Thursday, May 10, 2018

What We Can Learn From Unique Breadth Data

Every day, I update dozens of spreadsheets with market statistics, many of which are not commonly found.  My experience is that unique data contain the most durable edges.  

From the Barchart site, I track the number of stocks across all exchanges that make fresh one-month highs and one-month lows.  Over the past year, when the number of stocks making new highs has been in the top half of its distribution, the next three days in SPY have averaged a gain of .22% versus .06% for the remaining occasions.  When the number of stocks making new lows has been in the bottom half of its distribution (few stocks making fresh monthly lows), the next three days in SPY have averaged a gain of +.23% versus .05% for the other occasions.

From the Index Indicators site, I track the percentage of SPX stocks closing above their five-day moving averages.  Over the past year, when that percentage has been above 70, the next five days in SPY have averaged a gain of .46%, with 79% of occasions profitable.  When the percentage has been below 30, the next five days in SPY have averaged a loss of -.13%.

When we see broad market strength, has that led to further strength (momentum), or has it led to weakness (reversal)?  Knowing the regime we're in can help us frame worthwhile hypotheses about market behavior.  Those become good trades when we see short-term flows lining up with those ideas.  

.