patterns
Patterns are probabilities
The rule
A chart pattern gives odds, never a promise. Like a bus that is usually on time, every pattern fails a real share of the time. Plan for that failure.
Where it flips
Knowing a pattern 'usually works' tempts you to bet too big on the good ones and make no plan for the bad ones. The fix is to read the reliability honestly (Module 026), assume this one may be in the failing share, and let the stop, not hope, set your downside.
Bulkowski actually counted. He tallied how often a head-and-shoulders or a cup-and-handle hit its target, and how often it broke the wrong way. The lesson is not that patterns are useless. Each one has a failure rate you can look up, and it is never zero. A pattern is a bet with a small edge, not a sure thing. So you size the trade and place the stop for the times it does not work.
A worked example
A stock forms a clean double bottom, the kind that works more often than not. A reader treats it as certain, buys full size with no stop, and gets caught when this one fails and falls. The pattern's edge was real. Ignoring its failure rate was the mistake. [illustrative]
How to spot it
- ·a pattern treated as a sure thing rather than a set of odds
- ·no stop placed because the setup 'cannot fail'
- ·the pattern's usual failure rate never looked up or admitted
Thomas Bulkowski · Encyclopedia of Chart Patterns