Thank you for posting this. One more of my mythical heroes deconstructed. Your thread you referenced reminded me of a recent book on gambling scams, whereby they stated that the majority of poker tournaments always seem to have the same group of winners (who are coincidentally part of the contest organizers) rise to the top table, even though the probability of that is close to nill considering the luck portion of poker.
Many people get their faith of an afterlife from a 2000 year old book. If the discussion is still relevant, then it should be revisited. Williams %R is still on every major trading platform. If it should not be there, then the elite traders should demonstrate why. The posts that we make here can be googled, did you know that? We here at ET are considered the vanguard of traders, believe it or not. So, if Larry Williams needs to be exposed, or validated, then we are the ones to do it. I do not use %R, as I understand that it is some sort reverse stochastics, but I do use Larry's risk allocation model when I trade futures, which would have been worth the price of the book. Best Regards Oddi
I bought his book "Long term Secrets for Short Term trading". This book contains methods which one can integrate into his or her own style of trading. Larry and Jake Bernstein are friends and reading this book I can see why. They both offer readers useful tools which may not be viable as a stand-alone system but can be integrated into one's own standard operating procedure.
Itâs not necessary to read this or any book to find "useful tools" that in and of themselves do not make a trading system. The steps require a bit of programming but conceptually it's not that difficult: 1) Acquire a Genetic Algorithm engine like PI-Blue or similar. 2) Get 20 or 30 years of price history. 3) Create a concept library of the base elements of technical trading. There are perhaps three dozen: A + B, A - B, A / B, Average ( Ax...Ay), Stdev(), MACD, and so on. 4) Create a rule function to generate rules applied to the concepts: If a >b, if <> max(Ax...Ay) 5) Create a time window function to vary how the concepts and rules are applied in term of the number of bars to use. 6) Set 3), 4) and 5) above to interact randomly to generate many system tools. This will be a combination of concept, a rule and a time window. 7) Turn the GA program loose on the results of these system tools, with the ability to mutate and evolve them. 8) Wait about 3 to 6 weeks, depending on the speed of the PC, while the GA locates the best candidate system tools and evolves them based on consistency and profitability. In the end you will have dozens of high quality system tools, many never published before. Refine and combine as you see fit to create an actual trading system. You'll never need to read a trading guru book again.