Augie, thank you so much for the thoughtful response. There's a lot to mull over here, I spend some time on it. Here are a few intial thoughts.
Price trends: My saying "prices tend to trend" may have been a bit too simplistic. What I mean is this; daily volatility is reasonably "contained". For example, a chart of the $compq doesn't look like an electrocardiogram, with prices gaining 50% one day, losing 15% the next, etc. Prices trend, HOWEVER, it's impossible to tell what the dominant trend is at any given time. It kills me when I hear someone angst over what
Ma they should use.
It kills me even more when I read about folks backtesting different moving averages. It doesn't really matter what worked best in the past because down the road every moving average is going to work fine some times and terribly
others. I totally agree with you and LG, I use multiple timeframes in my timing models.
Breadth trends: Your points are well taken. I wish Stockcharts would create symbols for total issues traded for the major indices so the breadth indicators can be normalized. I asked Chip if he could do this and he never responded.
Indicators like the BPI, the percentage of stocks over __MAs, and the Summation Index are great breadth trend indicators. I prefer to use the A/D and AV/DV data to single breadth momentum as opposed to trends.
"'Sentiment Thrusts' may also be bullish IT/LT."
I agree, that's why it's important to wait for sentiment to reverse. A lot of folks on these boards like to say "look how low VIX is...the market's gonna tank". Absolute levels are one thing, relative levels are another, and reversals from relatively high/low levels are still another. I tend to stick with the later.
Well, I continue to read and learn. Thanks again for the excellent insight and ideas.
Market Data 
Markets 
AI
