Agree Tripack. Especially this part: But the turtle method will outperform buy and hold in other commodity markets that don't have a long term upward bias. And yes.... read the PDF. peace axeman
Well, you're half right. I write long-term and ultra-long-term trend-following systems that are similar to turtle systems. Currently, I have developed 12 independent systems. Proper money management is, indeed, the key to success. Entry/exit technique is largely irrelevant. The goal is to write a simple, robust, "bulletproof" entry/exit technique that will provide a CONSISTENT framework within which your money management system can operate. However, success (at least, my success) has nothing to do with pyramiding or discretion. My systems have been long the indices, copper, metals, currencies and short the dollar for quite some time. Any properly constructed trend-following system would have to be. I do not ever countermand my systems, and I do not ever pyramid positions. These systems can be easily backtested. Average position size is ~3 contracts per market per million, and is independent of past or present system performance. jj
I did read and understand the PDF. Thats why I asked if anyone had tried this on the DOW just for fun It was just interesting to see what kind of system stats a known trend following system like this would produce on the DOW, knowing full well this was not how the system was really used. I just find it interesting that buy and hold beat it by such a huge margin when run ONLY on the dow. peace axeman
The Dow has an upward bias, since the Turtle system uses up some of that upward bias to generate a signal, it's unlevered performance is not going to be superior to the Dow. In addition, since the Turtle system functions like an option straddle, it pays the straddle premium when the market chops rather than trends. So, those two forces, I think, conspire to hold back performance of the Turtle system. Ideally, the Turtle system should be able to compensate by making money on shorts (which buy & hold is excluded from), but since short opportunities occur twice as infrequently as long opportunities and occur at greater speed, the Turtle system will probably not be particularly effective at generating profits on the short side. Most commodity markets are really downward declining due to cost reductions from technology, and spike upwards in natural disasters or squeezes, so the Turtle system, which is essentially a straddle, should be able to harvest volatility profits. The genius of the Turtle system really, as other posters have mentioned, is in the iron discipline that keeps you from betting too big in the beginning (since this "ante" is going to be frequently chopped up).
Did you trade the 20 day breakout, or the 55 day? And if the 20 day, did you stick to the "pass-on-trade-if-last-trade-would've-been-a-winner" rule?
axeman, I think the system is designed to be traded on multiple markets b/c that keeps it closer to fully invested. Also, there are diversification benefits that improve avg return.
Futurestruth lists performance of the turtle system for a couple of markets. I have a one year old magazine somewhere and can give you some numbers if you want.
dont have it for the dow, but here are some other markets futurestruth follows. the annualized results for the turtle system system from 1/90 - 2 /2002 were Jap Yen : 67.6 % Bonds: 26.4% Coffee: 12.5 % Crude: 12.2 % B Pound: 12.1 % Copper: -12.5 % all calculations on 3 x margin. this should be without specific money management rules like pyramiding. btw, other multi market trend following systems like aberration performed much better.