Why do you believe its better to use the distance of ATR(20) from 200MA, instead of something like absolute mean deviation(also called average deviation) from the 200MA?The reason I ask is because ATR being just a measure of daily price changes doesn't tell you much about dispersion, only the range of the last 20 days. The avg deviation from the last 200 days will quickly tell how how the current price extreme compares to an historical context
I don't think it is better, however it is much easier for me to calculate with historical data. If I had charting software that gave me the average deviation then I would definitely look at that, and I'd be interested if there is a better indicator or method for measuring this. Another point is that this only applies after a lengthy trend (at least 2-3 months and a good % move for the market in question - the longer and bigger, the better). You can get 'overbought' readings fairly early (1-2 months) in a new trend after a narrow trading range, because the volatility was very low and the MA is flat.
Not the news but reaction to the news -- poor TXN performance on positive guidance bodes ill for semis