by ROBERT R. PRECHTER, JR.
A ny discipline that can be explained and quantifed is ultimately programmable, and the Elloitt Wave
Principle is no exception. In fact, many aspects probably could be computerized relatively easily.
However, in attempting to program "Elliott" in its entirety, we run into a problem, namely, the element
that is not easily programmable happens to be the most important and fundamental element of Elliott
analysis: wave counting. Wave counting precedes all other analysis under Elliott. If that aspect could be
programmed, it would be easy to program the rest of Elliott wave analysis, including channeling,
retracements and Fibonacci targeting.
The difficulty is that the standard in the Wave Principle is wave form, not time periodicity (as in cyclic
analysis), or even fixed time or price relationships within wave structures. Wave subdivisions of the same
degree often differ greatly in size and complexity, and can include variations such as extensions, diagonal
triangles and failures. Programming them would involve a tremendous number of "If ... then's"; and
instructing the computer to count waves in real time, considering all alternate possibilities, would require
a huge amount of statistical research that to date hasn't been done. The analyst would need to quantify
the probabilities of each structural shape occurring at each conceivable step in the progression of the
market. Such research could take half a lifetime.