You may have heard about the presidential cycle in stocks.

Or maybe you haven't. If not, here it is in a nutshell: Stocks do their best in the third year of the four-year presidential term--that is, in the year preceding the next presidential election year.

Since 2007 is the third year of the current presidential cycle--the next election will be in 2008--let's see whether there's any truth to the presidential cycle, or if it is just an urban myth.

Various studies have been done on the phenomenon, covering different time periods and using different indexes as proxies for "the market." All of the studies are in agreement. The presidential cycle is not a myth. Stocks, in fact, have historically done their best in the third year of the election cycle. They have been doing so for decades, it doesn't matter what index you look at, and the data is not even close.

In fact, the data supports recurring trends for each year of the four-year presidential election cycle, with the third year consistently being the best.

One study used the S&P 500 to represent the market and the time period from 1952 to 2003. The results were that the average annual total stock market gain has been about 6 percent in year one of the presidential cycle (that is, the first post-election year), 8 percent in year two, 23 percent in year three (the pre-election year), and 11 percent in year four. Returns from the last completed cycle just about matched those numbers. Year three of that cycle (2003) produced a gain of 26 percent and year four (2004) produced 9 percent.

Another study covered the years 1889 through 2005, also using the S&P 500 (and its predecessors) as the proxy for the market. Its conclusions were that returns were about 3 percent in year one, 3 percent in year two, 11 percent in year three, and 8 percent in year four.

The same study also looked at the data from another angle, measuring the percentage of years that the market was up for each year. The result: The market was up in 57 percent of year one's, 55 percent of year two's, 79 percent of year three's, and 73 percent of year four's.

Other studies support the strong-year-three indicator: Since 1945, the S&P has gained an average of 18 percent in the third years of election cycles, compared with an average of 9 percent in all years. Year three has not had a down year since 1939. And so on.

Clearly, there is a pattern. Does this make sense, or is it a mere a statistical accident? Do political considerations affect the stock market?

Sure it makes sense. The leading theory is that in the first year or two of a president's term, economic sacrifices are made. Painful decisions come early, such as fighting inflation, cutting back spending, and even starting wars. New priorities are introduced, fresh ideas abound. But by the third year of its hold on the White House, the incumbent administration emphasizes economic stimuli to gain favor for the coming election campaign.

Congress--no matter which party holds the White House--also wants to gain favor for the upcoming election. Presidential election years have the biggest stakes of any national election. All of the House's seats are in play, along with a third of the Senate's seats, and of course the presidency itself. Whether the Democrats or Republicans currently control the White House, each party wants to win it. So Congress wants to juice the economy too.

In essence, year three is the "setup" year for both parties to hit next year's campaign trail with their best arguments in place. The incumbent party wants voters go to the polls with jobs and a feeling of economic well being. The party out of power wants to have an economic record that they can argue is even stronger than the incumbent party's.

Will the cycle repeat itself in 2007? Well, we've had the capture of both houses of Congress by the Democrats, who will set the congressional agenda for the next two years. But the White House is still controlled by the Republicans. Does that equate to uncertainty, which is historically felt to be an enemy of a rising market? Or does it just mean stalemate, which is not usually felt to be a bad thing for the market?

Only time will tell. But in the absence of a disastrous negative event (which does not appear on the horizon at this time), or compelling negative data (as you might have if the market were wildly overvalued, which it is not), the historical pattern very strongly suggests that 2007 will be a good year for the stock market. After all, the long-term data itself covers election cycles in which there were all kinds of economic conditions and every combination of party control of the White House and Congress. The clear pattern of strong returns in year three has emerged from all those varying conditions.

Let's hope that this year three of the cycle, 2007, matches the last one, 2003, when the market returned 26 percent. That would be something that all stock investors could celebrate.