While all of these indicators have shown some improvement in the past year, according to Gallup they all remain at levels that suggest trouble for the incumbent. For example, only 24% of Americans said that they were satisfied with the direction of the country and 66% cited the economy as the most important problem facing the nation.

There is little evidence about how indicators like satisfaction with the direction of the country or perceptions of the most important problem facing the nation affect the outcomes of presidential elections. However, there is strong evidence that an incumbent president’s approval rating, even several months before Election Day, has a strong relationship to the eventual outcome of the election.

After examining the approval ratings in May of the election year for all incumbents who have run for a second term since 1964, the Gallup article noted that President Obama’s 47% approval rating is lower than that of all of the incumbents in this group who were reelected, including Lyndon Johnson, Richard Nixon, Ronald Reagan and George W. Bush, although it is only slightly lower than Bush’s 49% approval rating in May 2004. However, Gallup also noted that Obama’s May approval rating is higher than that of all of the incumbents in this group who were defeated: Gerald Ford, Jimmy Carter and George H.W. Bush.

One problem with the Gallup analysis is that it leaves out two other postwar incumbents — Harry Truman and Dwight Eisenhower. Eisenhower’s 69% approval rating in May 1956 was the highest of any postwar incumbent except Lyndon Johnson in 1964, so it is hardly surprising that he was easily reelected. However, Harry Truman’s 39% approval rating in May 1948 was the lowest of any president seeking a second term, including the three who were defeated. Yet Truman, much to the surprise of many pollsters and political observers at the time, was reelected. So what conclusion should we draw from Obama’s 47% approval rating?

Figure 1: Incumbent vote by May approval rating

Source: Gallup Poll and data compiled by author

Figure 1 displays a scatterplot of the relationship between the May approval ratings of the 10 incumbents seeking a second term since the end of World War II and the share of the major party vote that each of them received in November. The data displayed here show that there is a fairly strong relationship between May approval and election outcomes. May approval explains about two-thirds of the variation in the incumbent’s share of the vote in November.

Based on President Obama’s May approval rating of 47% and the prediction line shown in Figure 1, we would expect Mr. Obama to receive 51.6% of the major party vote in November. These results suggest that President Obama is currently a slight favorite to win a second term. However, our ability to predict the November vote based on the incumbent’s May approval rating is limited, as evidenced from the fact that several of the points for individual elections in Figure 1 are well above or below the prediction line.

Fortunately, my previous research shows that we can greatly improve the accuracy of our November forecast by including two additional predictors along with the incumbent’s approval rating: the growth rate of the economy during the first two quarters of the election year and the “time for change” factor, which indicates whether the incumbent’s party has held the White House for only one term or for two or more terms. Incumbents whose party has held the White House for two or more terms do significantly worse than those whose party has held the White House for only one term, even after taking into account the incumbent’s approval rating and the condition of the economy.

Figure 2: Incumbent vote by predicted incumbent vote based on Time for Change Model

Source: Gallup Poll and data compiled by author

The results displayed in Figure 2 show that a forecasting model including the change in real GDP during the first two quarters of the election year and the time for change factor along with the incumbent’s approval rating in May produces much more accurate predictions of the results of these 10 presidential elections. Despite the limited number of elections included in the analysis, the effects of all three predictors are statistically significant. We are now explaining over 90% of the variation in the incumbent’s share of the vote in November instead of only 67%, and all of the points representing individual elections in Figure 2 are very close to the prediction line.

So what does our improved forecasting model indicate about President Obama’s chances of winning a second term in the White House? That depends of course on the growth rate of the economy during the first two quarters of the year. The government’s initial estimate of real GDP growth during the first quarter of 2012 was about 2%, and growth during the second quarter is expected to be similar. If we assume that real GDP will grow by 2% during the first half of 2012, the full forecasting model predicts that Obama will end up with 51% of the major party vote in November, not much different from what the simple approval-based model predicted.

Conclusion

Whether we base our prediction on President Obama’s 47% approval rating in the Gallup Poll in early May or a more sophisticated forecasting model incorporating economic conditions and the “time for change” factor, it appears likely that we are headed for a very close election in November. Both models make Obama a slight favorite to win a second term. However, the final outcome will depend on the actual performance of the economy and the public’s evaluation of the president’s job performance in the months ahead. Those interested in assessing where the presidential race stands should focus on these two indicators rather than the day-to-day events of the campaign, which tend to dominate media coverage of the election.

Emory University professor and Crystal Ball Senior Columnist Alan Abramowitz takes an in-depth look at the seemingly irreconcilable divide between Republicans and Democrats in his new book, The Polarized Public? Why American Government is So Dysfunctional. Abramowitz argues that bipartisanship remains elusive, not because of politicians in Washington, but because of the American public and its fixation on party membership and loyalty. Now available on Amazon, click here to get your copy today.

Mitt Romney faces a complicated vice presidential choice, and his predicament traces to two factors: His campaign has multiple needs and the pool of potential candidates offers imperfect options. Romney’s situation is not, however, novel. If history is a guide, his options will sort out over time and, like his predecessors, he will ultimately choose from imperfect alternatives.

Vice presidential selection is inherently contextual and relational. It is contextual because the choice invariably depends upon a range of factors over which the candidate has relatively little control, including the race’s apparent competitiveness, the issues most likely to be important and the candidates who are available. Facing an unfavorable political landscape in 1984, Walter F. Mondale saw a need to reshape the terrain and tried to do so by choosing the first woman running mate, Geraldine Ferraro. He might have made a different choice if polls had forecast a closer race. Tax and economic issues seemed a favorable place for Bob Dole to make his stand in 1996, which helped explain the choice of Jack Kemp.

The choice of a running mate is also relational because a presidential nominee must always choose a running mate based upon needs particular to the candidate doing the selecting. The familiar concept of “ticket-balancing,” implies just that — the presidential nominee selects a running mate in relation to his or her own strengths and weaknesses. George H.W. Bush was a good choice in 1980 for Ronald Reagan, who wanted a running mate from the more moderate wing of the party and one with some national security credentials, but Bush would have been a less compelling running mate in 1980 for Gerald Ford. Joe Biden made great sense for Barack Obama, but might have been less likely had Hillary Clinton been the nominee. Mondale was a perfect fit for Jimmy Carter, but an unlikely running mate for Frank Church, Morris Udall or Hubert H. Humphrey. A reinforcing choice, as Bill Clinton made in selecting Al Gore, shares this relational character.  In choosing Gore, Clinton underscored attributes — that he was a southern centrist from the baby boomer generation — that were linked to his own biography.

Most polls forecast a competitive presidential race. If this expectation holds into the summer, Romney will be less likely to attempt to remake the political landscape with an unconventional selection and will be more likely to seek a running mate who appears to be a plausible president. Recent history suggests that most presidential nominees facing competitive races choose a running mate who would make a credible president. Determining whether someone satisfies that test involves subjective judgments, although some other criteria will help inform the decision. These include the amount of experience candidates have in traditional vice presidential feeder positions, their prior consideration for the presidency, their prior consideration for vice president and, most subjective, the way other officials and opinion-molders perceive them.

Since 1976, most first-time vice presidential candidates have been plausible presidential candidates based on these sorts of measures. Those chosen averaged 14.5 years in traditional feeder positions (governor, member of Congress, high executive official) for the vice presidency. All but Ferraro, John Edwards and Sarah Palin had at least 10 years experience in such positions, and Edwards had the compensating credentials of having been runner-up to John Kerry in the 2004 presidential primaries and a serious VP contender for Al Gore’s ticket four years earlier. Surely Dole, Mondale, Bush, Lloyd Bentsen, Al Gore, Jack Kemp, Joe Lieberman, Dick Cheney and Biden were among the leading political figures of their generation when chosen. Dan Quayle had served 12 years in Congress and was building a record in the Senate.

Many of those around whom vice presidential speculation has centered this time do not measure up based on these criteria. Indeed, many of those most prominently mentioned this year have extremely limited experience, have not demonstrated their presidential qualities through a presidential race, or have not even previously been considered for the second spot on a ticket. Chris Christie (NJ) and Bob McDonnell (VA) have been governors for three years; Marco Rubio (FL), Kelly Ayotte (NH) and Rand Paul (KY) have been senators for only two years; and Nikki Haley (SC), Susana Martinez (NM) and Brian Sandoval (NV) have been governors for only two years (all figures rounded up). Based on prior service in traditional feeder positions, these levels place these candidates near the Sarah Palin-Spiro Agnew line (i.e. two years as governor), hardly an aspirational standard.

Of course, one or more of these people may be among those extraordinary figures who are ready to perform well on the national stage in 2012 even though their resumes are short and they have not yet demonstrated success in presidential politics. Many are now trying to make their case during this several-month vice presidential audition period.  And perhaps anti-Washington sentiment makes this year an anomaly. Selecting a political neophyte presents risks, however. These figures have not been tested by campaigns comparable to a presidential contest, and such a choice would draw inevitable comparisons to John McCain’s designation of Palin. If the person selected did not perform well, Romney’s judgment would be questioned.

The risk for Romney would be even greater because his own career in public service is short, consisting of a single term as governor of Massachusetts, and he does not have experience in foreign policy or national government. These gaps in Romney’s resume further diminish the prospects of those mentioned above. Indeed, every governor nominated for president since 1976 has chosen a running mate with extensive experience in national government. Thus, Carter chose Mondale, Reagan chose Bush, Michael Dukakis picked Bentsen, Clinton selected Gore and George W. Bush selected Cheney. For Romney to choose a relative newcomer to high public office would represent a break from the path of these recent governor-presidential candidates.

Conversely, Romney might try to emphasize his theme that he is an economic turnaround expert by selecting a running mate who reinforces that message. That might lead Romney to choose a running mate with experience, but in creating jobs rather than in national public service.

To be sure, the Republican Party does include figures who fare better on some of the standard measures for qualifying running mates. Gov. Mitch Daniels (IN, 10 years), Sen. Rob Portman (OH, 16), Sen. John Thune (SD, 14), Gov. Bobby Jindal (LA, 8), Rep. Paul Ryan (WI, 14) and former Govs. Jeb Bush (FL, 8) and Tim Pawlenty (MN, 8) have all served at least eight years in traditional feeder positions and several were viewed as plausible presidential candidates earlier in this cycle. Jindal and Pawlenty were among those who made McCain’s vice presidential shortlist. Former Sen. Rick Santorum (PA) and former Gov. Mike Huckabee (AR) might also qualify based on the relative success of their presidential runs this year and in 2008, respectively. Yet the acrimonious nature of this year’s campaign will surely disqualify Santorum. A runner-up occasionally has been chosen (Bush, 1980, Edwards, 2004), but they left the race once the ultimate result was evident and conducted less divisive campaigns than did Santorum. Those who run but finish further back in the pack rarely are designated, and Pawlenty’s poor showing certainly did not bolster his chance to become Romney’s choice. Biden is the one exception to this latter pattern, yet he ran against a stronger field in 2008 than this year’s Republican contest featured and, unlike Pawlenty, brought a long record of national public service to the ticket.

In addition to being presidential, potential running mates will need to prove to Romney’s vetting team of lawyers, accountants and other nitpicking specialists that their (and their families’) backgrounds contain no blemishes that would be too difficult to manage. Generally speaking, some past possibilities have been unable to survive a vetting screen because of reputed (or confirmed) philandering, questionable financial dealings, bizarre habits or a tendency to go rogue.  These or other such stains may remove some of those who figure in speculation. And some will not get the nod because of their association with policy positions Romney may seek to avoid. Howard Baker’s role in securing ratification for the Panama Canal treaties damaged his chance of being Reagan’s running mate. Being pro-choice has, in the past, disqualified Sen. Alan Simpson (1988), Gov. Tom Ridge (2000, 2008) and Lieberman (2008). Sandoval will likely be this year’s casualty on these grounds, as will former Secretary of State Condoleezza Rice, who has described herself in the past as “mildly pro-choice” (as well as supportive of some affirmative action). Ryan might be an appealing choice, but he comes with his controversial budget plan. Romney has endorsed it, but will he really wish to be defined by that blueprint by choosing him?

Like any presidential nominee, Romney’s choice will respond to his needs as a presidential candidate. Romney enters the race with some obvious weaknesses. First, as mentioned above, Romney’s lack of national security credentials makes it less likely that he will choose someone whose exclusive experience has been as a state governor.

Moreover, the conservative base of the Republican Party has had misgivings about Romney, which contributed to his difficulty securing the nomination against this year’s weak field. Every recent Republican nominee from the more moderate wing of the party (Bush 1988 and 1992, Dole 1996, McCain, 2008) has felt compelled to select a running mate popular with the party’s conservative base. It seems likely that he will need to choose someone who at least will not offend, and perhaps will excite, the Republican base. Romney will want his choice to produce a happy and unified convention.

Yet Romney must balance his desire to pacify the conservative, evangelical base of the Republican Party against his need to appeal to independent and undecided voters in swing states. If he chooses a right-wing hero to prove he’s a true believer, he may offend independent voters. If he makes a choice to appeal to independents, he could provoke a conservative revolt or, perhaps more likely, an apathetic base.

Romney faces problems with other demographic groups, women and Hispanics among them.  Some recent polls showed him far behind among Hispanics, an important voting bloc in some swing states, and lagging among women, especially unmarried women. Yet choices thought to appeal to Hispanic voters, like Rubio, Martinez, Sandoval or Jeb Bush, or women, like Ayotte, Haley, Martinez or Rice have other drawbacks, including those suggested elsewhere in this discussion.

Romney’s affluence, coupled with his occasional gaffes that emphasized his economic status (e.g., his $10,000 bet, his wife’s two Cadillacs, etc.) may cause him to look for someone who would not replicate his elite pedigree. Such considerations could work against Portman or Bush among others, and in favor of someone like Pawlenty.

Finally, Romney will be constrained by some filters that are independent of anything he has done. Although past Republican nominees have frequently chosen running mates with substantial experience in the executive branch (e.g., Bush in 1980, Kemp in 1996, Cheney in 2000), this qualification may have less appeal this time. Many of the possible candidates are associated with the George W. Bush administration either based on service (Daniels, Portman, Rice) or family (Jeb Bush). Does Romney wish to associate himself with that administration by choosing a running mate with such baggage? To do so would emphasize the extent to which America’s economic problems predated President Barack Obama’s inauguration.

Republicans in Congress will also have one or two strikes against them. Sitting members of the House of Representatives are almost never selected as running mates, in part because they are perceived to have a stature deficit relative to senators, governors or members of the executive branch. Ferraro is the only sitting House member to be selected since 1976, and she was chosen at a time when there were no Democratic women in the Senate and only one recently elected woman Democratic governor. William Miller, in 1964, is the only other House member chosen going back more than 75 years. Barry Goldwater, who had limited options, chose Miller due to his proclivity at provoking Lyndon B. Johnson. Ford was selected under the 25th Amendment, but that was an extraordinary situation.

The bias against choosing a member of the House might count against Ryan, House Majority Leader Eric Cantor (VA) or Rep. Cathy McMorris Rodgers (WA). Moreover, such a selection would associate Romney with the unpopular Congress and perhaps enhance Obama’s ability to tie Romney to it. That consideration might also affect the likelihood that Romney would select Thune, Portman or another Republican senator. Incidentally, although the Senate is an incubator for Democratic running mates, only 20% of Republican vice presidential candidates since 1952 (Nixon, Dole and Quayle) were sitting senators when chosen.

If history is a guide, Romney’s options will clarify in the coming months. The ultimate nominee is almost never apparent in late May of election year. Few would have predicted Mondale, Dole, Bush, Ferraro, Bentsen, Quayle, Gore, Kemp, Lieberman, Cheney, Biden or Palin at that time in the year each was chosen. As the convention approaches, Romney’s relative electoral strengths and weaknesses will become more apparent, and his options will narrow as various prospective choices rise or fall based on the outcomes of the vetting, their own conduct and strategic considerations. And he will probably achieve a better sense of his relative comfort level with the various alternatives as a political partner.

Ultimately, Romney will need to choose between imperfect options. Presidential candidates always do. Bush was a great choice for Reagan in 1980, yet he had labeled Reagan’s signature policies “voodoo economics,” had never been elected to anything outside of Texas’s Seventh Congressional District (although he had won seven primaries or caucuses and previously been considered for the vice presidency) and Reagan had reservations about him. Bentsen disagreed with Dukakis on various issues. Clinton’s selection of Gore violated conventional practices regarding ticket-balancing. Dole chose Kemp although the two had well-publicized differences, which Kemp had exacerbated when he endorsed Steve Forbes at a point when Dole’s nomination was inevitable. Cheney, Biden and Palin were from tiny, safe states.

In picking a running mate, Romney will tell us something about himself. In addition to being contextual and relational, a vice presidential choice is idiosyncratic. It matters who is making the choice and who has his or her ear. Within the constraints a nominee faces, the choice tells us something about the selector’s values and decision-making style and ability.  McCain was perceived to undermine his promise to put “Country First” when he chose Palin. Reagan, Dukakis and Dole signaled they were open to a spectrum of views when they chose Bush, Bentsen and Kemp, respectively. The choice of Cheney in 2000 reassured voters that Bush would choose experienced and able advisers whereas Gore’s selection of Lieberman, an early critic of Clinton regarding the Lewinsky affair, signaled an emphasis on values, independence from Clinton, and boldness in selecting the first Jewish American to run on a national ticket.

Romney, like his predecessors, faces constraints in making his vice-presidential choice. How he decides from a menu of imperfect options will reveal something about him. That, after all, is part of what presidents do, and how they should be judged.

Chart 1: State-by-state unemployment rates, April 2012

Notes: P – preliminary figure; R – revised figure based on populations, model reestimation and new seasonal adjustment.

Source: National Bureau of Labor Statistics

Obviously, the safe Blue and Red states on the chart are going to stick with their preferred party, even if their unemployment is high, as in the case of Democratic states (like California), or low, as in the case of Republican states (like the Dakotas). At the same time, the president would have trouble making a compelling case about jobs in Nevada or North Carolina, for instance — two states that he won in 2008 after George W. Bush won them in 2000 and 2004, and where unemployment, despite a downward trend, remains high. But he has a better argument in states where unemployment is below the national average.

So far, the Obama campaign has run ads promoting the president’s handling of the economy, such as spots that tout the auto industry bailout and mention increased job growth. But are voters buying the pitch and giving Obama credit? That’s up for debate, especially with Republican governors in key swing states, such as Virginia and Ohio, competing with the president for the public’s applause.

In Virginia, in what can mainly be described as a campaign to improve his chances of being Romney’s running mate, Gov. Bob McDonnell’s (R) Opportunity Virginia PAC has run an ad highlighting Virginia’s economic improvement during McDonnell’s tenure. The spot notes that Virginia has its lowest unemployment rate in three years and the lowest in the Southeast. As our chart shows, Virginia’s 5.6% figure is at least 1% better than any other Southern state. Federal spending, particularly defense expenditures, is a big reason why, of course — a point often left unmade in a state whose politicians regularly launch broadsides against “wasteful spending by Washington.”

Meanwhile, Ohio and much of the Rust Belt have seen stirrings of economic improvement. But the president has not necessarily received a significant bump from this news. A recent Quinnipiac poll found that Ohioans who think the Buckeye State’s economy has improved give Gov. John Kasich (R) credit for the change by a 68% to 22% margin over President Obama. Voters who think the economy is worse also blame the sitting governor more than the president, 49% to 27%. Considering Ohio’s unemployment rate has gone from 8.8% in April 2011 to 7.4% last month, both incumbents can brag about the change. But it is far more important for Obama, who is on the ballot this November while Kasich isn’t up for reelection until 2014.

Strategically, the Obama campaign wants to convince voters that the economy is in fact improving. Tactically, this has meant running ads in key swing states that generally promote Obama’s economic stewardship. Yet the campaign might be losing an opportunity if it doesn’t take greater ownership of positive state-specific numbers. Obama’s generic television ads might do more than simply target all the swing states as a bloc. Instead, he could focus on each state separately. If a state’s unemployment rate has improved over the past year, then the president’s campaign could run general election ads that trumpet the success. Ohio and especially Virginia are ideal for such advertising.

In politics, a president gets the blame for anything bad that happens on his watch. Conversely, he gets the credit for anything good that unfolds during his term — that is, if he doesn’t let others take the credit from him. To this point, President Obama has failed to take advantage of the improved jobs numbers in some competitive states with unemployment lower than the national average. In this close election, Obama has little margin for error.

After the Illinois primary on March 20, which signaled the beginning of the end of the Republican presidential nominating process, Mitt Romney adviser Eric Fehrnstrom discussed how his candidate would pivot toward the general election: “It’s almost like an Etch A Sketch. You can kind of shake it up and restart all over again.”

Several days later, there was President Obama, who told Russian President Dmitri Medvedev that he would have “flexibility” to negotiate on missile defense after the election.

Then there were the comments from Hilary Rosen, a Democratic strategist, who said that Ann Romney (wife of Mitt) “never worked a day in her life.”

And last week, after his vice president pushed him into a corner, Obama surprised no one by endorsing gay marriage, a Darwinian evolution from his earlier position of supporting just civil unions. The very next day — in a bit of timing we’ll simply call curiously coincidental — the Washington Post reported that Mitt Romney, as a prep school boy, cut the hair of a purportedly gay classmate in an act of bullying at his elite boarding school.

What do all of these things have in common? They are all, to at least some degree, “feeding frenzies.”

Candidates beware: With the presidential silly season upon us, you are entering what can be described, contra Bill O’Reilly, as the “All-Spin Zone.” Without actual news, the press will fixate on any gaffe, big or small. But how many of these frenzies actually matter?

What is a feeding frenzy?

A feeding frenzy — as defined by the book of the same name — is “the press coverage attending any political event or circumstance where a critical mass of journalists leap to cover the same embarrassing or scandalous subject and pursue it intensely, often excessively, and sometimes uncontrollably.”

Not all feeding frenzies are the same, but they generally can be classified into three levels of severity:

• Mega Frenzy: The infrequent “black hole” frenzy that sucks in most available media light and dominates the news for weeks or months. It is remembered as the dominant story of a particular election or era: Watergate, the Iran-Contra Affair, and Bill Clinton’s affair with Monica Lewinsky all qualify.
• Full Frenzy: A major event that connects to a candidate’s existing subtext — that is, the way journalists think the real person underneath the public exterior is defined. An example is Gary Hart’s reckless extramarital relationship with Donna Rice, which he was pursuing even as the candidate invited reporters to “follow me around. You’ll be very bored.” The revelation effectively knocked Hart out of the 1988 Democratic presidential nominating contest and confirmed lingering suspicions about his character.
• Frenziette: The summer squall of feeding frenzies, which seems important at the moment but fades quickly and has no real impact. A New York Times article about John McCain’s interactions with a lobbyist in 2008 is an example. The story, which seemed to suggest far more than it proved, didn’t appear to hurt McCain in the slightest.

The context matters

So far at least, all the examples listed above from the 2012 campaign season probably belong in the “frenziette” category. But that doesn’t mean they are meaningless.

When assessing media feeding frenzies, the context matters. Incidents that seem to confirm or reinforce an existing stereotype about a candidate have a better chance of being covered. For instance, when President Gerald Ford said during a 1976 debate that “There is no Soviet dominanation of Eastern Europe,” it played into the (false) image of Ford as a clumsy bumbler. In the same election, Jimmy Carter gave an interview to Playboy saying that he had “committed adultery in my heart many times.” In an interview setting that was clearly designed to change Carter’s image as a sanctimonious, holier-than-thou candidate, he came off sounding… well, sanctimonious and holier-than-thou.

The four incidents from recent weeks all have important subtexts:

• The “Etch A Sketch” comment attracted attention because of Romney’s well-known propensity to flip-flop on major issues.
• Rosen, though little-known outside the Beltway, expressed a critique of Ann Romney’s traditional motherhood and wealth that was custom-made to rile conservatives and fire up the nation’s recurring culture wars about the role of the modern woman.
• Obama’s overheard promise of “flexibility” to the Russians played into a conservative fear about Obama, which is that he has partly hidden his real agenda and that his second term will be more liberal than his first. Obama’s turnabout on gay marriage, his hand forced prematurely by Joe Biden, has reinforced those same apprehensions.
• Romney’s high school bullying is more proof for his critics that he is a spoiled rich guy with no empathy for those who are different or poor. As was once said about George H.W. Bush, Romney is characterized by his foes as someone who was born on third base and thinks he’s hit a triple. That the target of this “prank” was thought to be gay made it an irresistible contrast to Obama’s new gay marriage endorsement. Romney can expect more of this since Democrats plan to build part of their campaign around the Republican’s alleged inhumanity while heading up Bain Capital — which is also why Romney’s earlier comments during the presidential campaign about how he liked to fire people who do a bad job became another frenziette.

From a policy standpoint, President Obama’s embrace of gay marriage is not a minor issue: It is a notable shift that might have some bearing on how some voters will vote, even in an election that almost assuredly will be dominated by the economy. For instance, we can imagine that Obama’s decision will boost his fundraising and help him turn out the youth vote. It also might play well in some libertarian-leaning states, such as Colorado (where white evangelicals, probably the main group that will react negatively to the announcement, made up only 21% of the 2008 electorate, according to exit polls). Meanwhile, Obama might be hurt in states with bigger evangelical populations, such as two states he won in 2008: Indiana (43% evangelical in 2008 exit polling) and North Carolina (44%). For context, exit polls indicated that white evangelicals made up 26% of the national electorate in 2008, and they favored John McCain, 74% to Obama’s 24%.

But the ephemeral stories wrapped around Obama’s announcement — Joe Biden’s forcing Obama’s hand by announcing his support for gay marriage a few days beforehand, and the revelation about Romney as a high school bully — probably do not matter much. Biden put his boss in a box by undercutting him and forcing him, reportedly, to make his announcement before he wanted to. But barring a switch in Veeps prompted by Biden’s goofs — a possibility we see as nearly inconceivable at this point — Obama’s actual change in opinion is the only thing that is important.

Timing is everything

Some elections are defined by major issues and profound disagreements about policy. For instance, 2004 was largely about the Iraq war and 2008 was about the economic collapse. The Iran hostage crisis and stagflation defined 1980. This election, as we’ve mentioned, is likely to be dominated by the slow recovery.

But not all elections are that way. The election of 1988 was largely about social issues such as crime. The most memorable artifact of that election was the Willie Horton ad, which stirred passions about race, law and order, and punishment. A dozen years later, the campaign between George W. Bush and Al Gore was considered so vapid it was called the “Seinfeld election”– an election about nothing.

In 2000 Gore’s “frenziettes” received substantial negative attention: his kissing now ex-wife Tipper at the Democratic convention, his “creating the Internet” and his odd sighing and creeping up on Bush during a debate, among other things. Meanwhile, Bush suffered a full-blown frenzy just days before the election, when reports surfaced of a drunk-driving arrest from the 1970s. Again, this played into the existing subtext some reporters had about Bush — that he was an unserious partier who had simply capitalized on the family name — and, because of the timing, it’s possible that the revelation actually cost him many votes.

Bush strategist Karl Rove claimed, in the days before the election — and the DUI story — that his candidate was on track to win 320 electoral votes. Obviously, Bush ended up losing the popular vote, and he barely won the presidency (271 electoral votes) after a legal battle over the razor-thin Florida result. Bush campaign officials believed that the DUI story undercut Bush’s main anti-Clinton campaign theme of “honor and integrity” and depressed turnout among evangelical Christians, possibly costing Bush the popular vote. Whatever the truth of that, the disclosure just before Election Day had to be damaging. If the story had surfaced in, say, July, it probably would have been forgotten by November. Former Bush spokesman Scott McClellan was correct when he argued in his memoir that Bush should have disclosed the drunk-driving incident earlier in the campaign so that he could have discussed it on his own terms.

Returning to an earlier example, Ford’s gaffe about Eastern Europe in 1976 clearly hurt him in his close race with Carter, given that the post-debate media frenzy changed viewers’ first polled reactions to that debate: They initially thought Ford had won, but days later gave the debate to Carter after the negative press coverage of Ford’s gaffe. The controversy went on and on, partly because of Ford’s stubbornness in refusing to admit he had misspoken. Given the closeness of the election — Ford would have won if he had flipped roughly 18,000 votes in Wisconsin and 6,000 in Ohio (both places with significant populations of voters with Eastern European lineages) — the gaffe might have cost Ford a full term as president.

Conclusion

One of the reasons why campaigns can be so boring and scripted, with candidates rarely saying anything interesting on the trail, is that they are desperate to avoid even a “frenziette.” Similarly, in frustration with the buttoned-up campaigns, journalists will jump on any little slip-up and report it as a gaffe. Call it a frenzied Catch-22 of the contemporary campaign.

Media storms and squalls will inevitably burst in the next few months, even though the only candidate-initiated move that might change the electoral calculus before the conventions is Romney’s selection of a vice president. Even that big announcement may not move the electoral needle very much, though we all recall the list of modern VP candidates who generated major controversies and full-fledged frenzies: Spiro Agnew (1968), Thomas Eagleton (1972), Geraldine Ferraro (1984), Dan Quayle (1988) and Sarah Palin (2008).

Therefore, if you like significant media frenzies that can have an impact on the election, you may have to wait for Mitt Romney’s running-mate unveiling. Until then, you’ll have plenty of frenziettes to occupy your time and attention.

But which state will put either Barack Obama or Mitt Romney over the top? In other words, which state will actually prove to be the decisive domino in the race to 270 electoral votes?

A simple yet meaningful way to look for the decisive state is to order the states according to the winner’s margin of victory. By arranging the states in this way, we can find the cut-off point where the winning candidate crosses the 270 electoral vote threshold. The state that gives a candidate a majority of the 538 votes in the Electoral College is the decisive state.

In 2008, that state was Colorado. Obama won the Centennial State by about nine percentage points over John McCain; had Obama failed to carry Colorado and every state that he won by a smaller margin, the Electoral College result would have been a 269-269 tie, a result that would have thrown the election into the House of Representatives. (Actually, the one electoral vote Obama won in the Second District of Nebraska would have given him a 270-268 victory, but our analysis here is state-based.) By carrying Colorado’s nine electoral votes, Obama clinched his victory.

Considering that the final tally in 2008 was Obama 365, McCain 173, it is easy to overlook the concept of the decisive state. In a blowout win, such as an FDR or Reagan landslide, the margin of victory in the decisive state can be nearly 20 percentage points. However, in a too-close-to-call election, that margin might be tenths, even hundredths, of a percentage point. Just look at 2000, when George W. Bush defeated Al Gore by .01% of the vote in Florida to reach 271 electoral votes. Flip Florida and Gore wins.

Examining the decisive states in recent elections naturally prompts the question: Which state has been decisive most often in American presidential elections? In the spirit of the 2000 election, it’s a tie, at least if we start our analysis with the first Democrat-versus-Republican contest in 1856. The chart below details the deciding state in each election, while the map illustrates which states have been most decisive. Keep in mind that the electoral vote count has changed over time as states were added and populations have shifted; the current total, 538, was not reached until 1964, when the District of Columbia first cast its three electoral votes.

Chart 1: Decisive states in presidential elections, 1856-2008

Map 1: The most decisive states in the Electoral College, 1856-2008

It may surprise contemporary election junkies that New York has been one of the two most decisive states in history. Yet it will shock no one that Ohio is the other. Both states have clinched election victories six times, accounting for 12 of the 39 elections between 1856 and 2008. Ohio did so most recently and memorably in the close 2004 contest. For Richard Nixon, Ohio proved to be a good friend — it went for him in his 1960 loss and ensured him victory in both 1968 and 1972. As for New York, the last time the Empire State put a winner over the top was in 1944, when its then-47 electoral votes guaranteed Franklin Roosevelt’s final presidential triumph. But New York’s heyday as the nation’s pivotal swing state really came in the late 19th century, when it decided three straight closely-fought elections (1880, 1884, and 1888).

The Land of Lincoln is the next most-decisive state, with five wins to its name. Appropriately, Illinois’ electoral votes were the deciding factor for Abraham Lincoln’s reelection in 1864. Next is three-timer Michigan, which put Ronald Reagan and George H.W. Bush over the top in 1984 and 1988, respectively. Four states have proven decisive twice (California, Florida, Pennsylvania and Tennessee), and 11 have done so once. The smallest state to settle an election is Rhode Island, whose then-five electoral votes certified Warren G. Harding’s win in his 1920 landslide victory.

What state will prove decisive in 2012? Ohio is always in contention, but so might be Colorado, Iowa, New Hampshire, or several other Purple swing states. Given that Virginia was the state closest to the national average in 2008 (Obama took 52.9% nationally and 52.6% in the Old Dominion), the Crystal Ball’s home state could be the crucial, final domino for one party or the other in November — which would be its first turn as the decisive state in the Democrat-versus-Republican era.