[Jasonn]

looks like a great pdf Nick. I look forward to reading through it... thanks!

[Andre]

I don't trade the exact turtle system, but I've tested it so many times, and here is what found:
- Yes it needs a huge account (probably a million or more)
- The amount of risk per trade needs to be pretty low otherwise you risk too much. Maybe 0,25% to 0,5%
- You definately needs much more than 10 markets like the pdf showed. I have 50.
- You need a GOOD, WELL BUILT continuous futures time series (most people don't know how to built it, and most vendors don't do it correctly either)
- Pyramiding is bullsh*t, doesn't improve your return to risk ratio more than increasing your risk %
- No you don't need to buy anything, I've done all in excel - and fortunately a bloomberg i have access to. But it took me more than a year. price series adjustments, little bugs, etc.

And, Richard Dennis is human like any of us, the model didn't blow up, he did, by not obeying the model or risking it too much.

[BlowFish]

Originally Posted by Andre »

I don't trade the exact turtle system, but I've tested it so many times, and here is what found:
- Yes it needs a huge account (probably a million or more)
- The amount of risk per trade needs to be pretty low otherwise you risk too much. Maybe 0,25% to 0,5%
- You definately needs much more than 10 markets like the pdf showed. I have 50.
- You need a GOOD, WELL BUILT continuous futures time series (most people don't know how to built it, and most vendors don't do it correctly either)
- Pyramiding is bullsh*t, doesn't improve your return to risk ratio more than increasing your risk %
- No you don't need to buy anything, I've done all in excel - and fortunately a bloomberg i have access to. But it took me more than a year. price series adjustments, little bugs, etc.

And, Richard Dennis is human like any of us, the model didn't blow up, he did, by not obeying the model or risking it too much.

Pyramiding is the turtle system. The money management is the very essence of the system. The trigger (simple BO) is pretty irrelevant, anyting that will ensure you are positioned when a trend starts.

When I looked at it (which was many years ago admittedly) you absolutely needed the aggressive adding of units to wining positions to cover the many small losers you get before a market really trends. It was this that actually turned the system profitable. Maybe markets have changed since then (though I would have thought for the worse). In any case it taught me what old fashion trend following was all about.

[Andre]

Originally Posted by BlowFish »

Pyramiding is the turtle system. The money management is the very essence of the system. The trigger (simple BO) is pretty irrelevant, anyting that will ensure you are positioned when a trend starts.

When I looked at it (which was many years ago admittedly) you absolutely needed the aggressive adding of units to wining positions to cover the many small losers you get before a market really trends. It was this that actually turned the system profitable. Maybe markets have changed since then (though I would have thought for the worse). In any case it taught me what old fashion trend following was all about.

It looks like Pyramiding is the turtle system, but it is not. Assuming a risk % like 0,25% per trade per market, all Pyramiding does is to increase returns from say 20%/yr to 30%/yr, while magnifying the drawdown in the same proportion. A better alternative is to increase the %risk and NOT use pyramiding, or alternatively increase the number of markets you can trade simultaneously.

I have always tried it with between 38 to 50 markets - everything mixed from indexes to currencies to rates to commodities. So maybe (I don't know about that because i didn't test it) you think pyramiding is a must because when using a small set of markets you definately need to make a killing in that one big trend. But for all I know, now after 2 years of backtesting, 20 slightly different versions of the model and live trading, all I can say is that, for 50 markets, daily time-frames, weekly time-frames, 20day breakouts or 20 week breakouts, or 50 week breakouts for that matter, pyramiding is definately not the key.

The key? the key is that markets trend, and you risk small to make a killing now and then.

[BlowFish]

Doesn't pyramiding ensures that you loosers are minimum size and your big winners are maximum sized? I guess you have read phantom of the pits? PoP does a much better job of describing why than I would. Links to the long and short versions of the ebook can be found here http://www.traderslaboratory.com/for...pits-1599.html

In any case if you remove the scaling the system simply becomes a basic donchian breakout it is no longer resembles the turtles system

Maybe it's a question of terminology. The management uses a constant risk system (adjusted for the volatility of the market). Contracts are added as the market moves N units in your direction. (N is dollar adjusted volatility). So as the position moves in your favour the risk is kept constant even though the position is growing. There is a cap on adds is it 4 units or maybe 4 adds? I'd call that 'pyramiding' though maybe others would not

Still trying to get my head round how it can work with 70-75% losers if they are all the same size as the winners. Another advantage of the turtle system is that it fully monopolised capital whilst controlling risk. Fairly simple stuff but quite elegant too.


Might I ask are you an old fashioned trend follower? If you have access to spread betting wherever you are located you could conceivably follow a lot of markets with a much smaller account.

[DugDug]

I have to agree with everything Andre says - we did the same thing - used Excel to backtest the turtle rules and our own variations of them.
We also had similar conclusions. What we found was that the most palatable results were when we only pyramided 2 times, and reducing the risk per trade from 1% to 0.25% of equity per unit.

Now on saying that - there were years/tests where you could have made 600% on a very scalable amount of money.(we tested data from 1994) Problem is to do this you needed to really increase the risk per position, and this leads to massive intra month drawdowns. (up to 80-90% (many people look at end of month - we looked at daily - can be hard to stomach being down 20% mid month to end up being up 3% at end of month)
Also many people find it hard to stomach the correlation between many products in the short term - especially when it turns against you and losses hit the portfolio all together.

Regardless the concepts are good to understand and the ideas interesting. You do need a broad portfolio of instruments, lots of money and a stomach for drawdowns - ie; a very intensive belief/faith in the system.
It works over the long term - the only way it stops is if you think trends will stop. Trend traders killed it in 2008 in terms of making money....not so good this year. Its the problem people have with such a system - they want to make money all the time and cannot understand that the key is in the cumulative returns over the long run when looking back at a return per annum average. (look at the equity markets - they outperform bonds over the long term - but may not over shorter periods - everything becomes relative to what you are trying to achieve and measure I guess.)

But for all the companies/funds that I have seen do it (trend trading) profitably in terms of scale usually incorporate some other measures - either short term trading, or variations on portfolio construction, discretion on when to roll contracts - pricing of these eg; OJ trades differently to corn, oil, bonds, and rubber. The turtle system is just a simple and aggressive version of trend trading. (look at the long terms returns of trend traders - their returns vary greatly but it works over the long term)
The other important thing that most people do not test for the turtle system- as its hard to accurately do, or as we used - requires a separate filter - is to incorporate the rule that goes something like - "after a winning trade do not take the first trade in the opposite direction (until that opposite direction trade has resulted in a loss)"

[Andre]

Originally Posted by BlowFish »

Doesn't pyramiding ensures that you loosers are minimum size and your big winners are maximum sized? I guess you have read phantom of the pits? PoP does a much better job of describing why than I would. Links to the long and short versions of the ebook can be found here http://www.traderslaboratory.com/for...pits-1599.html

Have never read it, thanks for the link. Pyramiding doesn't ensure anything, otherwise you could just add a pyramiding scheme to a trading system and make money out of it. So far, I know only one person who did it - Madoff
Think about it, if the edge on each additional entry level of the pyramid were similar, adding a pyramid to the system won't improve the reward/risk ratio, you're just leveraging up. If the edge was better for each additional entry level, then a pyramid would improve the system. But in this case, you could improve it even more by only taking the upper signals and not using a pyramid.

Originally Posted by BlowFish »

In any case if you remove the scaling the system simply becomes a basic donchian breakout it is no longer resembles the turtles system

Exactly, the turtle system is nothing more than a donchian channel, at least to me. It's no secret formula, the only innovation from the turtles came from showing you can teach a group of complete newbies (no offense inteded whatsoever) to make money in trading.

Originally Posted by BlowFish »

Maybe it's a question of terminology. The management uses a constant risk system (adjusted for the volatility of the market). Contracts are added as the market moves N units in your direction. (N is dollar adjusted volatility). So as the position moves in your favour the risk is kept constant even though the position is growing. There is a cap on adds is it 4 units or maybe 4 adds? I'd call that 'pyramiding' though maybe others would not

I've tried a pyramid the way you describe it, also adding a system on top of another system, adjusting the sizes of each, etc. Nothing really improves it. You know, it's a bit of the same improvement you get by trading a really short-term TF system mixed with a very long-term TF system. It gets a tiny little better, you may also feel more comfortable with following it, thinking "ok now if one of the time-frames stop working i'll have the other one to make money...", but they're just too positively correlated to make a real difference in the numbers.

Originally Posted by BlowFish »

Still trying to get my head round how it can work with 70-75% losers if they are all the same size as the winners. Another advantage of the turtle system is that it fully monopolised capital whilst controlling risk. Fairly simple stuff but quite elegant too

It does work, it has nothing to do with the % of losers, and i don't know what kind of nrs you're seeing but the size of the winners is much more than the losers. For a very long term system the size decreases while the % increases.

Originally Posted by BlowFish »

Might I ask are you an old fashioned trend follower? If you have access to spread betting wherever you are located you could conceivably follow a lot of markets with a much smaller account.

There are no such thing as a "old fashioned trend follower" I only trade futures, if that's what you ask. I don't see (so far, maybe the future will be different) the upside in opening an account in a spread betting. Risk with your cash deposit (something i can't live with), much higher commissions and bid-offer spreads. Not that it reeeeally kills a trend following if you trade 4 or 5 times a yr in each market, but even in futures you can pay 1%/yr in commissions, and that's a lot to me. I prefer to pick smaller sized futures instead.

[robertm]

Originally Posted by Andre »

Pyramiding doesn't ensure anything, otherwise you could just add a pyramiding scheme to a trading system and make money out of it.

I'm sorry what?

I can see why you could think this if you only trade futures. Oil isn't going from $70 up to $1000 over the next 3 years. Somewhere out there though, is a stock that WILL go from $7 to $100. Our risk on each instrument is the same, our potential for reward is MUCH greater if we can pyramid aggressively into an instrument that can have large sustained moves. This isn't to say we should only trade stocks, each market/instrument has it's unique benefits.

How your pyramid (if it all) has more to do with the time frame & instrument than anything else. Small targets = no room (but we can opt for profit risk strategies), multi year trends = load it up as much as you can while controlling heat & risk.

CFD/spread bet is the ultimate pyramiding tool once you understand it. Without pyramiding it's just an expensive margin loan facility.....