Oldtrader,
Thanks. As far as backtesting goes, I would throw that out along with the TA. If any backtesting is worth doing, it should be done manually in my opinion -- preferably in a fashion where you can't see the entire graph.
I like to play back the previous week's activity and just sketch down my observations about what happens, what patterns occur on a frequent basis -- what candle patterns have higher statistical outcomes, etc etc.
I hate these systems that say, "When you get an MA crossover, you buy, sell, etc." After you put it into a computer, the computer spits out this number that has no relation to your emotions or mental psych while trading.
What boggles my mind most is that people would rather trust a computer to come up with a system analysis rather than use their own brain to interpret price data as it happens.
It would be like me going up to someone and saying, "Hey, I'll swap your Pentium IV computer for an 8086 machine -- it has great graphics and the candles are always green!" Well, the human brain is the best computer we have -- and it is very good at adapting fuzzy-logic to certain situations.
If you were going to catch a football, you wouldn't stop, pick up a piece of paper and calculate the calculus required to control vectors of your muscle motions. You look at the ball, estimate where it will be at point X and catch it. Why shouldn't the same philosophy apply to trading?
The reason I got fed-up so quickly with TA is because I was doing a lot of investigation of moving averages and then I'd look at the charts -- lots of them -- and suddenly my brain said, "There is a pattern here, and you can trade it with high success, why are you messing with wavelets?" That's a good question! Why do I need a momentum indicator when the momentum is on my tick charts? The faster the ticks are created, the more momentum -- and if they are going up, I should be long. How does it NOT get any simpler?
If I go long and it doesn't act the way I thought it was, it is a scratch trade and I go on to the next signal -- but all this TA BS is just going to get a trader thinking about things other than the most important -- price, volume and direction.
Bollinger Bands are great for an academic study in price movements, but do I want to make money or do research on the fly with real money? No! I want to make money and the only way I ever consistently make money on simulated trading is by sticking to a simple system, getting with the market and accepting losses when I am wrong. I do that, and I make money -- every damn time.
TA might prove better for longer trading techniques, where you have time to think about your trade before you make it -- but on the fly daytrading? You don't have time to look at anything else except the ticker, chart and price movements -- spend any more time on anything else and you're cutting into your edge.
aphie