Buy in Fall, Sell in Spring ... a continuation of September returns

We were looking at the monthly returns, noting that (usually) September is the worstest month.
That suggest a strategy, like ...

>BUY in Fall, SELL in Spring. Right?
We'll see.

Here's what we'll do:

1. Pick a stock and download ten years of monthly closing prices.
2. Decide when we'll BUY and when we'll SELL ... but always at the closing price for the month.
3. Vary the BUY-month and SELL-month to see what the "best" strategy is.

>But the past is no indication of what might happen in the future! You should know that!
Uh ... yes, but I also have a secret tool.

>Yeah, sure, but there are hundreds of combinations. Am I supposed to ...?

If you're too lazy, you can click a button:

>Where's the spreadsheet?
Just click on the picture to download:

>What's the spreadsheet saying?
That the "best" strategy is to BUY each October (that's month 10) and SELL at the end of June.
That's for Microsoft, of course ... but you'd get a total return of 332.9% instead of the buy-and-hold return of a measly 256.2%.

>And that's guaranteeed?
Wait till I check my secret tool.

Sell when Congress is in Session !

There are other correlations between stock market returns and periodic events (like sunshine effects).
They include Super Bowl, sunspot cycles, astrology ...

>Yeah, so?
Thanks to a note from Ron Mc, I found this interesting (and recent: March, 2006) paper entitled: Congress and the Stock Market which ...

>Don't tell me! The U.S. Congress depresses the market, right?

As Will Rogers said:
      This country has come to feel the same when Congress is in session as we do when a baby gets hold of a hammer.
      It's just a question of how much damage he can do with it before we take it away from him.

>So how does Congress compare to September?
In the paper it notes that if $1.00 were invested in the DOW, in January of 1897, and withdrawn when Congress was "in" session ...

>Huh?
If you invest in the DOW when Congress is in session, and in cash when it's out of session (noting Will Rogers' comment) , then by the year 2000 you'd have a portfolio worth $216.10 and ...

>And you'd have lost money if you did the opposite, eh?
You'd end up with $2.00