Storytelling and Technical Analysis

Humans can’t not tell stories: that’s the result of decades of research in cognitive science, psychology, neuroscience, and philosophy into our tendency to string together disparate events into coherent narratives. If you’re philosophically inclined, you know that David Hume got it right centuries ago: when we identify causation, what we are really doing is composing fictions to help us make sense of the buzzing world around us. But it would be a mistake to confuse our convenient fictions with the underlying nature of reality. The problem is that it’s very hard to know when we’ve hit upon a law of nature and when we’re just making things up. Consider this famous example of an experiment involving a split-brain patient who was presented with two images, each beyond the view of the eye associated with the other lobe:

“The man’s left hemisphere saw a chicken claw; his right saw a snow scene. Afterward, the man chose the most appropriate matches from an array of pictures visible to both hemispheres. He chose a chicken to go with the claw, and a shovel to go with the snow. So far, so good. But then Dr. Gazzaniga asked him why he chose those items — and struck gold. The man had a ready answer for one choice: The chicken goes with the claw. His left hemisphere had seen the claw, after all. Yet it had not seen the picture of the snow, only the shovel. Looking down at the picture of the shovel, the man said, “And you need a shovel to clean out the chicken shed.” The left hemisphere was just concocting an explanation, Dr. Gazzaniga said. In studies in the 1980s and ’90s, he and others showed that the pattern was consistent: The left hemisphere takes what information it has and delivers a coherent tale to conscious awareness. It happens continually in daily life, and most everyone has caught himself or herself in the act — overhearing a fragment of gossip, for instance, and filling in the blanks with assumptions.” Decoding the Brain’s Cacophonysee this recent post from Barry Ritholtz for more of this sort of thing, or go here for the heavy-duty stuff.

What goes for apples falling from trees also goes for AAPL falling from recent highs: we can either be vigilant about excising spurious causal claims from our investing, or be ready to lose money for the sake of good feelings.

For instance, many traders rely heavily on subjective technical analysis to inform trading decisions, a practice which might uncharitably be called the systematic and deliberate confusion of storytelling with truth-telling. The fact that I can make a plausible case for a future stock price based on some lines on a chart doesn’t mean anything by itself. If there were quantifiable, repeatable, testable methods for checking the validity of a given chart pattern or rule set, then we could use reliable performance metrics to identify which circumstances were likely to be profitable.

Objective technical analysis can meet this test, potentially (e.g. rule sets relying on moving averages, quantified breadth or volume indicators, autocorrelation with a paired asset, etc.), but, by definition, subjective technical analysis is in the eye of the beholder.

Objective technical analysis tells stories in the form of buy and sell signals based on a concrete rule set. Those stories are precise, and can be confirmed as true or denied based on the evidence. Subjective technical analysis, in contrast, tells vague stories that cannot be proven or disproven: you might see a head and shoulders where I do not, or an ascending triangle or a cup and handle or…, or begin your Fibonacci retracement at a different point, or surf a different set of Elliott waves…

Academic researchers have made hundreds of attempts in published papers to test the validity of technical analysis patterns and rule sets. The unfortunate conclusion is that, in most cases, we are confusing fiction with reality (123) – and the exceptions that stand out, such as trend-following methods (4), are again entirely objective and rules-based rather than intuitive or subjective.

Stories that can be passed perfectly from one hearer to the next (objective and quantifiable) seem to have more predictive power than entertaining but elusive yarns like you might hear around a campfire. This presents a challenging resolution for 2012. Life is short, and capital is limited. Why waste time and energy on unverifiable prognostications when we could be pursuing methods with a rational, quantifiable chance of success? Since we can’t not tell stories about the events we see, I’d rather construct fictions that I can later test for accuracy.

This article was originally published at TheStreet’s Options Profits.

Tags: ,

2 Comments For This Post

  1. Market Rewind Says:

    Every picture tells a story; don’t it?

  2. Jared Woodard Says:

    Sure, but many stories don’t warrant tradeable inferences.

4 Trackbacks For This Post

  1. Sunday links: surprising scale | Abnormal Returns Says:

    [...] Technical analysis is done in two very different ways.  (Condor Options) [...]

  2. Sunday links: surprising scale Says:

    [...] Technical analysis is done in two very different ways.  (Condor Options) [...]

  3. FT Alphaville » Further reading Says:

    [...] Of subjectivity and [...]

  4. Quote of the Month « Systematic Relative Strength Says:

    [...] Jared Woodard, Condor Options: Life is short, and capital is limited. Why waste time and energy on unverifiable prognostications when we could be pursuing methods with a rational, quantifiable chance of success? [...]

Leave a Reply


Jared Woodard specializes in trading volatility as an asset class. With over a decade of experience trading options and other volatility products ... Read More


Open All | Close All