Forget the jobs numbers. They aren’t that important.

Even economic growth is secondary. Gas prices provide a mere distraction, and inflation hardly matters at all.

In fact, none of the most watched economic barometers has much value in predicting who will win the presidential election.

Only the stock market really matters.

This decidedly unconventional wisdom is the conclusion of some recent scholarly research done at Emory University. No beach read – its title is “SocialMood, Stock Market Performance and U.S. Presidential Elections” – the Emory research, nevertheless, is both timely and provocative.

The scholars, led by Robert Prechter, examined one of the oldest riddles in American politics: what best predicts the outcome of presidential elections?

Now it’s not exactly a national secret that most political analysts believe “the economy” decides presidential winners and losers. Believers in “the economy” did it aren’t scarce. But numerous as they are, many are vague about how to measure precisely what in “the economy” matters so much. Indeed, there is substantial argument about whether it is the unemployment rate, GDP, inflation or some other macro statistic that tells the tale.

Consequently, the Emory team first honed in on a basketful of traditional economic variables. These included unemployment rate, GDP and inflation. They also looked at changes in the stock market – up, down and sideways – to discover if there was a link between market performance and incumbent re-election.

They analyzed popular vote data as far back as the 1820s and Electoral College results back to George Washington. Relying on this massive data trove encompassing almost two centuries, their conclusions were stunning.

They found little evidence that traditional economic variables like the rate of inflation or the unemployment rate had much direct impact on the election. Economic growth (GDP) does seem to have some predictive value but the real 800-pound gorilla in the room is the stock market.

The researchers found that presidential incumbents were almost always re-elected if they presided over a rising stock market the last three years of their term. Conversely, incumbents almost always lost if their tenure coincided with declining markets. Moreover, the larger the gains or losses recorded by the market, the more certain the electoral outcome.

The researchers also uncovered what they called the “recency effect.” Essentially, stock market fluctuations that occurred last month influence voting much more than fluctuations that occurred last year.

For example, market performance in the ten months prior to an election is a much better election forecaster than GDP over the last three years. So the closer to the election the stock market goes up or down, the more influence it has on whether an incumbent is re-elected.

Cause and effect are tricky here. According to the Emory team, it’s not actually the stock market that causes voters to vote for or against the incumbent. Instead the researchers believe stock market up or down trends simply reflect the national mood.

If voters feel optimistic the stock market goes up and incumbents are re-elected. If voters feel pessimistic the market goes down and incumbents are defeated. It’s the mood that matters.

What does this tell us about the 2012 presidential election now underway?

At first blush it seems to strongly favor Barack Obama’s re-election chances since the stock market has recovered more than 50 percent since his 2009 inauguration. The Dow Jones stood around 8,000 in January 2009. Recently, it was trading just north of 12,500. A rise of that magnitude normally augurs a happy election day for the incumbent.

But this election campaign still has some five months to play out, while the stock market recently has been volatile. The major market averages are still up for the year but most lost 6 percent or more in May. Problems in Europe remain unresolved while at home the national mood seems glum. The most recent Conference Board’s Confidence Index fell to a 4-month low in May.

All of this is this is clearly inauspicious for Obama. Voters are in a foul mood, and according to the Emory researchers, mood matters. Worse perhaps, under the “recency effect,” what happens last is what matters most. In particular, stock performance in the third quarter of the year (July through August) influences voting most because voters remember it best.

So the stock market’s performance between now and Election Day may hold Obama’s fate. If there is a continuing slide in stock prices, President Obama may find himself ex-President Obama. On the other hand, if prices stabilize, he probably hangs on.

Only fools and madmen hazard to forecast what the stock market will do any given day, week, month or year. Given recent market gyrations, however, it’s likely to be a wild ride.

One thing we do know for sure: Election Day is Nov. 6. If the Emory researchers are correct, America’s mood on Nov. 6 will determine Barack Obama’s mood Nov. 7.