Monday, January 11, 2010

Buying Strength vs Buying Weakness

on November 6, 2009, I noticed that there were 9 S&P 500 components with RSI(14) readings above 70 (generally considered the 'overbought' line). there were also 9 components with RSI(14) readings below 30 (generally considered 'oversold').

the overboughts were: EL, RX, AMZN, MCD, HRS, SRCL, ADP, PAYX, and NSC.

the oversolds were: NTRS, PTV, BSX, AYE, FSLR, CVS, JAVA, TAP,and APOL.

on finviz.com I created two equally weighted portfolios to track performances of each group. one portfolio was long all the overboughts, while the other was long the oversolds. I wanted to see how each group fared over a mid-term horizon. it's been a couple months now, so let's see how we stand.

first, it is notable that all 18 stocks are showing a profit from the theoretical buy price on 11/6, which, if nothing else, is a testament to the strength of the broad market over the period in question.

the overbought basket is up 3.95%.

the oversold basket is up 8.84%.

that is a substantial out-performance by the oversold issues and at first thought the results seem to reinforce my personal preference of buying weakness, as opposed to the strategy of buying strength popular in the trend-following set. (though, to be fair, I have never heard of a trend-following strategy as simplistic as buying all the >70 RSI issues on the market and holding them for a couple months.)

but if we examine the results a little more closely, we see that the oversold basket, so far at least, has not drastically out-performed the S&P 500 as a whole, as the benchmark advanced 7.53% over the same period.

but it gets worse. remember I said the two baskets were equally weighted, meaning we placed an equal dollar amount of each stock into the baskets. so to be really fair, shouldn't we be comparing these baskets against the equal weight version of the S&P?

well, the Equal Weight S&P 500 (SPXEW) gained 10% over the period in question:



I am not sure whether we should draw any conclusions from this little study, but I thought it was kinda interesting...

oh, and by the way, there are currently 79 S&P stocks with RSI(14) above 70. how many stocks have RSIs below 30? 0 (Zero). if you will pardon my french, that's fucking sick. Waterloo for the bears indeed. or, just maybe, it's Operation Barbarossa. say, mid-October, 1941.