Reverse technical analysis

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
neil9327
Contributor
Contributor
Posts: 3
Joined: Fri Nov 03, 2006 3:52 pm
Location: London, UK

Reverse technical analysis

Post by neil9327 » Sun Nov 05, 2006 6:35 am

I am fairly new to trading. I set up an account with Tradestation in the USA on the advice of a book I've read "High Probability Trading" by Marcel Link.
I typed in to this trading platform program a sample Easylanguage automated trading program from the book (page 263), which the author says works, and run a back test on a 1-minute chart for 100 days. It is a breakout strategy trading the S&P500 E-mini December futures contract. The back-test assumed no commission or slippage.

To my surprise it actually lost money, and quite alot of money at that: It made I think $15000 and lost $18000 making a total of around 600 trades.

So I am wondering whether technical analysis works any more? One scneario that might cause it to fail is if all the traders are using it, then the banks and market makers will surely have an opportunity to know when they have put trades on, and move prices to hit their stops, so causing these traders to lose.
In fact, it is almost worth using a reverse technical analysis strategy putting on opposite positions to those recommended (but commissions would probably eat profits).

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Sun Nov 05, 2006 2:26 pm

Neil,

If only trading were as simple as finding a system that lost a lot of money and reversing the entries - everyone could be successful since it's pretty easy to develop something that loses lots.

Normally it is the exits that make the money in a trading system, so reversing the entry on a losing system will just carry on losing. Correspondingly, a trading system that has good exits should make money regardless of which way round you trade the entries (or even if you take random entries).

Also, testing a system on 100 days on one contract is not exactly a fully representative test of whether the system has any value or not.

I suggest you decide whether you believe the concepts upon which the system is based are valid, understandable, and reasonable to you before you decide whether to use it or ditch it.

Hope this helps

Paul

p.s. I'm not familiar with the actual system you are using, but my comments should be generally applicable to any trading system.
Last edited by Paul King on Sun Nov 05, 2006 6:15 pm, edited 1 time in total.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Re: Reverse technical analysis

Post by BARLI » Sun Nov 05, 2006 6:03 pm

neil9327 wrote:In fact, it is almost worth using a reverse technical analysis strategy putting on opposite positions to those recommended (but commissions would probably eat profits).
neil, its what I do a lot when building new trading systems, I get an idea about the market, I code it, I run back tests, if there are lotsa trades (over 2000) on 20 years of data and results are not good and neither the equity curve is I simply reverse it and get something statistically significant. I agree with Paul about importance of exits, there's another side: How much pain the one can take untill the profitable exit. Its definitaly something to work with

Jason
Roundtable Fellow
Roundtable Fellow
Posts: 57
Joined: Sat Apr 26, 2003 10:31 pm
Location: Oakville, Canada
Contact:

Post by Jason » Mon Nov 06, 2006 9:22 am

Hi Neal,

Welcome to the forum.

I can really relate to your post as I have asked the same question of technical analysis myself. My answer provides a healthy perspective to operate within for ME. That answer is NO - technical analysis does not work. Rather, most on this forum use tested analysis. The distinction is subtle but critical. The veterans on this forum and in this business have poured through all the essential books and articles out there and have tested many of the systems, ideas and concepts. Most simply do not work. However, some do.

Determining what works for you is a deeply personal question. Myself, I get a thrill out of finding a concept that works across wide time frames, markets and systems. I consider a 100 day entry/50 day exit breakout system to be an idea. Whereas, "breakouts" are a concept.

In the end, it seems to pay to be an open minded critic who is numb to rejection. That is you need to be willing and eager to test anything. Yet, you must be completely ruthless and unforgiving in your analysis. Then, when yet another one of your ideas or concepts that you have spent minutes/hours/days/weeks/months/years on ends up in the trash heap, you must gleefully hop on the next project without hesitation.

So the 1 minute, 1 market, 100 day idea does not work. That says a little something about breakouts. It is good to be critical. However, for me the critic does not yet have enough ammo to stop the testing. More markets and different timeframes to be tested are required before any real conclusion can be made.

Good luck. You are in the right place. The people and the software here are very supportive.

Jason

neil9327
Contributor
Contributor
Posts: 3
Joined: Fri Nov 03, 2006 3:52 pm
Location: London, UK

Post by neil9327 » Mon Nov 06, 2006 4:26 pm

Thanks for your answers and the welcome.

I would like to clarify that my "reverse" strategy does include exits as well as entries, in the sense that it runs exactly the same as the recommended strategy, but substitutes buy orders for sell orders at every stage, so (assuming zero commissions and slippage) the result is always opposite in money terms.

I've posted the strategy and reverse strategy here:

http://dohat.com/link.html

(BTW I think I must have got something wrong here - the reverse strategy now only breaks even when I back-test it. But beforehand it was showing a good profit - but you can see the principle).

cramnhoj
Contributing Member
Contributing Member
Posts: 8
Joined: Thu Jul 06, 2006 5:27 am

Post by cramnhoj » Mon Nov 06, 2006 5:42 pm

Welcome to reality, where things aren't as simple as what's said in books. :lol:

The thing is technical analysis is merely a part of a trading system. Trading veterans usually consider the entry to contribute just 10% of a system's success. You haven't mentioned what money management the system uses, this makes a bigger factor to a system's success.

Regarding the reverse entry, yes the other guys understood it correctly and I agree that simply getting a system that losses a lot and reversing the entry won't give you a profitable system. This concept has been talked about over and over again in different trading forums, and that generally ends up with the idea being disproved.

The most repeated saying in trading is "Cut your losses, Let your profits run" and this saying is talking about exits not entries. This is because one can have a good entry yet have a lousy exit, hence end up with an unprofitable system. This is why Paul King said that it's the exits that makes your profits. Is your exit strategy able to keep the open profits or does it let profits go back to losses?

Also, in my opinion, 1 minute bars only lets you see the noise in the market. Noise being almost random means systems trading noise may look profitable at first but prove unprofitable in the long run.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Mon Nov 06, 2006 6:21 pm

cramnhoj wrote:
Regarding the reverse entry, yes the other guys understood it correctly and I agree that simply getting a system that losses a lot and reversing the entry won't give you a profitable system. This concept has been talked about over and over again in different trading forums, and that generally ends up with the idea being disproved.
how did they disprove the idea?

cramnhoj
Contributing Member
Contributing Member
Posts: 8
Joined: Thu Jul 06, 2006 5:27 am

Post by cramnhoj » Tue Nov 07, 2006 3:01 am

I'm not really sure. The discussion goes something like someone looking for a system that gives big losses per trade with the idea that by simply reversing the direction it'll produce big gains instead. But it was improper exits that let losses run that caused losses.

From what I remember in Van Tharp's book, some oscillators have winning/losing trade ratio of almost 1:1 in which case I guess reversing the entry signal might not matter as long as it also has good exits, money management and run on a diversified portfolio.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Tue Nov 07, 2006 4:30 am

what if you have a system that trades every day and has 30 % Win. trades?

Post Reply