I've heard ETFs mean revert more than stocks and certain ETFs mean revert more than other ETFs. Is this true? Does prior performance indicate future returns? I have a mean reversion system I'm backtesting. I ran it with 2400+ symbols (Stocks & ETFs) for the '94-'99 period. It generated 7,732 trades and showed good results (e.g. Sharpe ratio > 1). Then I sorted the symbol list by profit and skimmed off all the profitable symbols. I know if I run just the profitable symbols on the same time period I'll get fantastic results, but that certainly seems like cheating or data fitting. So I took just those profitable symbols (648 symbols), ran the backtest for the period '00-'09 and got worse results, but not terrible (9960 trades). So, does it make any sense to pick symbols that were profitable in the past?

No, it doesn't make sense. What does make sense it to create a more specific filter (per volume or avg. price or any other metric you see fit) that doesn't trade symbols the next day which display unfavorable characterisitics for some set period of time prior. Then forward test of course. From the sound of it, this system should work across all products. By simply removing poor-performing products you're introducing severe bias into your model. The best thing you can do is focus your development efforts entirely on the losing symbols to see what behaviors caused your model to fail.

ETF's represent weighted indicies of different sectors/groups of stocks. 1) So a more relevant backtest might be to take a look at each individual stock in relation to its indice. 2) Analyze the relative strength of the stock vs indicie to see whether it is weaker or stronger than the group of stocks which are its peers 3) Then devise two strategies: (i) is a trending strategy for stocks which are stronger than its indicie, and the indicie is in an uptrend. (ii) a mean reverting (counter trending) strategy for stocks which are in weaker than its indicie, and the indicie is in an uptrend. 4) Reverse the logic for indices which are in a downtrend. Good Luck