LottaBits is the culmination of more than 35 years of thrashing around. Back in the late 60's I returned from a tour in an exotic and dangerous Asian locale. My idea back then was to build a trimaran sailboat, sail the world and write novels. Only one small problem kept me beach-bound. My savings account was too small to do the whole job. Then, one day while traveling, dreaming and writing I stumbled on a small brokerage office in the Northern California town of Susanville and was introduced to commodities trading. Here was the way to multiply my savings to a level needed to build and sail the trimaran. Or so I thought. I was advised to go to San Francisco, where the real traders lurked and learn their trade first-hand.
One thing led to another. I went to San Francisco, learned to trade, traded pork bellies and cocoa, lost my savings, got a job, got married, grew up a little ... but never lost the bug for trading. Then in the late 70's I discovered computers. Wow, what analytical power that would provide, what ease. But the necessity to key in prices by hand from the Wall Street Journal, $60/trade commissions and other impediments kept the effort theoretical rather than practical.
I never stopped studying, began to observe patterns in stocks and concluded that they had meaning. Others had, of course, observed patterns, and I studied these too. But being an inventor at heart, and concluding that all commonly known methodologies would be immediately discounted by the market spurred me on to develop my own method. At first I thought I could figure out a trading system that would work for every stock all the time. Under rigorous testing using 20 years of daily data on 7,000 stocks, my system failed that test. Then I tried a method that "should have worked" for some stocks all of the time -- and lost a lot of money proving again that the market changes too fast for that to be true. But after additional years of effort, I realized that my methods could work for some stocks some of the time, that this was sufficient, and refined my methods with this objective in mind. But which stocks were "some stocks", and when was "some of the time"?
The breakthrough came when I realized that my methodology would yield two important components I'd never had before. The first component was a way to combine the information for all stocks on the market in such a way as to determine what conditions were tradable at a particular time or state of the market. Then, secondly, based on this information, it was now possible to determine when a stock was or was not tradable, and in so doing to pick a list of stocks having signals suitable for trading on a given day. This system, using all the stocks in the market, or on a large list, provided good signals and consistently good performance. This system is LottaBits. It doesn't work for any particular stock all the time. It doesn't work for all stocks most of the time. But it does know when a given stock is tradable, and that's fine with me, because the overall result is very good indeed.