Dead Turtles

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.
AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Tue Oct 25, 2011 12:44 pm

Now look at the period 2003 to 2007. A very different story.
Attachments
Ab2.JPG
Ab2.JPG (85.17 KiB) Viewed 9372 times
Last edited by AFJ Garner on Tue Oct 25, 2011 12:46 pm, edited 1 time in total.

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Tue Oct 25, 2011 12:46 pm

The final screen shot is of the entire period on the same portfolio and parameters from 1995 to date. It worked out fine, if sub-optimal. A better curve could have been had, with hindsight, with a larger portfolio with risk limits.

Guess when I was trading it!
Attachments
Ab3.JPG
Ab3.JPG (71.01 KiB) Viewed 9368 times

marriot
Roundtable Knight
Roundtable Knight
Posts: 350
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Tue Oct 25, 2011 3:07 pm

> Guess when I was trading it!

From 1995 to 2003 ?

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Tue Oct 25, 2011 4:19 pm

Would that I were that prescient.

Bounty
Senior Member
Senior Member
Posts: 33
Joined: Tue Jun 19, 2007 8:59 am

Post by Bounty » Wed Oct 26, 2011 8:17 am

What changes have you made based on this experience--any advice for use newbies?

I understand that any system should be tested over multiple time frames. So is the idea to try to determine the type of market a system does well in and then include it in the suite for those types of markets?

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Wed Oct 26, 2011 9:29 am

There are others on this forum better to qualified to answer your questions than I. I have learnt, ignored and then learnt again some of the most basic lessons you will have read about in all the standard and splendid books on trading such as Wizards and Reminiscences. The markets are about fear and greed. Several times on bull runs I have become greedy and over-leveraged; when the inevitable steep draw down starts, the fear sets in and I regret having over-stepped my psychological limits.

Back testing is a fascinating and extremely valuable resource. Use it well. I and others have pointed out ad nauseam its drawbacks and complexities; the flap of the butterfly's wings. Flap your wings a few times in back testing and you will see what I mean.

Trend following works; given a basic TF system it is down to money management and keeping leverage at manageable levels. Most people over-estimate what they can put up with. Whatever back testing purports to show, you can place a fair sized bet that you will be in for a rougher ride.

That is the most frank and honest answer I can give you.

babelproofreader
Roundtable Knight
Roundtable Knight
Posts: 138
Joined: Wed Nov 10, 2004 4:36 pm

Post by babelproofreader » Wed Oct 26, 2011 3:26 pm

So is the idea to try to determine the type of market a system does well in and then include it in the suite for those types of markets?
In my opinion yes, but of course easier said than done.

marriot
Roundtable Knight
Roundtable Knight
Posts: 350
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Thu Oct 27, 2011 2:22 am

To make short the story, i think that Afj is suggesting us :
No wonder how tests looks, thing can become worst and remain worst for a long, long time.
Ok,got it.
But this is not going to happen to me :D

rgd
Roundtable Knight
Roundtable Knight
Posts: 396
Joined: Thu Aug 10, 2006 5:21 pm

Post by rgd » Thu Oct 27, 2011 11:24 am

To me, the most important part of backtesting is understanding when you are fooling yourself.

To the degree you are certain you have found the holy grail, and are about to embark on a career that will make Winton look bad, remember... pride goeth before the fall!

To the degree you live in constant fear that your models won't even generate half the performance of your backtest, then you may be on the right track!

Bravochico
Full Member
Full Member
Posts: 24
Joined: Wed Sep 14, 2011 11:46 pm
Location: Southern California, Ca

Post by Bravochico » Wed Nov 02, 2011 12:21 pm

Afj Garner brings up great points.

A rougher ride is almost the rule, not the exception.

When you deciding on your strategy, you also have to consider that there is high probability that you will do some tweaking to the strategy. I don't know any trader that has not made at least a few changes over the years.

Well, except maybe John Henry. Look at how his performance went limp as the years went by. To be fair, perhaps his 30 year old girlfriend and Red Sox had something to do with it. Correction- just Red Sox.

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Wed Mar 14, 2012 1:15 pm

“Occasionally, someone trying to promote something or start a debate will argue that trend following rules must always change due to changing market conditions. This is nonsense. It is a specious argument.â€

drm7
Roundtable Fellow
Roundtable Fellow
Posts: 96
Joined: Sun Apr 20, 2003 9:02 pm
Location: Richmond, VA

Post by drm7 » Wed Mar 14, 2012 1:25 pm

AFJ,

I'm a little dense today, so bear with me...are you implying that Dunn should have left their systems alone, or were correct in modifying them. I couldn't tease it out of the context of your post.

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Wed Mar 14, 2012 1:31 pm

Oh, sorry. Its an old story relating to a web marketer and seller of systems (please no one mention his name or this whole thread gets deleted by Tim).

The guy made the statement above on his sales website with particular reference to Dunn. I contested this assertion vigorously on another forum. Things DO change. That's why the original Turtle System is now a loser. My argument is that change has to occur in systems to keep up with changing conditions in markets. The due diligence documents of CTAs proudly vaunt their research and the reason for continued research is to make sure your systems continue to work.

Dunn was correct to modify. In my view. He made the modification following the super steep drawdowns and his systems have been recovering ever since.

JL25
Contributing Member
Contributing Member
Posts: 8
Joined: Thu Sep 08, 2011 5:23 pm

Post by JL25 » Wed Mar 14, 2012 4:05 pm

These are all such interesting and critical issues re system trading. If markets do change over time, at what point in ones trading do you modify/alter the system to better adapt to the current trading environment?

What evidence is required to suggest a a change is required? Perhaps if the system is currently in a drawdown that is greater that any previously experienced - actual or sim? Just thinking out loud, as I have no clear answers to these questions....many thanks to all for sharing your thoughts.

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Wed Mar 14, 2012 5:56 pm

JL25 wrote:These are all such interesting and critical issues re system trading. If markets do change over time, at what point in ones trading do you modify/alter the system to better adapt to the current trading environment?

What evidence is required to suggest a a change is required? Perhaps if the system is currently in a drawdown that is greater that any previously experienced - actual or sim? Just thinking out loud, as I have no clear answers to these questions....many thanks to all for sharing your thoughts.
Frankly, do any of us have much of a view? Dunn left it awfully late: apparently our Bill personally sat with much of his own money through that terrible 70% DD. At least if the sales guy I met is to be believed. Ouch.

rossb34
Full Member
Full Member
Posts: 12
Joined: Fri Dec 31, 2010 7:04 pm
Location: Middleton, WI

Post by rossb34 » Wed Mar 14, 2012 8:23 pm

AFJ Garner,

Thank you for sharing that story, it is always interesting hearing such unique experiences like that.

Here is a quote from a Dunn interview where he talks about making annual "adjustments" to his system in an article in Futures Mag.
The World Monetary program currently trades 13 markets: four currencies, gold, crude oil, T-bonds, Eurodollars, gilts, sterling, bunds, the FT-SE 100 and the S&P 500. Back testing 10 to 12 years (each year has equal weighting), Dunn annually adjusts the parameters of trading signals and each markets weighting. In February - just as the grains were about to take off - he dumped the entire grain sector. But Dunn has no regrets.
"For the last six to seven years, the grains have been difficult and less liquid than they used to be. They were more trouble than they used to be. They were more trouble than they were worth." Dunn says. By contrast, the currencies "have been good as long as theyve been around."
Copyright (c) 1996, Futures Magazine

Aaron01
Roundtable Knight
Roundtable Knight
Posts: 113
Joined: Thu Jun 02, 2011 1:21 pm

Post by Aaron01 » Wed Mar 14, 2012 10:33 pm

Is adjusting your parameters really changing your system though? Perhaps we should ask Theseus...

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Thu Mar 15, 2012 4:24 am

rossb34 wrote:AFJ Garner,

Thank you for sharing that story, it is always interesting hearing such unique experiences like that.

Here is a quote from a Dunn interview where he talks about making annual "adjustments" to his system in an article in Futures Mag.
The World Monetary program currently trades 13 markets: four currencies, gold, crude oil, T-bonds, Eurodollars, gilts, sterling, bunds, the FT-SE 100 and the S&P 500. Back testing 10 to 12 years (each year has equal weighting), Dunn annually adjusts the parameters of trading signals and each markets weighting. In February - just as the grains were about to take off - he dumped the entire grain sector. But Dunn has no regrets.
"For the last six to seven years, the grains have been difficult and less liquid than they used to be. They were more trouble than they used to be. They were more trouble than they were worth." Dunn says. By contrast, the currencies "have been good as long as theyve been around."
Copyright (c) 1996, Futures Magazine
Yesterday the salesman and analyst I met reported that “agriculturalsâ€

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Sun Apr 22, 2012 11:44 am

Here is an update on the Dead Turtles System. Doubtless it will have its difficulties in the months and years to come but it has fared pretty well over the past 5 years. Same parameters exactly as those I used for the article although the data may have changed a bit (even disregarding the addition of data since the article was written).
Attachments
Dead Turtles.JPG
Dead Turtles.JPG (77.31 KiB) Viewed 7441 times

AceofAce
Senior Member
Senior Member
Posts: 49
Joined: Thu Sep 30, 2010 5:06 am

Re: Dead Turtles

Post by AceofAce » Tue May 01, 2012 12:56 pm

[quote="AFJ Garner"][quote][b]HotTip
Modified Turtle System
I've read your posts and your article in ActiveTrader magazine this month. Excellent article. Are you able to share your parameters from the modified Turtle system? I'm considering purchasing Trading Blox and would like to recreate your equity chart.
Thanks for any insight you might provide.
[/quote] [/b]

I received the above PM on the Truly Awful “elitetraderâ€

Post Reply