Everyone has their favorite TA methods. Personally, I use BB's, volume and MACD histogram to come up with trade candidates. I never got into Stochastics, OBV, Fibonacci, etc. Some people swear by them, others don't.
The tightening of the bands is another useful application to see volatility constricting. Look at a 5 year, weekly VIX chart. It tightened tremendously in the summer of '98 and then, an explosion of volatility leading to readings above 50. Well what do you know, the same thing happened recently, a sharp tightening of the bands, then an explosive move upwards, with historic readings in the 50 range. I don't view any indicator as the holy grail, but simply as a useful tool.
My method is to sell short at X% above the upperband, covering as it re-enters the band (within a few days max). Using MACD histogram and volume to weed out the powerful uptrends likely to continue. EK recently is quite an extreme example of this. Shorting something like YHOO in the last week is suicide. 52 weeks highs in MACD histogram and volume. I don't step in front of freight trains.
There are other methods of determining when a stock has reached an extreme, then deciding to go the other way. For example, there is some finite limit that a stock can reach over it's 10 dma (say for NYSE stocks). In GE's history, when reaching X% over it's 10 dma, the odds of reversing become very high. It might not happen very often for a particular security. Most stocks I trade only exhibit these characteristics a few times a year.
Even 200 dma. The NASDAQ in 2000 was more overbought than either 1929 or the Nikkei in 1990 using this criteria.
My whole trading philosophy is based on the premise that stocks don't go in one direction forever. When they reach a certain extreme, the odds of reversing become much higher. Measuring that extreme then going the other way is what I've made money from.
