FLOYD NORRIS

Notions on high and low finance.

If the Super Bowl game in two weeks is not close, Barack Obama may be in trouble when November rolls around. On the other hand, a close game would give him a better-than-even chance of winning the election.

That, anyway, is the result of my efforts today to find some relationship between election-year football games and November election results. I’ll explain why later.

Of course, there used to be a theory that held that the Super Bowl could forecast the stock market. That worked really well for a couple of decades. Since then, it has been useless.

Before we get to those examples of spurious correlation, let me note one fact about the game coming up. If the New York Giants win, they will be the worst team ever (as measured by regular season records) to win. As it is, they appear to be the worst team ever to get into the game.

They finished the regular season with 9 wins and 7 losses.

Two other teams with a record that bad made it to the final game. They lost.

In 1980, the Pittsburgh Steelers (12-4) beat the Los Angeles Rams (9-7) by a score of 31-19.

In 2009, Pittsburgh (12-4) beat the Arizona Cardinals (9-7), by 27-23.

Each of those losers had managed to outscore their opponents during the regular season, although the Cardinals did so by a one-point margin. The 2011 Giants gave up 400 points during the regular season while scoring 394 points.

Three 10-6 teams have won, the most recent being  the Green Bay Packers a year ago. The others were the Giants in 2008 and the San Francisco 49ers in 1989.

Over all, the best team (as measured by regular season record) has won 25 and lost 13. The other seven games involved teams with identical regular season records.

But of the last six Super Bowls, the (regular season) better team won only one. That was the Pittsburgh victory in 2009.

There is a case that election-year history favors the Giants. In election years, a team from the old National Football League has won the Super Bowl nine times, the most recent being the Giants four years ago. The Pittsburgh Steelers, an old N.F.L. team now in the American Football Conference, won two of the nine, in 1976 and 1980. Only two teams from the old American Football League have won in election years — the Los Angeles Raiders in 1984 and the New England Patriots in 2004. The Republicans won both those elections.

There used to be a theory that the winner of the Super Bowl foretold how the stock market would do. If an old N.F.L. team won, the market would rise. If an old A.F.L. team won, the market would fall.

Of course, the theory was developed and tweaked based on early Super Bowls, including the decision to count Pittsburgh as being an old N.F.L. team, rather than as a representative of the A.F.C. But it kept working even after that, and it developed a real following on Wall Street.

From 1967, the year Green Bay (old N.F.L.) won the first Super Bowl, through 1989, when the San Francisco 49ers (N.F.L.) won, the market was up after all 16 old-N.F.L. teams won, and down six of seven times after old A.F.L. teams prevailed. The only time it failed was after the Oakland Raiders won in 1984, and that year the S.P. 500 rose just 1.4 percent.

Since then, the theory has forecast the market correctly 11 times, while getting it wrong 10 times. (I am not counting 2003, when Tampa Bay won. That team came into the league after the merger, first joining the A.F.C. and then moving to the N.F.C. It is not clear what the theory would have forecast that year. If you go by the idea that the team was first in the A.F.C., and therefore that its victory was a bearish indicator, that was another loss for the theory. Stocks soared in 2003.)

The Giants’ victory in 2008 was followed by the worst year for stocks since 1931, before there was an N.F.L. championship game.

I searched for evidence that the Super Bowl result somehow forecasts the presidential election. And I found a relationship, which proves that if you try hard enough, you can almost always find some correlation.

Here it is:

If the game is a runaway, the Republicans will win in November.

Of the 11 election-year Super Bowls so far, the Democrats are 4-3 when the game was decided by less than 14 points. But when the game was not close, the Republicans won all four times. The last such romp was in 1988.

Article source: http://feeds.nytimes.com/click.phdo?i=1da2d8070af48b720581c0a3a325c18e