"Instead of trying to identify profitable trading algos on in-sample data that validate out-of-sample and remain profitable forward, one could instead try to identify unprofitable algos in some data sample that turn profitable in a forward sample. This often works because markets have become more mean-reverting in recent years." Article Has anyone tried this approach?
I've toyed with it. There are weak mean reverting effects in some models, the emphasis being weak. I'm very uncomfortable with it, due to the explosion in the likelihood of overfitting. Make it simple suppose there are two 'states' in the market, where different models are profitable. You could fit an average model with N parameters. Or you could fit a model to each state, so you have 2N parameters. You then need a model to determine which state you're likely to be in over the period you're going to hold your positions for. That will probably need at least one parameter. So we've gone from N to 2N+1 parameters. Of course if you're fitting method is robust enough, properly out of sample, blah, blah, blah.... . The other alternative is to make the two models much simpler with say N/3 parameters each. Thus you end up with the same or slightly fewer parameters; and you've effectively gone from capturing behaviour in one way, to capturing it in a more non linear way. Once again I've never seen this work that well, as you often end up with something very similar to the original one state model. "This often works because markets have become more mean-reverting in recent years." - obvious question, why not fit the mean reversion directly?
Well I don't know what the original authors meant by "mean reverting", but a very simple univariate mean reversion model would be something like -(Pt - E(P)) where E(P) is an EWMA of the price series P0.... Pt. Note the minus sign; we buy when we're below the trend and sell above.
I find those to be the least reliable. It would be best to compare your in and out sample data. Have you been finding strategies that benefit from a regime change? Can you expect that regime to continue for the forseeable future? Perhaps, you can use this strategy intermittently if you can forecast an impending favorable regime (holy grail).
I think the article is referring to profitable and unprofitable phases of a trading system's equity curve. This works for a system that has a regular tendency to trade profitably, then go negative, then go positive again. So you are timing the equity curve. You could apply an indicator (e.g., moving average, oscillator) to the equity curve just like any other time series. This isn't "mean reversion" as the term is normally used. But the title of this post is well-named, as it is indeed an interesting idea.