Crazy Accurate Election Cycle Stock Market Indicator

I haven't had much interest in calendar trades since the Stock Trader's Almanac got popular because, as we saw yesterday, although some anomalies remain even after becoming widely known, the question to ask is:
"Are they exploitable?"

The only reason to bother posting this is that the author, Steve Sjuggerud, is one of the highest ranked forecasters according to CXO Advisory, dramatically outpacing such pontificators as Doug Kass and Comstock Partners .

From Daily Wealth:

"I've hero-worshipped the presidential cycle," investment legend Jeremy Grantham told Bloomberg Businessweek.

Grantham, you may recall, made what I call "The Most Accurate Investment Forecast Ever." (If you don't recall, I urge you to go back and take a look at that story. It's a quick read.) 

And the "presidential cycle" – as Grantham calls it – has been a surprisingly accurate predictor of future stock market returns...

Grantham told Bloomberg: "Going back to 1932, if you take the first and second year [of a presidential term] together, they've had no real return in the market." 

When I first heard of this idea of a presidential cycle for stock market returns, I thought it sounded absolutely ridiculous... But then I crunched the numbers – and it turned out to be way more accurate than I ever imagined. Take a look: 

Year	Return*	% of losing years
Year 1	-0.7%	50%
Year 2	-3.3%	50%
Year 3	26.2%	5%
Year 4	6.9%	20%
* Not including dividends

The first two years have no return. Then the numbers get silly...

All of the return [from the presidential cycle] has been compressed into a gigantic Year Three [return] and a respectable Year Four [return].

It turns out, since 1932, the return during the third year of a presidential term was negative only once... and it was a loss of less than 1%. If you add in dividends, we haven't had a single negative Year Three since 1932!...MORE

HT: Mebane Faber who has some additional insight.