In chessNwine’s nightly market recap, he mentioned the $SPY moving average cross where the 20 day has fallen beneath the 50 day average. Like most moving average crosses where the shorter average crosses beneath the longer average, popular lore considers these crosses to be bearish. Chess noted that I have previously written on this topic, and the research showed that the 20 day, 50 day cross is not a solidly bearish event.
Surprisingly, it has been over 2 years since I looked at these crosses. Man how time flies. Tonight I re-ran tests using $SPY, buying the close when the 20 day average crosses beneath the 50 day average. There have been 8 more of these crosses since I last tested them.
The results are below…
The median of the winning trades is 5.70% and the median of the losing trades is -5.72%.
I have included the $SPY buy-n-hold calculation. To arrive at this calculation I have simply chopped all $SPY history into 100 day segments and then averaged those segments.
The bottom line is that a 20, 50 day cross, at least in terms of $SPY, is really nothing to be worried about. In terms of buy and sell signals, on average $SPY doesn’t usually trade significantly lower after the cross. It remains a better buy signal than sell signal, over the course of almost 20 years of $SPY history, and tracks very similarly to the average $SPY buy-n-hold performance.
