Hi North.....yes, you understood correctly...it is very rare indeed that it does not make the opposite "100" level, once it has passed through zero. That doesn't mean it never happens...just that it's unusual. Like anything in the market, you don't want to bet your firstborn on it.
Everyone plays with parameter settings...I do too. I think the CCI(20) is the best compromise, but someone could curve-fit another time period for a specific era that might be better.
Another of the advantages (it can also be a disadvantage in certain circumstances), is that the CCI seems to be somewhat "early"...I'm not proposing a set of rules..but perhaps shorting when the CCI crossed +100 might be too early..perhaps waiting to +150,+175 might be a better place to commence shorting..or if the CCI never gets that high, you could almost always get a reasonable short (a few points, or more) if it barely got above +100, and you shorted when it crossed below +100. In a few cases you might have to endure a little sideways action for a few days, but generally once it heads to the other side, it makes it.
The proof might be in just studying how the CCI levels respond to price. It is not a panacea, and I don't want to suggest that it is..it just seems better to me as a "VTO type" strategy than the RSI (perhaps).
Market Data 
Markets 
AI
