A Different Way to Use Traditional Technical Indicators

The above chart depicts a running cumulative total of buy signals minus sell signals for all NYSE stocks for the Wilder Parabolic SAR indicator (raw data from the excellent Stock Charts site).  It has been a useful overbought/oversold measure for the broad stock indexes.  Note how we've bounced off an oversold level recently.

Interestingly, going back to June, 2014 when I began collecting these data, daily buy signals have correlated with daily sell signals by only -.35.  If we just perform a median split of the buy signals, we find that when we have a high number of buy signals, the next five days in SPY average a gain of only +.01%.  When we have had a low number of buy signals, the next five days in SPY have averaged a gain of +.31%.  

I compute similar stats for the Bollinger Band; CCI; ADX; and RSI indicators and find value in tracking buy signals and sell signals, as well as cumulative running totals of their differences.  Treating buy signals and sell signals as distinct variables ends up offering unique value--it's a very different way of making use of traditional technical indicators.

Further Reading:  Tracking Market Strength
.