Original turtle rules

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Original turtle rules

Post by LeviF » Tue Jan 13, 2009 12:56 am

I was revisiting the "Original Turtle Rules" tonight and I noticed CF's statement about the drawdown reduction technique:

"There are other, perhaps better strategies for reducing or increasing equity as the account goes up or down..."

Any ideas what these other, and/or better methods could be?

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Re: Original turtle rules

Post by sluggo » Tue Jan 13, 2009 7:50 am

"Other methods" are easy to imagine. Anything that isn't the O.T.R. "drawdown reduction technique" is, by definition, an "other method." Anything that does not reduce betsize-in-percent when drawdowns occur. Thus to name an example, keeping betsize-in-percent constant during drawdowns (a/k/a "Fixed Fractional" betting) is an other method. Explicitly causing betsize-in-percent to increase during a drawdown (an example of Martingale position sizing) is an other method.

One interesting idea is found in this magazine article: http://tinyurl.com/7svch9 written by a member of this Forum.

Another interesting idea may be found on this Trader's Roundtable Forum website by searching for, and reading, all articles here containing the word "piecewise". (There are only three)

A third idea, easy to talk about in general terms, hard to pin down in specifics, is a Betting Ratchet. As equity grows, your position sizes get larger, as they would in a typical fixed fractional methodology. However, during drawdowns, you don't let your position sizes get smaller. Position size is allowed to ratchet up (get bigger) when equity rises, but it is not allowed to shrink when equity falls. Traders who embrace the Betting Ratchet want to trade their way out of a drawdown as quickly as possible, so they don't reduce betsize when drawdowns occur.

Which is "better"? Better is in the eye of the beholder. That's why there are so many different ways to measure What Is Good. (Sharpe Ratio, Sortino Ratio, Lake Ratio, Return Retracement Ratio, R Squared, R Cubed, MAR Ratio, Kestner K-Ratio, Ulcer Index, ad nauseum). Different people have different opinions of what is Good, and of which is Better. In MY opinion, the only way to decide which is better is to compare them yourself and see which one YOU prefer. You can call it "better" if you like.

J Randall
Full Member
Full Member
Posts: 15
Joined: Thu Jan 29, 2009 4:50 am

Post by J Randall » Sat Feb 21, 2009 10:28 am

Tradingblox does not have the ability to properly test the exact drawdown reduction technique which was used by the Turtles. Does it?

I mean that as far as my understanding goes, and very generally speaking, whatever the total equity was in a given Turtle's notional account on December 31st became "closed equity" on January 1st.

Then for the duration of that calendar year the total equity which was used to calculate the number of contracts in each market using the 20-day ATR was static. Therefore a position taken in January would generally tend to fluctuate the same in dollar terms up or down as a subsequent position taken in December of that same year.

However, the exception to the above was in the event that someone experienced a 10% drawdown in "closed equity", defined as being 10% below the starting equity on January 1st of that given year. It was only then that the individual would begin trading as if he or she had an equity level that was 20% less than before.

So it was in a way a betting ratchet, which was ratcheted up every January. BUT it could be ratcheted down quite quickly if the January 1st equity level was breached to the downside by 10% or more.

I have not tested this technique, and it is not immediately clear to me how I would go about doing that with the Blox platform. But, well, it seems like it is something that should probably be checked out. My guess would be that C.F. probably did this during the development and debugging of the Blox Turtle system and likely judged it to be impractical or without significant merit. But of course I don't know.

My gut feeling tells me that somehow there is something significant about this betting ratchet concept. Maybe for certain scenarios where one is trading at extremely aggressive risk levels, as Dennis and the Turtles did, then maybe there is something in this. Again, unfortunately, I don't know because I have not yet done the testing.

My main point in any case was just to mention that I can't immediately see a way in Blox to test the actual Turtle drawdown reduction method, or annually-fixed-notional-equity money management method for that matter. But please anybody feel free to meet out a quick reprimand and set me straight if I have overlooked something.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Sat Feb 21, 2009 10:49 am

The documentation for Trading Blox software is online and freely downloadable, have a look at http://www.tradingblox.com/tradingblox/ ... tation.htm to see whether the program will or won't do what you have in mind.

The Drawdown Reduction Threshold is a built-in feature of the testing software itself; it can be used with any system (or any Suite of simultaneously traded systems). It's not just limited to the Turtle system.

I fired up my copy of the software and took a screen image snapshot, shown below. In it, I've told Blox to simulate my system (Suite) 66 times, and report the results: Eleven different values of DRT%, times six different values of DRA%, yields 66 different simulation combinations. In this example I've specified "Closed Equity" as the Trading Equity Base, but there are other options available if you click on the black triangle and pull down the selection menu. Since Blox is lightning fast, it takes very little time to run these 66 simulations.
Attachments
blox_settings.png
blox_settings.png (10.23 KiB) Viewed 4805 times

J Randall
Full Member
Full Member
Posts: 15
Joined: Thu Jan 29, 2009 4:50 am

Post by J Randall » Sat Feb 21, 2009 12:17 pm

Thanks, Sluggo, for your reply.

Yes, I am delighted with Blox, and your point is very well taken regarding the speed and functionality of the program. In a way, you could say that these realities render my observations above rather moot.

And I agree. It's always a danger to become fixated on some particular parameter or entry technique or money management algorithm that you read about in a book or whatever, and then pretty soon you are searching for the Holy Grail, entirely missing the forest for the trees.

My opinion regarding there being something of value with Dennis' betting approach is really, I suppose, more philosophical. As I believe has been discussed in other posts on this forum one could - theoretically - blind oneself to all equity fluctuations other that those that penetrated below some predetermined level that one deemed to be of significance. In other words, drawdowns above this level would be disregarded. In the Turtle case, maybe this level was the one that existed on January 1st. But it is in any case just a way of looking at the world and kind of playing psychological games with oneself. My gut feeling says there might be something to it for some personality types when trading at elevated risk levels, but I doubt really that there is a whole lot to it in objective terms.

Therefore once again I would imagine that my original observations above regarding Blox - while maybe true - are probably a moot point in terms of system design.

alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Post by alp » Sat Feb 21, 2009 4:44 pm

J Randall wrote:It's always a danger to become fixated on some particular parameter or entry technique or money management algorithm that you read about in a book or whatever, and then pretty soon you are searching for the Holy Grail, entirely missing the forest for the trees.
...
My gut feeling says there might be something to it for some personality types when trading at elevated risk levels
One interesting method, as suggested by Ed Seykota, is using your gut feelings to measure risk.

Roger Rines
Roundtable Knight
Roundtable Knight
Posts: 1946
Joined: Wed Oct 06, 2004 10:52 am
Location: San Jose, CA

Post by Roger Rines » Sun Feb 22, 2009 10:59 am

When you said this,...
J Randall wrote:Therefore once again I would imagine that my original observations above regarding Blox - while maybe true - are probably a moot point in terms of system design.
Were you refering to this?
J Randall wrote:Tradingblox does not have the ability to properly test the exact drawdown reduction technique which was used by the Turtles. Does it?

I mean that as far as my understanding goes, and very generally speaking, whatever the total equity was in a given Turtle's notional account on December 31st became "closed equity" on January 1st.
If so, I can see a way for Trading Blox Builder to make account decisions by following the stated year end rule. One of the nice things about TBB, is how creative the user can be in generating new data streams, and then using those new streams to make decisions.

J Randall
Full Member
Full Member
Posts: 15
Joined: Thu Jan 29, 2009 4:50 am

Post by J Randall » Sun Feb 22, 2009 1:21 pm

Roger-

Yes, that's what I was thinking. And then that the position size should then be maintained throughout each calendar year, and of course that the DRT applies only to the starting equity on January 1st. You can read my original post. I think that was the money management method they used as far as I am aware . . . but . . . I imagine that Tim would know for sure as well as others on the forum who have been in closer contact over the years with people who have first-hand knowledge.

But as I said above for me it is really more a curiousity and just something I chose to point out. I don't think I want to trade that way or probably spend a lot of time looking into it. And I don't mean to slight Tim or Blox. I think the program is the best on the market for non-programmers by far.

With all that said if you have something you want to share that would be great! :D

Right now I have only the Turtle version of Blox and in the coming few days I will upgrade to Pro. I was not 100% sure about Builder because I am not really a programmer. I did do the 7-day trial of TBB, and I agree all around it's a great program.

I have plenty of other ideas that I would like to test that are beyond the scope of the built-in systems and the included blox to tell you the truth. My plan was to sort of exhaust all the functionality of the Pro version first and then exhaust all of the free additional blox on the blox marketplace. After that I was then planning to pay for someone to program a few simple additional blox regarding entries and exits that I am particularly interested in. Once this is all done, then I may or may not consider the TBB version.

Thanks for your reply,
J.

Roger Rines
Roundtable Knight
Roundtable Knight
Posts: 1946
Joined: Wed Oct 06, 2004 10:52 am
Location: San Jose, CA

Post by Roger Rines » Sun Feb 22, 2009 2:41 pm

J Randall wrote:With all that said if you have something you want to share that would be great! :D
My posting was refering to the fact that Trading Blox Builder allows a user to generate streams of data that can be derived from another source. In this case, your comment would divert the usual process of depending upon the built-in equity streams, to a modified source of that information where the closed component would contain the Open-Equity component of all active positions on the close of each year. This would mean that on the first day of any new year period, there would be no Open-Equity. This process would show that the user developed and maintained equity streams would show the Close and Total equity values at the start of the first day of each year would be equal, regardless of the active position count.

To do this, a separate equity-series set could be kept so the math could be applied daily, or all active trades could be closed on the last day, and those same positions could then be opened again on the open of the first day of the new year, but the restarted positions would be filled with the Close price of when the active positions were closed on the last day of the previous year. Closing the positions would add cost, but that could also be adjusted out as well so the closing component wouldn't be affected.

While I agree with you, I don't see a useful application for mind-game processes either, but I do hope my comment exhibits how capable Trading Blox Builder can be in enabling new ideas become part of a new reality.

Post Reply