This is my last serious post for 2023. There will of course be the traditional Christmas edition, but otherwise, I will be back in January 2024. Before I go, I want to write about one of the most common chart crimes you can find in investment reports and on social media.
Back in the late '80s in my first job on Wall Street, there was a mysterious group of about four people in an old vault on the 15th floor with the door permanently propped open, with the walls covered in huge sheets of plotter paper. Curious, I asked someone who they were.
"Those are the technical analysts, a.k.a. the chartists. They sit in that windowless vault all day and painstakingly draw out two-line graphs shifting the y-axis scales and x-axis timeframes until they find something that looks as though it's correlated enough to predict the future, and then they print it in a report."
I said, "but we're a fundamental shop that subscribes to the Random Walk Hypothesis https://en.wikipedia.org/wiki/Random_walk_hypothesis so why do we have these folks at all? Doesn't it send the wrong message to our clients who expect us to be coldly rational and facts-based?"
The response was: "Every other house has a small group of technical analyst tea leaf readers for the same reason that newspapers still print daily horoscopes ... it's entertainment."
The ultimate excuse seems to be that every trader I've ever met says technical analysis of very short term trends *does* have some value to gauge whether sell-offs are over- or under-done https://en.wikipedia.org/wiki/Head_and_shoulders_(chart_pattern) , but with what appeared to be the relative veracity of three-to-five-day weather forecasts in comparison to those for a whole year in an almanac.
It is exactly the other way around: as a pure technically Driven trader, I Place my entry and Exit points based in Supply and demand points in the Price Chart instead of Hoping that one of the narratives will come true it Not. Fundamentals (at least on Time Frames of few months or lower) Are totally useless and just Entertainment. Trust me..
Thank goodness someone is calling this nonsense out. If you want to create these things, here is an easy way: start with a picture of the first time series, the one you intend to "forecast." Then photocopy the other time series, to an appropriate scale (which you can change later,) on a sheet of transparent film. (Occasionally, if the trends are in the opposite direction, invert one of the series.) Then line up the two series, one on top of the other. Shift the sheet of clear film horizontally until the two time series line up (that's the fictional "lag" between the two series, of course.) You are done: the series should look pretty much alike. You can make virtually any time series "forecast" any other time series. A Google search of the news, or an appropriate "theory" of some kind, "explains" the relationship and completes the package.
Ironically, that is pretty much what my Excel tool does. It’s so easy, yet apparently you can build entire businesses around that crap (not naming names here…)
Back in the late '80s in my first job on Wall Street, there was a mysterious group of about four people in an old vault on the 15th floor with the door permanently propped open, with the walls covered in huge sheets of plotter paper. Curious, I asked someone who they were.
"Those are the technical analysts, a.k.a. the chartists. They sit in that windowless vault all day and painstakingly draw out two-line graphs shifting the y-axis scales and x-axis timeframes until they find something that looks as though it's correlated enough to predict the future, and then they print it in a report."
I said, "but we're a fundamental shop that subscribes to the Random Walk Hypothesis https://en.wikipedia.org/wiki/Random_walk_hypothesis so why do we have these folks at all? Doesn't it send the wrong message to our clients who expect us to be coldly rational and facts-based?"
The response was: "Every other house has a small group of technical analyst tea leaf readers for the same reason that newspapers still print daily horoscopes ... it's entertainment."
The ultimate excuse seems to be that every trader I've ever met says technical analysis of very short term trends *does* have some value to gauge whether sell-offs are over- or under-done https://en.wikipedia.org/wiki/Head_and_shoulders_(chart_pattern) , but with what appeared to be the relative veracity of three-to-five-day weather forecasts in comparison to those for a whole year in an almanac.
Frohe Weihnachten.
It is exactly the other way around: as a pure technically Driven trader, I Place my entry and Exit points based in Supply and demand points in the Price Chart instead of Hoping that one of the narratives will come true it Not. Fundamentals (at least on Time Frames of few months or lower) Are totally useless and just Entertainment. Trust me..
Did you even read my comment? I said the only thing people seemed to agree technical analysis was good for was as short-term trading signals.
Thank goodness someone is calling this nonsense out. If you want to create these things, here is an easy way: start with a picture of the first time series, the one you intend to "forecast." Then photocopy the other time series, to an appropriate scale (which you can change later,) on a sheet of transparent film. (Occasionally, if the trends are in the opposite direction, invert one of the series.) Then line up the two series, one on top of the other. Shift the sheet of clear film horizontally until the two time series line up (that's the fictional "lag" between the two series, of course.) You are done: the series should look pretty much alike. You can make virtually any time series "forecast" any other time series. A Google search of the news, or an appropriate "theory" of some kind, "explains" the relationship and completes the package.
Ironically, that is pretty much what my Excel tool does. It’s so easy, yet apparently you can build entire businesses around that crap (not naming names here…)
what's worse, the charts or the truisms?
Like the one I repeatedly heard a few weeks ago, "nothing good ever happens below the 200-day moving average".
Yeah right, I hope those folks went to cash at the end of October.
It reminds me of that line in "Key Largo": "What's worse, a wise guy or a dumb bell?"
"Lies, damned lies, and cherry-picked time series charts.", as Oscar Wilde said (or was it Franco Modigliani when he won Strictly Come Dancing?)
Nice post BTW.
Wishing you a Merry Bah Humbug
Beautifully done...
Thanks for all the thought provoking writing you've shared thus year. Hope your break is enjoyable!