This problem seems to manifest itself once trends are underway - my best backtest results happened when I caught the big moving trends really early on. Later on, they got more volatile but still capable of making big moves. I would then be stopped out while things were volatile and of course traded smaller when the trend resumed. Since these were long-term systems, I would be riding trends for quite a while with very small size - that does not feel very good.
I found it to be very difficult to adjust trading size effectively using a system. Yes, you can use ATR to normalise volatility across stocks, but I found it to be ineffective as big movers have their size adjusted downwards too much. Nowadays, I will adjust my size based upon how well I am doing in the current month, how well my current trades are holding up, whether I have managed to at least book some profits, and how well stocks are acting in their bases. For instance, for August, the first three criteria already had me on the back foot, but last week several stocks just had complete base failures and then I knew I wanted as little exposure to the markets as possible. That is something that I probably can backtest, but right now I just don't feel the need to do so - my exposure mirrors those of some of the better stock traders so I am happy to just use judgement for that.
Can't really provide any insight on the mean-reversion strategies as I don't trade any. However, if you want to hold long-term then using options to hold through earnings seems to be your best bet. The tax benefit I get from having my gains classified as capital, will more than offset the cost of the options. Thus, in my situation it makes sense, but I am not sure what your situation looks like.
The following user says Thank You to grausch for this post:
I agree that the sharpe ratio seems to be better if I just accept the volatility - that's why I struggle some with it - but at the same time I try to accomodate my risk averse nature and accept lower performance for lower risk. I can trade "small" with the model and the volatility doesn't hurt as bad because it's a small part of everything else going on, but if I were trading a large part of portfolio w/ high volatility I think the swings would bother me more.