Strategy Central

[Walkingshadow]

Yes. Consistency and discipline are the key concepts, once you have a good feel for the strengths and limitations of each indicator you feel comfortable with. And as you know, they all have strengths and limitations.

If you can achieve even a 60% win rate and the average profit per winner is significantly greater than the average loss per loser, you will do very well over time. Simple mathematics.

I prefer AskResearch, but unfortunately they are more limited. Some things cannot be charted there, e.g. the dollar.

I also divide things into preliminary (early) indicators which are highly sensitive but not very specific, primary indicators that have lower sensitivity (or are slower) but higher specificity, and confirming indicators that are extremely specific but signal late. I use the first for a "heads-up", the second as a primary signal, and the third as a confidence measure. Sometimes I use these to adjust position size.

As for divergence, I think it is not so much how rapidly the stochastics (or other indicators) are downsloping that is important. Rather, the degree of discordance between the slop of the indicator and the slope of the price. You must examine both. Duration is also important, and corroboration from other sources (e.g., other technical indicators, other related indices, sentiment extremes developing, candles, limits of the regression channel, volumes, etc.).

I also like to be short and long at the same time (in different markets, that is). Helps keep me honest. It is one thing to say you are not concerned with which direction the market goes, quite another to keep bullish or bearish bias from affecting your decisions.

I realize the problems you point out. My answer is to develop a decision flow-chart with go/no-go decision points that indicate exactly when to go long or short using categorical criteria that are mutually exclusive. With that comes an automatic way to decide when you have been wrong, since the decision making process is repeated each day.

Exactly what inputs to use is something I have been trying to fine tune without getting to complex.


Who is this Mr. Burk? I am very interested in the Presidential Cycle, and have done extensive analysis myself on that. I would be very interested to see his analyses.

Thanks,

T