Pullbacks and LTTF methods

How do you know when a trend has started? Ended? This forum is for discussions about trend indicators and signals.
Post Reply
ladadriver
Senior Member
Senior Member
Posts: 25
Joined: Tue Feb 27, 2007 10:08 am

Pullbacks and LTTF methods

Post by ladadriver » Sat May 24, 2008 6:02 am

One other testing idea I've yet to investigate fully relates to market structure. Let me expand with the hope of attracting other ideas from the board.

Typically, trading breakouts allows one to jump on markets moving beyond certain historic highs and lows.
But is breakout trading efficient in terms of risk:reward? I would argue the technique is functional but not very efficient.

Any amount of market study reveals that prices move in waves, constantly probing higher and lower for direction. In this line, it's my belief that price action is king. Experience shows me that breakout trades often retrace quite quickly, giving the trader unwanted heat. From discretionary trading experience (mostly in Stock Index Futures) I can safely say my favourite place to trade is on a pullback in a trending market. Trading in this fashion substantially improves the risk:reward profile, relative to a vanilla breakout trade. However, when one looks at classic trend following methods (Donchian BO, DMAX, TMAX, BBBO, PGO etc), pullbacks don't feature once. Am I barking up the wrong tree? Or is there opportunity in applying pullbacks to LTTF?

Look forward to comments.

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

Post by BARLI » Sun May 25, 2008 1:02 am

ladadriver, I agree that stock index futures is the best to trade on pullbacks.. have you done any research concerning markets moving in waves?

ladadriver
Senior Member
Senior Member
Posts: 25
Joined: Tue Feb 27, 2007 10:08 am

Post by ladadriver » Sun May 25, 2008 1:01 pm

BARLI wrote:ladadriver, I agree that stock index futures is the best to trade on pullbacks.. have you done any research concerning markets moving in waves?
BARLI, I'll need to rearrange your statement to agree with you. I prefer, 'on stock index futures, pullbacks are the most favourable short term trades'. In terms of your question on wave research, I haven't consciously researched with that word in mind. But then LTTF might be considered the business of riding long term waves. Emotive language combined with the results of 'complex' R&D does make great sales material though ;) Ironically, simplicity is rarely convincing. I digress...

One idea that may be of interest is to consider that there is no such thing as the trend - rather, that there are two trends. Consider two traders playing one market. One is has the mandate to sell, exclusively, and the other the mandate to buy, exclusively. Is it possible for them both to make money over a given time period? Yes, if they avoid choppy/directionless conditions, manage risk and play the swings accordingly. This is not to say that one sided trends do not exist - studies of HHs, HLs, LLs & LHs will determine whether the bulls, bears or neither are in charge. The concept of concurrent bullish & bearish trends does, however, provide an understanding of waves in prices. Secondly, it increases the number of trading opportunities if one agrees to play both sides in agreement with context.

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

Post by BARLI » Sun May 25, 2008 2:04 pm

I agree that everything in comsos is binnary("+" and "-"; love and hate etc) but I disagree that there's always two trends in the action, there' are buy and sell but just like in particles, the chrage of the other one is greater and "dominates" the second one choosing the trajectory
ladadriver wrote: One idea that may be of interest is to consider that there is no such thing as the trend - rather, that there are two trends. Consider two traders playing one market. One is has the mandate to sell, exclusively, and the other the mandate to buy, exclusively. Is it possible for them both to make money over a given time period? Yes, if they avoid choppy/directionless conditions, manage risk and play the swings accordingly. .

TrendsCatcher
Roundtable Fellow
Roundtable Fellow
Posts: 97
Joined: Thu Jan 10, 2008 3:28 pm
Location: Southern California, USA

Post by TrendsCatcher » Tue May 27, 2008 4:55 pm

One problem with trading pullbacks is that in a strongly trending market, the pullback may never happen, so you may miss a big opportunity. This would be a major concern in Futures trading, but not in individual stocks trading. In futures trading, you have only so many markets, and therefore, you cannot afford to miss one big opprotunity. By trading breakouts, you assure that you will be in the big trend if it happens, becuase if a trend develops, by definition, sooner or later it will make a new high/low - thus, making new high/low is the NECESSARY CONDITION for a significant trend. Therefore, assuming one doesn't miss a breakout signal, he/she will be assured to be in significant trend, if there is one.

but in individual stocks, it's no big deal: there are so many stocks, and stocks as a class are so able to make significant trends in the intermediate-term time frame. There is no pressure for the trader to capture all the trends. There is always another one on its way to make a nice setup, in case you've missed the first one. So, in stocks trading, the trader can afford to fine tune the entry (eg, pullback vs breakout), while not giving up much - because missing a lot of trends will not necessarily prevent you from catching a lot of trends with with full exposure without violating money managment rules.

As far as the equity market is concerned, my experince is that an a brand new bull maket, when stocks have great momentum, trading breakout works fine, and pullbacks don't happen as frequently as one wishes in the best of the best stocks. But in almost all other general market enviorment (older bull market, consolidation phases, bear market, bear market rallies), trading pullbacks offers much better Reward/Risk ratio, with better odds that you don't get stopped out - higher reliability.

The key to entry is to find reliable "pivotal points" that once the market crosses, it will never look back, and will move significantly in your favor). In this type of situation, you get a good reward/risk ratio (because the market is not supposed to look back, so you can afford to use tight stop losses, reducing risk/contract|share, in turn boosting Reward/Risk) with a good reliabiltiy, regardless if this point is at a new high of certain trading period (eg, 20 bars, 55 bars etc - breakout) or not (pullbacks or other atypical setups). Another aspect to consider is how often this entry setup will happen in the markets you've chosen, the more often, the better.

Just my 2 cents

ladadriver
Senior Member
Senior Member
Posts: 25
Joined: Tue Feb 27, 2007 10:08 am

Post by ladadriver » Wed May 28, 2008 2:42 am

TrendsCatcher, thanks for the interesting post.

I agree with your points entirely. Using a pullback strategy exclusively could lead to missing one or more of the crucial 15R+ winners, adversely affecting one's trading statistics. As you say, this issue is not such a concern in stock trading, given the significantly broader universe of securities. My gut reaction, though, is not to discount pullback methods in a futures context. If one had a couple of different LTTF methods on the go already, missing the big trends wouldn't be a problem. From there, adding a pullback (and profit target?) method would hopefully add diversification and improve the risk profile.

On a tangent, in the far and distant future when I'm happy with my futures trading, I'd love to develop some stock trading models. For general interest, here's a stock trading pullback method I found on the web:

http://www.daisydogger.com/Trade%20Setup%27s.htm

One last tangent, ETFs. I'm most interested in AFJ Garner's Equities Without Tears document. Mechanical ETF trading sounds like fun. Especially if one expands it beyond equities. An ET member called Cutten posted this breakdown of a "passive long-term investment" portfolio. Food for thought, perhaps.

http://www.elitetrader.com/vb/showthrea ... ost1900935

TrendsCatcher wrote:One problem with trading pullbacks is that in a strongly trending market, the pullback may never happen, so you may miss a big opportunity. This would be a major concern in Futures trading, but not in individual stocks trading. In futures trading, you have only so many markets, and therefore, you cannot afford to miss one big opprotunity. By trading breakouts, you assure that you will be in the big trend if it happens, becuase if a trend develops, by definition, sooner or later it will make a new high/low - thus, making new high/low is the NECESSARY CONDITION for a significant trend. Therefore, assuming one doesn't miss a breakout signal, he/she will be assured to be in significant trend, if there is one.

but in individual stocks, it's no big deal: there are so many stocks, and stocks as a class are so able to make significant trends in the intermediate-term time frame. There is no pressure for the trader to capture all the trends. There is always another one on its way to make a nice setup, in case you've missed the first one. So, in stocks trading, the trader can afford to fine tune the entry (eg, pullback vs breakout), while not giving up much - because missing a lot of trends will not necessarily prevent you from catching a lot of trends with with full exposure without violating money managment rules.

As far as the equity market is concerned, my experince is that an a brand new bull maket, when stocks have great momentum, trading breakout works fine, and pullbacks don't happen as frequently as one wishes in the best of the best stocks. But in almost all other general market enviorment (older bull market, consolidation phases, bear market, bear market rallies), trading pullbacks offers much better Reward/Risk ratio, with better odds that you don't get stopped out - higher reliability.

The key to entry is to find reliable "pivotal points" that once the market crosses, it will never look back, and will move significantly in your favor). In this type of situation, you get a good reward/risk ratio (because the market is not supposed to look back, so you can afford to use tight stop losses, reducing risk/contract|share, in turn boosting Reward/Risk) with a good reliabiltiy, regardless if this point is at a new high of certain trading period (eg, 20 bars, 55 bars etc - breakout) or not (pullbacks or other atypical setups). Another aspect to consider is how often this entry setup will happen in the markets you've chosen, the more often, the better.

Just my 2 cents

TrendsCatcher
Roundtable Fellow
Roundtable Fellow
Posts: 97
Joined: Thu Jan 10, 2008 3:28 pm
Location: Southern California, USA

Post by TrendsCatcher » Fri May 30, 2008 10:38 am

See what Ed Seykota has to say regarding buying on dips

http://www.seykota.com/tribe/FAQ/2008_May/01/index.htm

See the last entry.

PS. I don't agree with him 100%.

I personally buy both breakouts and pullbacks, the core concept is something I got out of reading Livermore's stuff - pivotal points!

Angelo
Roundtable Fellow
Roundtable Fellow
Posts: 90
Joined: Fri Apr 29, 2005 4:31 am
Location: Italy

Post by Angelo » Sat May 31, 2008 2:44 am

TrendsCatcher wrote:See what Ed Seykota has to say regarding buying on dips

http://www.seykota.com/tribe/FAQ/2008_May/01/index.htm

See the last entry.

PS. I don't agree with him 100%.

I personally buy both breakouts and pullbacks, the core concept is something I got out of reading Livermore's stuff - pivotal points!

In the same FAQ page not on the pullback's subject, but really worth noting (at least to me):

_______________________________________
Fri, 9 May 2008 Back to the Future
Ed,
Are you still trading?
Was wondering from your perspective if the markets remind you of the 70's?

Ed's reply: Yes, and the 80's, 90's and 00's.
_____________________________________


Given the public real time results of long term trend followers like Bill Dunn or John Henry, this is a bit amazing, isn'it?

paloma
Full Member
Full Member
Posts: 22
Joined: Tue Oct 17, 2006 11:50 pm
Location: Stouffville

Post by paloma » Sat May 31, 2008 2:24 pm

Angelo,

Im am interested to see why you feel that long term trendfollowers like Dunn and Henry are accurate proxies for the systems that Seykota is using now.

I find it interesting that you do not mention the performance of long term trendfollowers like Eckhardt, Winton etc, Salem, Parker etc. whose systems are going in the opposite direction of Dunn's and Henry's.

Dunn and Henry seem to not have adapted as well as these other unmentioned trendfollowers have.

What Seykota most likely meant by the comment is that the only constant in the markets is change, but you may want to email him and see what he says.

Sebastian

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

Post by AFJ Garner » Mon Jun 02, 2008 3:19 am

paloma wrote:Angelo,
you may want to email him and see what he says.
Sebastian
I can hazard a fair guess at what Ed is likley to say:

"You might consider taking your feelings about <wanting to know> to Tribe." :D

Angelo
Roundtable Fellow
Roundtable Fellow
Posts: 90
Joined: Fri Apr 29, 2005 4:31 am
Location: Italy

Post by Angelo » Mon Jun 02, 2008 5:35 am

paloma wrote:
Im am interested to see why you feel that long term trendfollowers like Dunn and Henry are accurate proxies for the systems that Seykota is using now.

Hi Sebastian,

in this day, talking about trend following is a bit hard, beacuse there is always the risk to spark ideological wars..... sometime with some more practical consequences (thare are a lot of people out there, who make a living not by trading but by selling the trendfollowing or the mean-reverting mantra).

I don't want to start a new stint of this useless (for me) debate, but I notice you did a big leap away from my words: I was puzzled about Ed's words about markets; I haven't said anything about Ed's systems.

What I really have seen after two decades in the markets (which I feel is a time long enough to start making some judgements) is that they do change and - AS A CONSEQUENCE - trading results of my systems change too, and change a lot.
You can claim someone is able "to adapt" better than others (someone by skill, but someone by chance alone.... have you read Taleb?), but that is not really my point.
If you need to "adapt" ... something has been changed, isn't it?


About CTA's figures:

1) as you surely know, Ed Seykota is out of the race for the public track records: he chose a long time ago not to publish numbers, with the little expections of those contained in Schwager's book, which are a bit too dated to move any feelings in me. So I like to read his nice and witty comments, but I don’t even know to which measure he is involved in the markets right now (AUM, his 5 year’s performance?).
Chances are, you share the same level of knowing-nothing, no matter how many emails you can send to him.

2) I normally look at Henry and Dunn (among others) because of the longer published track record (Ed chose a long time ago not to publish numbers, with the little expections of those contained in Schwager's book, which are a bit too dated to move any feelings in me) and - yes - there are maybe the slower to change and - therefore - the more appropriate proxy of what "classic" trend following was.

3) ça va sans dire, I could have obtained some different results looking at the names you mention, but - at the end of the day - almost all of them have shown degrading performance in recent years, the mighty JP as well as the smaller ones like Hawskbill or Liz Cheval, so I don't really think there's something to quarrel about.



PS for a recent interview to Salem Abraham, that gives the pulse of how much one need to adapt his trading philosophy to trade OPM in today's markets:

http://www.sfomag.com/article.aspx?ID=1167


================================

For the majority of years, trend following was the bread and butter of ATC (which also has a fund for accredited investors; it’s a private placement, so the principals cannot discuss it). Yet smoothing out returns to keep customers from “using ulcer medicine and barf bagsâ€

paloma
Full Member
Full Member
Posts: 22
Joined: Tue Oct 17, 2006 11:50 pm
Location: Stouffville

Post by paloma » Mon Jun 02, 2008 1:02 pm

I never fully "believed" in the classic Dennis approach, it is simply one approach that Dennis used in the past, Dennis himself doesn't believe in this approach as much anymore. I believe in using reason to discover scientific methods of mechanically extracting profits from the markets that have shown a great enough level of robustness in my research for me to believe in the method.

I do not to equate trend following with buying 20 day breakouts and selling 20 day breakdowns (ie. Dennis approach).

I feel trend following is simply looking at the long term chart, use quantifiable methods to ensure this chart looks like it is going up alot or down alot in the long term (ie. >6mth basis), then use chart patterns to enter buy/sell stops in the same direction as the long term chart suggests the price is going to continue to go. The actual entry/exit method can get as creative as you want.

Also, I never stated that the trend followers I mentioned above were proxy for Dennis' classic trend following approach. But I do feel they are proxy for trend following as I have defined above in this post.

On a lighter note, I do agree that discussing trading philosophy is a touchy area and traders usually get nowhere discussing it with anybody. lol, the method that usually opens up my eyes the most to the truth about markets is my own research.

I wish you well Angelo. Happy trading.

Post Reply