Win and Marginal Loss instead for the definition of Ratio of Wins/Losses, I get the equation ... Profitability (%win) = PF / (PF + MW/ML) Solving for Marginal Win, MW, we get... MW = (PF * ML)/probability of trade working Using a PF of 1.5 representing my preference for higher accuracy and 75% representing the % chance of hitting the initial target, Marginal Win = (1.5*ML)/.75 MW = 2ML So with a early exit target of 1% (and a risk of 1%) we get a total Marginal Loss of 2%. To take the trade, I should generate 4% extra by holding on. Is my math off here? I'll be the first one to admit this is my first try at crunching my own numbers, just looking for input.
Marginal win is defined as the extra % gain by holding on after the early exit target to the original exit target. Original - Early = MW Marginal Loss is defined as the % loss if we go back down to my stop, which occurs 25% of the time. ML = Early Exit + Initial risk (my initial risk is always 1%) Looking at both methods I presented, each seems viable over a coin flip method. In trading sometimes good enough is good enough, just have to pick one method that is more or less aggressive based on my perference maybe...