Chart Pattern Statistics Book

heavenskrow

heavenskrow

Does anyone here know a good book that goes over the statistics for patterns in the market??
Such as the success/failure rate of a head and shoulders pattern, ascending triangle, bull flag, etc.
Thanks in advance!
 
heavenskrow said:
Such as the success/failure rate of a head and shoulders pattern, ascending triangle, bull flag, etc.
More...

The success rate is high enough to have given such patterns, that have passed the test of time, the status of 'classic' patterns. IMHO The failure rate is most often the result of the trader misreading the pattern or seeing what is not really there or forcing the price movement to look like a pattern that fits in with his bias or whatever. But you could easily enough go through old charts picking out 100 of each pattern and see for yourself what the success ratio is. And if you find the failure rate to be higher than expected, such an exercise might prove to have been worth your time!
 
Last edited:
One of the problems with patterns and charts is that most people look at a chart and they will say this looks like a '_____ pattern', basically assuming that a pattern actually exists. If one has not mathematically defined a pattern, no matter what it's components (price, volume, time, or any other input), they are basically making two assumptions that they cannot prove:

  1. That the pattern even exists
  2. That the pattern has been historically significant
 
trhudson said:
One of the problems with patterns and charts is that most people look at a chart and they will say this looks like a '_____ pattern', basically assuming that a pattern actually exists. If one has not mathematically defined a pattern, no matter what it's components (price, volume, time, or any other input), they are basically making two assumptions that they cannot prove:

  1. That the pattern even exists
  2. That the pattern has been historically significant
More...


Good points.

Toby/Tony Crabel wrote a book a while back with stats on outcomes based on simple patterns/phenomena. He's a legit trader now running a CTA based on his research.

Outside days, inside days that kind of stuff. None of this horse shit like trendlines, double/tripple/tops/bottoms and other stuff that is really only there in the eye of the beholder -and (as you suggest) difficult to quantify.

Even if one was to quantify such silliness, I imagine as soon as one has a definition, occurrences that meet the exact definition will be rarer than wings on a pig; thus useless.
 
trhudson said:
One of the problems with patterns and charts is that most people look at a chart and they will say this looks like a '_____ pattern', basically assuming that a pattern actually exists. If one has not mathematically defined a pattern, no matter what it's components (price, volume, time, or any other input), they are basically making two assumptions that they cannot prove:

  1. That the pattern even exists
  2. That the pattern has been historically significant
More...

What exactly do you mean by mathematically defined a pattern? How would you even go about doing that?
 
You must log in or register to reply here.
Back
Top