Help - Search - Members - Calendar



Full Version: The Trend
Traders-Talk.com > TTHQ Directory > Investors University
Chart Guru Doug
I would like hear how traders here decide which way the trend is going. Can you give us your simple description?

Thanks a bunch! biggrin.gif
selecto
BB centerline (20 sma) with an 18 ema warning and a 7sma signal - in all windows.

"Mojo" also derived from various ma's & technicals of sentiment stuff.

(That's trend, not necessarily trading triggers.)
SemiBizz
This is easy, I apply Wyckoff's techniques, I look at volumes at swing points in the market. Looking at the Nasdaq composite, the two biggest days of volume in this timeframe were 12/7/04 and 1/3/05

Here's the datapoints:

................. High...... Low ...... Volume

12/7..... 2161.30...... 2114.65 ..... 2.59B

1/3 ..... 2159.64..... 2100.56..... 2.69B

First of all, we go to a lower low on higher volume, that's the first part. Secondly, even with higher volume we were unable to penetrate the high on 12/7.

Conclusion: Weakness, Downtrend.

Volume/price interaction says it all in my World.

What I'd be looking for in a trend reversal here would be a trading day over the 2.69B shares and a swing high over the 2161.30... Although I have my hunches about the market trend and where it goes, the volume/price interaction is what I use to guide my trading....
Sentient Being
I tend to take my computer generted buys then:

1. I draw a line across the tops and the bottoms in the time frame I'm looking at.

2. Or I'll drop in a channel tool and let it draw the lines as I establish the time frame.

3. And if all else fails, look at my position, if I'm long the trend is down and if I'm short the trend is up. I'm really good at getting it wrong. cry.gif
iron cross
two ways

Daily chart with 20, 50,200 ema

Weekly chart with 8 20 50 ema

The weekly carries more weight
SemiBizz
TREND TRADING STRATEGY:

Using Wyckoff, when the Nasdaq hit the 2159, 1/3, on what appeared to be a blowout volume day, that set up a LOW RISK ENTRY. I shorted NVDA at 24 that morning, NTGR at 19. This was easy, because if the Nasdaq subsequently had gone over the 2161, that would be my stop... That's the beauty of Wyckoff, Risk/Reward.
Rogerdodger
Which trend? 5 min, hourly, daily, weekly?
SemiBizz
QUOTE (Rogerdodger @ Jan 30 2005, 01:16 AM)
Which trend? 5 min, hourly, daily, weekly?
*


Exactly...
Chart Guru Doug
QUOTE (Rogerdodger @ Jan 30 2005, 12:16 AM)
Which trend? 5 min, hourly, daily, weekly?
*

If you use different methods for different timframes, feel free to post them.

Good thread going guys, keep it up! biggrin.gif
Stickan
I find that regression channels with 2% deviation is great at catching trends on all levles. Experimenting with other % settings can also give good results (like fib 1.618 etc). Different markets has different % alsodifferent dimeframes can vary in the same market.
Combining regr.channels with Median Lines can sometimes support findings. Like in this chart
The regressin channels are with 2% deviation (blue down and red up) and the green Pitchfork is parallell with th]e channels, which supports the opinion that the main tend is up (unless you drop below the grey line at D).
I use tops and bottoms for the regression channels. Thus if you suspect we have a bottom at hand, use the previous top together with the supected bottom. If ithe channel fits and you the have a clear break of the upper channel line a tredn break is probably what happend.



In this chart below I have used data from what some say is the start of this bullmarket 1982 up until today. Std deviation is set to 3% upper and 2% lower.
stig

da_cheif
QUOTE (Chart Guru Doug @ Jan 29 2005, 11:03 PM)
I would like hear how traders here decide which way the trend is going. Can you give us your simple description?

Thanks a bunch! biggrin.gif
*

how i do it is to watch the chart of what most stocks are doing by the broadest measure there is.....the weekly ad line.....which has been going straight up since it bottomed in march 2000....also look at the monthly charts of the Sydney all ords.....dj 65 comp...nyse comp....value line arithmatic....morgan stanley cyclical...sp600...russell 2000....mexico ipc...... cool.gif
esther231
I keep it simple. I start with higher highs, higher lows, lower highs, lower lows on the hourly. When I see a change there, I switch to daily.
I look at the 9,20,50,70 EMA. I might close a position based on what I'm seeing but I do not take a new position till I see how the market responds at the 20EMA.

To catch a trend change, I watch RSI over or under 50, STO 30,3 and OBV as confirming. I watch for strong market movement with a trend change. You can bottom slowly or top slowly but to be a believer a trend has changed, I want high drama with strong volume.
IYB
Great question, Doug. Personally, I don't think there is a more important question a trader could ask! Yet, the answer is allusive - which may well account for about 99% of our frustrations in this biz. wink.gif I'll lay out my thoughts on this, fwiw. Sorry if they are not as simple as some would like.

In the chart below, the 4-year cyclical market trend (what we normally refer to as bull market or bear market) is approximated by the direction of the 55 week MA, and in the next chart, the IT Trend is approximated by the 55 day MA. Finally, in the third chart, the ST trend is approximated by the 55 hour MA.

But that's a huge oversimplification. There is NO easy answer. I would submit that identifying the "trend" in the market is both one of the most difficult, yet one of the single most important determinates of success that we face as market traders. If we get it wrong - we generally lose. Period. And vice-versa.

Of key importantance to me is the "maturation" of a trend. Each has a beginning, a middle, and an end. As Yogi Berra has said "It's not over till it's over", and this seeming nonsensical sentence turns out to have huge implications for traders. My biggest errors occur in proclaiming a trend OVER too soon or too late. Each trend goes thru several phases of its life-as we do, for example. It has been said that a bull market is born in fear, grows in skeptiscism, matures on confidence, and finally dies in euphoria. This is true of each uptrend - no matter what the time frame. Downtrends have thier own maturation characteristics. So in order to know what the current trend is, we must be watching to see if a trend that began at a particular point has or has not gone through the necessary steps to have completed, and if not - just where we are within that progression. Tricky stuff- again, no easy answers.

Following the charts are my "Four Friends" thoughts on how we know where we are in the cycle at any given time, last updated in October, 2004. Imperfect as it is, I think it addresses the "trend" question.

[img]http://stockcharts.com/def/servlet/SharpChartv05.ServletDriver?chart=$SPX,uu[g,a]whclyyay[d19990323,20051223][pb55][vc60][iuah12,26,9!ld20!lo14!lh14,3][J32449460,Y].gif[/img]

[img]http://stockcharts.com/def/servlet/SharpChartv05.ServletDriver?chart=qqqq,uu[r,a]dhclyyay[dc][pb55][vc60][iuah12,26,9!ld20!lo14!lh14,3][j21152059,y]&r=5749.gif[/img]

[img]http://stockcharts.com/def/servlet/SharpChartv05.ServletDriver?chart=QQQQ,uu[g,a]ehclyyay[d40][pb55][vc60][iuah12,26,9!ld20!lo14!lh14,3][J19218173,Y].gif[/img]

QUOTE (IYB @ Oct 2 2004, 12:07 PM)
Four Faithful Friends – October 2, 2004

Monetary trend
Cyclical trend
Sentiment
Momentum

Originally posted in May, 2003. Here is another update, as this cyclical bull enters its end stages, highly likely, in the opinion of this author, to be marked by a magnificent final rally into 2005. The following is a very long winded explanation of why I say "the trend is up".  wink.gif Ideas and opinions expressed here are those of this writer only, and do not represent recommendations to buy or sell any security. Caveat Emptor.
_____________________________________________________

I. Monetary trend: The plain fact is this: when monetary policy is in an expansionary phase, cyclical bull markets ultimately follow. Likewise, when monetary policy is in a contraction phase, bear markets eventually follow. But there is virtually always a lag time. New bull or bear markets begin 6-18 months after the Fed policy reverses policy- as the cyclical bull market did from the October 2002 bottom, after more than a year and a half of Fed easing. Subsequently, rates continued downward, bottomed, and held at low levels - but then the Fed began to raise rates several months ago, after a three-year period of accommodation. The Fed cycle has reversed, and the window for a cyclical bull market peak is now upon us. Sentiment and momentum measures need to be monitored for signs of the cyclical bull market peak. But, meanwhile:

We are currently in a cyclical bull market.
_________________________________________________________

II. Cycles: Tim Wood in a May 9, 2003 article posted here, said it best: ‘Going back to 1896 the 4-year cycle has averaged 47.08 months in duration. The last confirmed 4-year cycle bottomed in September 1998 at 7,400.30 on the Dow. The next 4-year cycle low was ideally due in the fall of 2002’. {it did so in October 2002} The plain fact is that the cyclical market has bottomed in the second year of the election cycle for decades, almost as faithfully as the moon comes up full every 28 days. 2002 was no exception, nor will be 2006. In a similar fashion, bull cycles peak right around, or shortly after, the presidential election, as it did in late 2000, for example. Look for the peak around Jan 2005, plus or minus a couple of months.

If you subscribe to Decision Point, you can see the 4-year cycle chart here:

Four Year Cycle

And following is a chart of the SPX since 1999. The Primary Trend can be seen a defined by the 55 week MA, and recent IT trends are labeled, up fron Mar '03 to Jan '04, then down into August '04, then having turned back up once again: 

[img]http://stockcharts.com/def/servlet/SharpChartv05.ServletDriver?chart=$SPX,uu[g,a]whclyyay[d19990323,20041223][pb55][vc60][iuah12,26,9!ld20!lo14!lh14,3][J32449460,Y].gif[/img]

But, meantime:

We are currently in a cyclical Bull Market.

____________________________________________________________

III. Sentiment: Earlier incarnations of this “Four Friends” post have detailed the EXTREME bearish sentiment in the fall of 2002, then again in Feb, March 2003, and the opposite – extreme bullish sentiment - at the first Intermediate Term peak of this four year cycle in December 2003/January 2004. The latter extremes set up the IT decline which followed throughout Jan-March 2004, to be followed by a protracted base which formed over March-Aug 2004.

By now we’ve come full circle. In May 2004, Sentiment readings were amazingly bullish once again. Ten-day Open ARMS readings, for example, at 1.48 in March, the internal low of this IT base, reached their best readings since September 2001 (following 9-11)! Now THAT’S extreme!

ARMS INDEX

Meanwhile, at 1.08, the CBOE Call/Put Ratio 10 DMA in May EXCEEDED its September 2001 reading, or any other reading this decade. Now, again, THAT’S extreme! And it remains (late in June) at levels associated with bottom areas, with CBOE 10-DMA Put/Call Ratio in the 90’s and the Equity P/C Ratio in the high 70’s:

Put/Call Ratio

In short:
Sentiment has reached multi-year record extremes, fully supportive of the next IT advance. We have a long, long way to go before these readings get to levels even approaching IT top levels. Meanwhile:

The market is in the very early stages of an Intermediate Term Advance.
________________________________________________________________

IV.  Momentum: Momentum, as shown by the seven sentinels linked below:

  Sentinels

as well as by MCO indicators for each of the various indices, by advance/decline cumulative data and A/D volume cumulative data as reported here and here (Fib’s board):

  Cummulative Data

..reached bottom extremes in March/May/August, and turned up. Notice that cumulative A/D lines, A/D Volume, and the rest, are now up and away from their consolidation areas, indicating that the new IT uptrend is underway.  Notice, too, that as all of the major stock indices moved to new yearly lows in May, most major measures of market internal momentum held above their March 2004 lows in May, then likewise in August, held above the May levels – A clear indication that the IT had bottomed by August 2004. Here is an excellent example of that – as the SPX made lower lows in Mar, May, August…the SPX MCO made a series of rising bottoms, each subsequent one higher than the previous low.

McClellan Oscillator

The Summation Index for EACH major stock index or group is in positive territory, and rising, indicative of a new Intermediate Term uptrend. This is occurring, of course, within the context of a cyclical bull market. The trend is your friend, and the trend is UP.

These and ALL other momentum measures that I follow are consistant in their message, which is:

The market is in the very early stages of an Intermediate Term Advance.
__________________________________________________________________

Overall Conclusion: We are in the last 30% of this Cyclical Bull Market, but just beginning a new IT run.
--------------------------------------------------------------------------------------------

As always, caveat emptor, and please don’t forget to place a stop with every entry order.
*
trinharder
Different tactics for different situations, but I agree with Stickan in that my primary indication of a change in trend is the break of the top or bottom rail of a regression channel. As a swing trader 99% of what I do is based on the daily chart.
Chart Guru Doug
QUOTE
Sorry if they are not as simple as some would like.

Damn, such is life, but then if it was easy, we wouldn't need all this discussion.
QUOTE
It has been said that a bull market is born in fear, grows in skeptiscism, matures on confidence, and finally dies in euphoria. This is true of each uptrend - no matter what the time frame.


That sure seems like words of wisdom to me.

QUOTE
Following the charts are my "Four Friends" thoughts on how we know where we are in the cycle at any given time, last updated in October, 2004. Imperfect as it is, I think it addresses the "trend" question.


Of your 4 friends,
would you say;
the monetary trend is close to/turning now?
The 4 year cycle is close to/turning now?

Those other indicators I cant get charts for but I am wondering if you could put annotations on your sentinials link charts showing what would trigger the buy/sell.
Thanks for your great post Don...

cheers cgd
IYB
>>Of your 4 friends,
would you say;
the monetary trend is close to/turning now?<<

Definately - the turn in short term rates typically leads the peak in the equities market by 6-15 months. We are well into that window.

>>The 4 year cycle is close to/turning now?<<

Yes, imho. Timewise - I believe we are very close. What I see as still conspicuously absent is an overwhelming sense that the market is going a lot higher, and has virtually NO downside. I didn't see December as nearly euphoric enough. I think that we need one last "run for the roses" and who knows...maybe this Iraqi event will be the catalyst... then a top formation, and then it's OVER.

>>Those other indicators I cant get charts for but I am wondering if you could put annotations on your sentinials link charts showing what would trigger the buy/sell.<<

6 of 7 are already on BUY. Just one small up day tomarrow would trigger a 7 of 7 FULL BUY SIGNAL, by taking the 7th - the $BPCOMPQ - back above lower boundry of the Bollinger Bands. It's looking very much like will happen, given the news background...

Thanks for your great post Don...

cheers cgd

Thank YOU for a great string, Doug! Aloha.
Spidergann
If you follow the Trend you get in the rear end....or you become an old man

Sorry

SG
uncleharley
rolleyes.gif I look for the sawtooth. If the drops are quicker than the rises, I call it a downward trend. If the market rises faster than it falls, I call the trend up.

uh
Chart Guru Doug
QUOTE (Spidergann @ Jan 31 2005, 10:39 AM)
If you follow the Trend you get in the rear end....or you become an old man

Sorry

SG
*

Not to worry, they make toilet paper for that! laugh.gif
Chart Guru Doug
BTW
A trend for me is usually defined by patterns and what they are doing. If I see a bunch of bearish patterns breaking down and hitting their targets I can trade them that way. If I see a bunch of bases exploding to the upside=I play them that way.
If bearish patterns are failing and breaking up, then I play them that way and usually watch the action in the individuals sector. My bias for trading is usually on the long side though...
cheers..
TrillionDollarMan
I like the way Colin Alexander described trend identification in his book: TIMING THE STOCK MARKET.

He uses a 25 and 40 period moving average in time frames.

So, if the 25 is above the 40 in whatever time frame we are talking, the trend is up, or vice versa if down.

Then, the trick is to figure out what the PROGRESSION of time frames is with respect to the trend...and one is in good shape.

Of course a decent way to do that, is as per Elder's triple screen, using MACD historgram as a guage of longer term trend, and stochastic as an indication of the next shorter time frame trend.

I suppose trendlines and Dave Landry's big blue arrow could do it also, if all else fails!

TDM
TrillionDollarMan
QUOTE (TrillionDollarMan @ Jan 31 2005, 09:19 PM)
I like the way Colin Alexander described trend identification in his book:  TIMING THE STOCK MARKET.

He uses a 25 and 40 period moving average in time frames.

So, if the 25 is above the 40 in whatever time frame we are talking, the trend is up, or vice versa if down.

Then, the trick is to figure out what the PROGRESSION of time frames is with respect to the trend...and one is in good shape.

Of course a decent way to do that, is as per Elder's triple screen, using MACD historgram as a guage of longer term trend, and stochastic as an indication of the next shorter time frame trend.

I suppose trendlines and Dave Landry's big blue arrow could do it also, if all else fails!

TDM
*



PS -- in perusing the posts, I also like the idea of regression channel, pitchfork, etc. as a more complex form of trendline, and one that can be more indicative perhaps, in terms of channel breaks, etc. And Bollinger Bands, also have a std. deviation component can work in some fashion also.
Douglas
On a daily chart trend direction called when Demand Index Crosses "0", 6 day + DI minus -DI Crosses "0" and the direction of the 7 day TRIX all agree.
deacon
here's some news plays you might want to pick from at 8am

another DSCO inferno perhaps

WIND ONNN EVOL LSBC

i like macd on weekly first for trend...ike always said 50MA 60min for trend

looking at some monthly charts last nite, the best monthly macd was XOI.X and EBAY(!)
cupboardbear
QUOTE (Chart Guru Doug @ Jan 30 2005, 12:03 AM)
I would like hear how traders here decide which way the trend is going. Can you give us your simple description?

Thanks a bunch! biggrin.gif
*


My view is that the market is always trending, in any timeframe. Either it's trending up, down or sideways. So, the first thing I look for is what it isn't doing ... the obvious is easiest to spot. smile.gif

Since sideways is what it does most of the time (what most techs call 'trendless'), I find the origin of that sideways pattern, then compare the prior up/down trending time and distance covered, for an early guess of how long the sideways could last.

Sideways trends are trading gold mines if you're not looking for an emminent breakout to possibly kill your trade. For all intents and purposes, I look for trends inside of trends which are embedded inside even shorter trends.

That's my mental attitude toward determining a trend. Safety first, trade/money management next, entry/exit targets last, and all that wrapped up in a timing blanket.
calmcookie
QUOTE (SemiBizz @ Jan 30 2005, 12:16 AM)
This is easy,  I apply Wyckoff's techniques, I look at volumes at swing points in the market.  Looking at the Nasdaq composite, the two biggest days of volume in this timeframe were 12/7/04 and 1/3/05

Here's the datapoints:

.................        High......          Low ......          Volume

12/7.....  2161.30......  2114.65 .....      2.59B

1/3 .....  2159.64.....    2100.56.....        2.69B

  First of all, we go to a lower low on higher volume, that's the first part.  Secondly, even with higher volume we were unable to penetrate the high on 12/7.

Conclusion: Weakness, Downtrend.

Volume/price interaction says it all in my World.

What I'd be looking for in a trend reversal here would be a trading day over the 2.69B shares and a swing high over the 2161.30... Although I have my hunches about the market trend and where it goes, the volume/price interaction is what I use to guide my trading....
*
disabledtrader
QUOTE (calmcookie @ Feb 13 2005, 10:14 AM)
QUOTE (SemiBizz @ Jan 30 2005, 12:16 AM)
This is easy,  I apply Wyckoff's techniques, I look at volumes at swing points in the market.  Looking at the Nasdaq composite, the two biggest days of volume in this timeframe were 12/7/04 and 1/3/05

Here's the datapoints:

.................        High......           Low ......           Volume

12/7.....  2161.30......   2114.65 .....       2.59B

1/3 .....   2159.64.....    2100.56.....        2.69B

   First of all, we go to a lower low on higher volume, that's the first part.  Secondly, even with higher volume we were unable to penetrate the high on 12/7.

Conclusion: Weakness, Downtrend.

Volume/price interaction says it all in my World.

What I'd be looking for in a trend reversal here would be a trading day over the 2.69B shares and a swing high over the 2161.30... Although I have my hunches about the market trend and where it goes, the volume/price interaction is what I use to guide my trading....
*

*



This is easy for my first post. Two or more higher lows in the time period you are trading defines an uptrend.
disabledtrader
Sorry, I did not mean to quote someone with my reply. This was my first post and I, at first, could not locate the post reply button.

sweatingbullets.gif rolleyes.gif
Chilidawgz
QUOTE
Which trend? 5 min, hourly, daily, weekly?



tada!

Channel lines, Chapman wave and MACD
greenie
Dan, you wrote:

>6 of 7 are already on BUY. Just one small up day tomarrow would trigger a 7 of 7 FULL BUY SIGNAL, by taking the 7th - the $BPCOMPQ - back above lower boundry of the Bollinger Bands. It's looking very much like will happen, given the news background...

I do not quite get it. To me all seven sentinels look like in sell. NAMO, NYMO, NAHL, NYHL are all below 6MA. TRINQ 3MA above 13MA. Only one close to buy is TRIN. Am I reading the sentinels wrong ? Thanks, G.
jabba
Ed Seykota once said... print the chart, post it on the wall at the other end of the room. Usually becomes clear if the direction is UP, DOWN, or SIDEWAYS
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.