It's rare that a week passes that I don't see what I consider to be an improperly drawn trendline. Even the example chart on the wikipedia page for 'trend lines' makes my skin crawl. (One of these days I'm going to try to edit that page and post a better chart.) I've posted the rules for drawing trendlines which I follow before:
For an uptrend within the period of consideration, draw a line from the lowest low, up and to the highest minor low point preceding the highest high so that the line does not pass though prices between the two low points (from Trader Vic’s excellent book ‘Methods of a Wall Street Master‘)
Duru prodded me today by sending me a link to a gold article titled " The Fuse Has Been Lit and The Rocket is Rising". Hmm, do you think the author might be a gold bull? If that title was enough of a clue, the site's name, 'Gold Stock Bull', makes it perfectly clear. The article begins by listing 16 fundamental reasons to be bullish on gold. ( Update: I see that Roger Nusbaum has also critiqued the same gold article but with more of a focus on the fundamental argument. ) The part that follows that list is what floored me though. That's where the author did some technical analysis of gold. Here's the first of his charts:
And here's the caption to go along with that chart:
Featured is the daily gold chart. Price has broken out from beneath 2 months of resistance. This matches the length of the pull-back in Nov. and Dec. Price is now free to soar! The RSI and MACD have turned positive (green lines). The rocket is rising!
My first reaction upon seeing the chart was "how in the world did he come up with that trendline?" After reading the caption I was really scratching my head. I was trying to find the "2 months of resistance" that price had just broken out of. I guess he meant the broken March - April down trendline. While it's true that the trendline was broken I'd hardly say that "price is now free to fly". What about resistance from the 50-day moving average? Or resistance from the three peaks in late January, March and April, near 950 -- not to mention the obvious resistance at 1,000.
As for the trendline, at best it's what John Murphy calls a tentative trendline in his book 'Technical Analysis of the Financial Markets'. (I'd call it a 'hopeful' trendline.) In order to be a valid trendline many would say that price needs to retrace at least 50% of the drop from the March peak. Others, like me, would say that price needs to break the March high before that trendline can be considered valid. The trendline as drawn clearly violates the rule above. It looks to me as though it was conveniently drawn to fit the author's fundamental thesis about where gold is headed. It reminds me of a class I once took about scientists and their paradigms. Many scientists get so caught up in proving their theses correct that they can't see conflicting data (anomalies) even when it's staring them right in the face.
My version of the chart & trendline is below. Gold could very well be head back over 1,000 but it certainly won't be clear sailing based on the chart.
The way I have read Trader Vic's method is slightly different from how I believe you have adopted your "trends," although what you drew is certainly consistent with his methodology. If one views long-term as months to years, intermediate as weeks to months, and short-term as days to weeks, I would view gold as being in a solid long-term up trend (based on a weekly or monthly chart), an intermediate down trend (based on the chart above) and a short-term up trend (again, based on the chart above). You chose moving averages to define these moves but I still find it useful to view such trends in three time frames. Your view of this would be interesting - especially a comment on why you chose moving averages to define these time frames vs. looking at the trend lines.
"draw a line from the lowest low, up and to the highest minor low point PRECEDING the highest high"
Perhaps I am missing something, but wouldn't you draw the line to "highest minor low point preceding the highest high" which would have been in March just before the 08 high? It seems you drew the line to a point after the highest high and not preceding it. Great article!
Carey,
You must be talking about the trend table that I use for the daily recaps. I just use moving averages there for simplicity -- it's easier than having to find & draw 3 trendlines every day. Also, it saves me the trouble of having to argue with folks about what's long, intermediate & short-term.
In this case, I'm simply talking about how to draw a trendline. I'd probably call the trendline that I drew on the gold chart the intermediate-term trendline.
Steve,
You forgot the part about "not going through other prices." I used the same time frame as the gold bull. In that case the lowest low is in September and the connecting low -- without going through other prices -- would have to be the late December lows.
If I was considering a slightly shorter time frame I'd draw a line from the late December low through the February intraday low.
Note that in either case we don't get the trendline that thye gold bull have created.
Good scientists set out to prove their theories are wrong (the null set), only when they fail to do that can they be proven right.
Great post!
That's very true about good scientists Jack. I think many good traders are also willing to examine the opposite side and to trade either long & short depending on what the market's telling them.
trendlines are all in the minds of those who see it..some technicians would agree w/ the article (of course there are resistances ahead)//remembet TA is an artform...(and sometimes B.S.)
Hi Mike,
Glad you mentioned Trader Vic's book. One of my favorites, and I also look to it as a simple and correct guide for reading charts and drawing trendlines, finding change of trend in chart patterns.
I see a lot of people drawing trendlines that go through prices, or seem to follow some other flight of fancy. I think anyone who is serious about using, discussing charts should read Vic's book! And there is much more wisdom to be found there, besides.