There's been a bit of back and forth tweeting about the value of divergences among market indicators. Jason Goepfert of the excellent SentimenTrader service rightly notes that divergences are not great predictive tools. Indeed, divergences on a chart are merely observations; their predictive value needs to be established through testing.
Still, I find divergences to be useful observations as a kind of heads up, a yellow caution light. If the broad market indexes are making new highs or new lows and a substantial proportion of shares are not participating, I want to be open to the hypothesis that the move reflects a handful of highly weighted shares within the indexes, not the broad market. This keeps me open to the possibility of reversal, combating any overconfidence bias I may have with respect to the recent market move.
Yesterday gave us a perfect example of the cautionary value of divergences. When we made fresh lows in SPX in the morning, my measure of selling programs showed considerable hitting of bids in my basket of highly liquid large cap names (top chart). These stocks were not only downticking, but doing so at the same moment, revealing an elevated level of basket executions of sell orders.
Despite this selling activity, the broad list of stocks (all listed U.S. shares) was showing reduced selling pressure (middle chart) as we made new lows. My measure of upticks vs. downticks across all stocks was showing considerably less selling pressure at yesterday's lows relative to the prior recent downturns. That was an indication that intense selling pressure was occurring in a smaller group of large cap names, but not across the broad market. Indeed, my cumulative measure of upticks vs. downticks among all listed shares was hitting new highs yesterday! (All data above from e-Signal).
On the bottom chart (data from Barchart), we can see how this divergence played out with respect to stocks making fresh three-month lows. Although we touched price lows intraday yesterday, the number of shares listed on all exchanges did not expand the number of new lows made from the last few days. Again, there was intense selling pressure in the visible large cap group, but a lack of selling impact across the broad market.
Keeping tabs on market strength and divergences, I find, helps keep me cognitively flexible. For someone like myself who was short the market going into the day, that flexibility proved to be quite useful. It is very easy to get caught up in the stock market and lose sight that it's actually a market of stocks. Looking at what the components are doing can be very useful in anticipating the movement of the broader market.
Further Reading: Views on Breadth