It probably hasn't made much difference to the outcome since 1960, when Lyndon B. Johnson helped put Texas in John F. Kennedy's column. Nevertheless, when pundits review their short- lists, the first question they usually ask is whether the prospective veep comes from a winnable swing state. Long ago, when state parties had tightly run political machines, that logic made some sense, and even into the postwar era it governed many candidates' thinking. But not so much today. Edmund Muskie might have helped the Democrats win Maine in 1968 (en route, it should be noted, to losing to Richard M. Nixon), and Walter F. Mondale's presence at Jimmy Carter's side probably enabled them to hang onto Minnesota when Hurricane Reagan swept away 44 states in 1980. More often, though, the No. 2 either fails to carry his or her home state or simply isn't chosen with such hopes in mind in the first place. Of this year's Democratic contenders, only Ohio Gov. Ted Strickland promises to nab a battleground state for Sen. Barack Obama. On the Republican side, not even Tim Pawlenty, Minnesota's moderate but little-known governor, seems likely to shore up that swing state for the GOP.
2. Ideological and regional balance are vital to a ticket.
This assumption, too, was once valid but no longer holds. In the 19th century, the heyday of political machines, voters felt a strong allegiance to one party or the other, and a race's outcome was determined by how well the party operatives turned out their vote. Because these same operatives picked the nominees — primaries didn't play a decisive role in nominations until 1972 — conventions typically featured nasty dustups between a party's ideological wings or regional blocs. Ticket-balancing arose as a way to preserve peace at convention time. So the bosses would pair a Northeasterner with a Midwesterner — and indeed, between 1864 and 1920, two-thirds of all national candidates came from New York, Indiana or Ohio.
But the once-powerful logic of balance has eroded as the parties have become more ideologically uniform: The Republicans are no longer fiercely divided between the Eastern Wall Street Establishment and the Midwestern Main Street wing, and the Democrats no longer have a sizable Southern conservative bloc deeply hostile to their Northern liberal base.
Moreover, voters today choose their candidates on the basis of image more than of party. And so although balancing efforts still sometimes occur — Ronald Reagan's selection of George H.W. Bush in 1980, Michael Dukakis' choice of Lloyd Bentsen in 1988 — nominees will more often use their moments in the spotlight to send signals about their own candidacies. They might wish to augment their own profiles, as Bill Clinton did in 1992 by choosing Al Gore — another young Southern moderate — or to seek a bounce from a bold, exciting pick, as Mondale did by tapping Geraldine Ferraro in 1984.
To the extent that ticket-balancing survives, it's in the form of signaling neglected constituent groups that they're not taken for granted. This year, Obama lacks strong support from enough key Democratic voting blocs — Hispanics, Jews, women, gays, seniors and the white working class — that he should be sensitive to their concerns in choosing a veep. John McCain, on the other hand, would do well to bolster his own reputation as a maverick by choosing someone like Colin Powell or Mike Huckabee.
3. Reaching across the aisle to form a bipartisan ticket would be smart politics.
Well, maybe, but it's not going to happen. Every four years, we hear titillating talk that one of the nominees will make a paradigm-shifting move by choosing an understudy from the opposite party. Favorite names served up this year are Sen. Joe Lieberman (who rumor has it used to be a Democrat) as a potential McCain lieutenant and Republican Sen. Chuck Hagel or New York Mayor Michael Bloomberg as possible Obama backstops. This isn't unthinkable; in 2004, John F. Kerry considered putting McCain on his ticket. But the chances of that pairing's coming to fruition were always exceedingly slim, as are the prospects for this year's cross-party dream teams.
The reason is that, as weak as parties have become, the choice of a president and a vice president still defines a party. To turn to the rival institution in this basic act of self-definition would be an abdication — a concession that the party is nothing but a vehicle for the ambitions of individual politicians, not a coherent body with a purpose.
Incidentally, the last president to choose a vice president from the opposite party was Abraham Lincoln, who selected Andrew Johnson in 1864 — a pairing that didn't work out very well. (Lincoln was assassinated, Johnson impeached.)
4. Candidates should think "outside the box."
For candidates, as for housecats, the box is there for a reason.
Those who aren't fantasizing about a bipartisan ticket often dream that a nominee will look outside politics altogether for his or her deputy. (Former Hewlett-Packard chief Carly Fiorina is one name bandied about this year.) Washington loves no cliche more than the cliche of Washington's bankruptcy — the myth that only someone with no political experience, usually a mogul or a financier, can fix our political mess. (It's a myth that got the nutty H. Ross Perot 19 percent of the vote in 1992.) But it hasn't happened yet — and probably won't. Occasionally an Arnold Schwarzenegger or a Bloomberg succeeds in high office without government experience. But most politicians canny enough to get a presidential nomination know that the fall campaign isn't the place for a novice to be learning the game.
5. All of this is beside the point: The choice of a running mate doesn't matter.
The aphorisms are legion: John Adams called the vice presidency "the most insignificant office that ever the invention of man contrived or his imagination conceived." Nelson Rockefeller called it "stand-by equipment." John Nance Garner, we often hear, said that the job wasn't "worth a bucket of warm spit," though he didn't actually say "spit." But these days, the vice presidency is worth a lot. Ever since Nixon held the office under Dwight Eisenhower, the office has been growing in power, reaching unprecedented influence under Dick Cheney.
But, does anyone vote based on the veep? The kibitzing over a nominee's choices is clearly out of proportion to the importance of the selection. Given that George H.W. Bush survived the calamitous selection of Dan Quayle in 1988, it's hard to argue that even a bad choice can sink a candidacy — though George McGovern's jettisoning of Tom Eagleton in 1972 sure didn't help his already floundering campaign.
Still, significant numbers of voters tell pollsters that the candidates' veep choices help them decide how to vote in the fall. A March poll by SurveyUSA showed that between 12 and 40 percent of voters, depending on the state and the top of the ticket, said that the vice presidential choice would influence them. And more recent polls show huge swings in how various states break depending on the choices. (In New Mexico, for example, McCain-Huckabee beats Obama-Hagel by 17 points, but McCain-Pawlenty loses to a pairing of Obama and former senator John Edwards by nine points.)
That kind of self-reporting is unreliable, of course. Name recognition alone, for example, probably accounts for the boost that Edwards gives Obama in some polls. But some political scientists do surmise that the choice can shift the popular vote by a smidgen — and even 1 or 2 percent can be enough, in the kinds of close presidential elections we've had in 2000 and 2004, to make all the difference.
David Greenberg is a professor of history and media studies at Rutgers University, a columnist for Slate and the author of "Nixon's Shadow: The History of an Image."