Technicals, from Victor Niederhoffer

April 19, 2012 |

I claim that all well-known technical chart patterns that are capable of quantification have an opposite predictive power to that posited in the books on technical analysis from Magee to the updates at the top the list is the triple and quadruple top (there's one right now).

Philip McDonnell added:

One is reminded of the paper by Andrew Lo which tested something like 65,000 technical patterns. It found that the double top and double bottom patterns worked and were statistically significant even after adjusting for the serious amount of data mining.

Andrew W. Lo writes:

I believe [Vic's claim]. No pattern can truly be 100% consistent, otherwise it would be exploited to its limit… The problem is when it seems to works 52% of the time.

Rocky Humbert writes:

My former partner was fond of saying, "There are double tops, and triple tops. But no quadruple tops." Importantly, he is my "former" partner….

Anatoly Veltman writes:

In the upside down, they say that only cats have four legs.

Craig Mee writes:

In my opinion, technicals have everything to do with quantification of patterns and allowing for the edges to play out just like any area of business you're trying to gain an edge. The larger the patterns IE 1 hr to 4hr, D to W, and the stronger the boxes ticked in one's strength meter, the more risk, and a little more risk/reward is expected on the trade. One might say all head and shoulders or double or triple tops are not equal.

The risk appears to lie in the fact that most technical traders move off the course, trade too many patterns from trend line breaks to wedges, etc, over too many markets over too many time frames, and thus the edge they're trying to work with or look for is blunted.

Comments

4 Comments so far

Enn on April 20, 2012 6:35 am
People do quantify these things….
http://thepatternsite.com/BestPatterns.html

Failure rates
http://thepatternsite.com/FailureRates.html

Interesting study on volume confirmation too
http://thepatternsite.com/VolumeStudy.html
Terry Lesniak on April 20, 2012 10:17 am
Is there really any predictive power from technicals or is it really chasing the last bit of inertia that happens to be moving in the direction you desire(if in fact you are a directional trader)?

Is there anyone out there really predicting anything? something beyond some ephemeral edge?
steve on April 21, 2012 11:16 am
All I will say about technicals is that they are a tool that is in your belt. Just as there is no magical tool there is no magical system. Granville had his GM and his transport indicator. Prechter had his Elliott Wave. Marty Zweig, Stan Weinstein, Bob Farrell, Carter Worth, Bollinger Bands They come and they go. The easiest way to make money using technicals is look for a stock that is way down in price and makes a big breakout on big volume Even sweeter if it is a island formation Then start buying

I used to say everything works and nothing works. And never confuse brains with a bull market. A rising tide lifts all boats and a receding tide leaves them on sand bars.

I am out of cliche’s
steve on April 21, 2012 11:22 am
I will state one further thing. Technical analysis is like the Dismal science. the analysts always speak in probabilities. Not absolutes they are the true politicians of investment analysis.

Here is an interesting primer from Bob Farrell. http://stockcharts.com/school/doku.php?id=chart_school:trading_strategies:bob_farrell_10_rules

Recent Posts

List of Authors

Apr

19