I was curious if anyone here uses a passive trend following methodology. In other words, an "always in" system that reverses based on certain conditions. Any comments or ideas on this? How do you handle position sizing or other? r
Interesting question and difficult problem at the same time. I tried something like that 15 years ago with currency futures. It worked well the first two years and I doubled my account. Then, suddenly this method stopped working and I was faced with a huge drawdown. I was very curious why that happened and after extensive research and analysis I found what seems to be the right answer. The performance of a passive trend following methodology depends on how sensitive are your criteria for reversing and how that impacts equity growth. If you set loose criteria you end up trading choppy markets and your equity goes down due to many losing trades. If you set the criteria very tight then you get large drawdowns because you reverse too late and you miss a good part of the correction at the same time. So the right methodology is one of timing the reversal points and as you might have understood already there is no solution to this problem because markets change all the time. In conclusion I believe that there is no "good" passive trend following methodology other than "buy and hold" in equities. Not for markets subject to cycles like commodities and currencies. I think that Michael Harris explains the problem of trend following well in one of his articles in Fidelity Active Trader: "A price trend is probably one of the most elusive concepts in trading because a trend can only be identified after a significant portion of it has already formed." http://personal.fidelity.com/myfidelity/atn/archives/august2003.html Ron
Other names for this same idea are "two phase system", "stop and reverse" trading system, "pure reversal trading system", and so forth. I trade a suite of several systems, chosen among other things, for diversity of trend length and for noncorrelation of performance. I find that mixing two-phase systems (either Long or Short but never Out) with three-phase systems (can be Long or Short or Out) is beneficial, at least in my case. The 2 phase systems are plenty different and, apparently, different in a good way. However, all the books are correct: two phase systems DO encounter whipsaws: numerous, frequent reversals. And it IS annoying, as the books caution. On the other hand the benefits outweigh the costs (monetary and psychological), in my experience. In my particular case, the suite of systems is applied to a portfolio of several dozen commodity futures markets. This requires a large account size, and so it might not be a good fit for everybody.
Ron, Thanks for the input. Do you mind sharing in general terms a little more about "how" you traded it? How many days were you in the market on average moving in a certain direction? Do you still trade currency futures? using a mechanical approach? Thanks again for the input. It was helpful.
Hi Richard, The average number of days when the system followed a trend was anywhere from 25 up to 120 or even more. I remember I closed my first trade after 3 months. Losing trades lasted a week or two. I now trade Forex instead of currency futures and I take advantage of the higher leverage and tight spreads. I do position/swing trading and use a combination of short-term trend indicators, overbought/oversold levels and price patterns. I know many people like trend following but I hate huge drawdowns. I prefer to make less and sleep more Ron
I attended a presentation a while back in which the the trader showed copies of actual statements where he took a $75k account to just over $1mm in one year. He said the results would have been $4mm except that his wife kept bugging him to take some money off the table. He was trading US bond futures from mid-1986 using a reversing system and aggressive position-sizing. I could never have tolerated the equity swings, especially when his positions sizes got really big. I recently investigated using a reversing system on index futures intraday, and the results were disappointing. I may go back and have another look someday, but I don't have any bright ideas for now. [rwk]
I have been having some general thoughts lately on trend following. In some ways, the trend of trend following is in a consolidation mode. Shaking out all the weak hands. I think it is going to breakout and re-trend again for the next few years. I won't argue that markets have changed dramatically, especially on the pariticpation side and the flow of information side. I believe this will ultimately build and new explosive trends will emerge in the Age of Information. It's part of the cycle on a macro type view. So just like bell bottoms, etc. there will again be stories of great fortunes made by trend followers for all the book writers in the next 20 years or so. I know as I get older, everyday I realize that I have no predictive edge in the market except my ability to react to and plan for price movement. In reality, the only thing any of us have is non-diluted price movement. No matter how much we like to think we know what is going to happen. Thanks for the input so far.
I recall testing long / short (always in, never out) systems and found they work much better for some securities than others.
My weakness has always been..... how do you identify when to stay out of the market? How do you identify consolidation periods effectively so you can wait on the sides? This is difficult for me to put into code. In fact, I can't even write it out in English so I am comfortable with it. Maybe I should start a thread that deals with "How not to be in the market". How do you guys do it?
Rm-I only trade the es and Im "always in", but I dont use trend; at least in the sense of trend lines, x-overs, etc.. Rather, I use only support and resistance prices for reversal points. The only time Im out of the market is when I quit. My real grief came, like you and everyone else, from defining and testing the method. As mentioned earlier, with the "always in", if you work the chop, the runs kill you...if you work the runs, the chop kills you. So, the real challenge is to get comfortable trading consolidation and breakouts. My method is sort of discretionary trading with rules, if that makes any sense. I dont use indicators since theres not alot of time to analysis indicators or confirm trades and, the "always in" approach does just fine without any position size progression or regression. I get some false breakouts during the day but this only results in two or three bad trades before the rules put me back in sync. If you want to study something, dont waste your time on some silly indicators that will only tell you what the market just did...or honeing your predictive skills. Pick a market and pick time-frame. Then, waste your time studying/working support and resistance prices without stops. I learned years ago not to mix time frames; I have one screen and I only watch one time frame. The pay off is a high return with mimimal risks, even on one lots. Good luck!