JLS - Thanks for your reply. I am familiar with the method and understand *in spades* the patience and discipline that are required to be successful trading a mechanical model. It will be wrong frequently, but if the statistics say that *over time* it is a winner, then you have to be in all the time to allow the statistics to work in your favor and to realize the gains. Re: the scientific method, it is rare that any experimental outcome is ever widely accepted on the basis of a single experiment, and normally requires independent verification from several other experimenters to be given validity. The "cold fusion" case comes to mind, and as in the case of global warming, you always have to be suspicious of the motivations and/or agendas of the experimenters. Many research scientists will endeavor to prove anything you want if you will agree to continue funding their research. If global warming is indeed for real, then it has been underway for a long time. The last ice age came to an end ~ 10,000 years ago, far in advance of primitive mankind making any sort of measurable impact on the environment, so something much larger, and moreover, a totally *natural* phenomena must have been at play. To say that A caused B simply because B followed A is just bad science.
Re: market behavior and trading systems, the fortunes of trading systems ebb and flow as the character of the market changes over time, and the system or algorithm that worked well at times in the past may not work well at all today. There was a marked shift in market behavior in 2000, which was perhaps due to the widespread proliferation of online trading (i.e. a massive number of relatively inexperienced traders learning the ropes), and the historic volatility levels made gains of over 80,000% possible, *if* you had gotten every single day's call correct. That level of volatility moderated by 2003, but there was another distinct shift in the market's trading behavior in July last year with the elimination of the Uptick Rule. It took a few months to recognize that the existing models were no longer working as well as previously and needed "tweaking", but the observed effect was very real and was widely reported in the quantitative trading world.
Re: EW and/or Fibonacci ratios, I for one do not believe that the market is preordained to behave in any particular way as those methods suggest, but if 6 out of 10 traders believe in Fibonnaci ratios, use them to set S/R levels, and actually base their trades on those levels, then there will be a statistical clustering of trades at that level, proportionate to the number of traders using them, which might be sufficient frequently enough to turn the overall market at that level. Hence, for those using the "method", it has a tendency to be self-fulfilling and therefore self-reinforcing.
