You're generally right about winning percentages in a buy-high/sell-low system, which is obviously the opposite both of good trading range styles as well as conventional wisdom about stock trading. Still, after all is said and done, buying new highs or lows works fine when an overall trend is in place and in your favor, and when you have a good exit strategy, as the big wins more than compensate for what is typically a greater number of small to moderate losses. Traded mechanically (or systematically) with appropriate filters and money management (easier to describe than to create), buying high and selling low also tends to work out mathematically, if not always emotionally. In trading a pure version of the Turtle System, for instance, you have to be willing at times (i.e., during extended non-trending periods) to undergo literally tens of consecutive losses and very large drawdowns from the most recent equity peak. You'd spend much of any given trading year, even the good ones, well below your most recent peak. You're more likely than not to start out on a losing streak. Still, if you have faith, you have a very high expectation of high long-term profitability.
I'd guess that the beauty of trading a Turtle-like system intraday is watching the equity curve rise much sooner, increasing confidence in the concept and thus willingess to accept the necessary drawdowns. Unfortunately, as PDT and I have both pointed out in different ways, designing, testing, and effectively trading an intraday Turtle-like system isn't easy, even if you're not aiming to trade dozens of vehicles at once like PDT. Among other things, the literature on true systems (i.e., as opposed to mere methods) designed for equities traded intraday is so scant it's almost non-existent. It's therefore not that surprising that PDT's firm is so secretive about what they're doing - for the moment, anyway, they appear to have a significant edge, though I have to wonder if the market, with typical perverseness, isn't about to ruin the fun. Though I'd expect the system has been back-tested on all manner of historical data, we may be approaching critically unprecedented "curves." |
