The January Barometer, Another Take

I know it's a bit past-due, but I wanted to offer another take on the so-called January Barometer ("as goes January, so goes the year"). Others have already weighed in on this subject, quite skeptically overall, and yet I'm not so sure that this skepticism is warranted.

From data in The Stock Trader's Almanac, I put together the following:

Since 1950, there have been 40 occasions when the S&P 500 has been up in January and 24 times when the S&P 500 has finished down in January. For the 40 years when January has been up, the S&P 500 finished the year higher 36 times and was down for the year in just three occasions (with one year finishing near exactly flat, excluded). For the 24 years when January has been down, the S&P 500 has finished those years higher in just 11 occasions, with 13 of the 24 years resulting in down years. The key column above is on the far right, "% UP," listing the percentage of up years. Overall, for the entire 1950-2013 period, the S&P 500 has been up in 47 years and down in 16 (again, excluding the one flat year) for a 73% up percentage. Compare this 73% base line figure to 90% up years for the S&P 500 when January has been up and just 46% up years for the S&P 500 when January has been down. Interesting.

The following is from Ned Davis Research:
  

In the period 1928-2012, there have been 54 up Januarys and 33 down. What has happened after these Januarys, for February-December, thus excluding any effect January may be having on results for the entire year? For up Januarys, the ensuing February-December period has returned 8.3% on average or a median return of 10.9% for a 78% up hit rate. Whereas for down Januarys, the February-December period returned just 2.8% on average and an anemic 2.3% median return for a 58% up hit rate or percentage.

Based on the data above, I wouldn't be so quick to dismiss the January Barometer. It appears there might be something to it.