
You can see clearly that new highs have consistently expanded during the recent market rally. Interestingly, going back to the start of 2004 (N = 908 trading days) when new highs have exceeded new lows by more than 20 issues (N = 30), the next ten days in SPY have averaged a loss of -.56% (14 up, 16 down). That is considerably weaker than the average 10 day gain in SPY for the remainder of the sample of .32%.
This suggests that, by the time a solid majority of issues are making new short-term highs, the buyers have generally gotten their business done and there has been no bullish edge in the near term.