In 1976, Washington insider Averell Harriman famously said of Georgia peanut farmer Jimmy Carter, the one-term governor and presidential aspirant, “He can’t be nominated, I don’t know him and I don’t know anyone who does.’’ Within months Jimmy Carter was president. Harriman’s predictive folly serves as an allegory of democratic politics. The unthinkable can happen, and when it does it becomes not only thinkable but natural, even commonplace. The many compelling elements of Carter’s unusual presidential quest remained shrouded from Harriman’s vision because they didn’t track with his particular experiences and political perceptions. Call it the Harriman syndrome.

The Harriman syndrome has been on full display during the presidential candidacy of Donald Trump. He couldn’t possibly get the Republican nomination. Too boorish. A political neophyte. No organization. No intellectual depth. A divisive character out of sync with Republicans’ true sensibilities. Then he got the nomination, and now those same perceptions are being trotted out to bolster the view that he can’t possibly become president. Besides, goes the conventional wisdom, demographic trends are impinging upon the Electoral College in ways that pretty much preclude any Republican from winning the presidency in our time.

But Trump actually can win, despite his gaffe-prone ways and his poor standing in the polls as the general-election campaign gets under way. I say this based upon my thesis, explored in my latest book (Where They Stand: The American Presidents in the Eyes of Voters and Historians), that presidential elections are largely referendums on the incumbent or incumbent party. If the incumbent’s record is adjudged by the electorate to be exemplary, it doesn’t matter who the challenger is or what he or she says or does. The incumbent wins. If that record is perceived as unacceptable, then again it doesn’t much matter who the challenger is or what he or she says or does. The incumbent or incumbent party loses.

We can never know what the electorate will do until it goes to the polls and unlocks the secret of its collective sentiment. But some political scientists have sought to parse the referendum concept through analytical frameworks that lay bare the essence of voters’ presidential decisionmaking. Of these, the most compelling was put forth by Allan J. Lichtman and Ken DeCell in their 1990 book, The 13 Keys to the Presidency. Lichtman and DeCell reject the notion that the electorate renders its presidential decisions based upon such things as negative ads, clever slogans, fund-raising disparities, campaign gaffes, or big-name endorsements. They believe, rather, that the voters, exercising their collective franchise, bring sound judgment to the task of choosing their leaders, that their decisions are based on big-picture considerations and not trivia, and that the country’s referendum guidance system has remained consistent through the country’s presidential history.

In this view, we have been looking in the wrong places as we assess the campaign. Instead of focusing on Trump and Democrat Hillary Clinton, we should be looking at the presidential performance of Barack Obama—and not his overall tenure but specifically his second-term record. Therein lie, in the Lichtman-DeCell framework, the levers of electoral outcomes.

The authors identify 13 “keys,’’ or fundamental analytical statements, that illuminate the political standing of the party in power. Assessing each presidential election since Lincoln’s 1860 victory, they note that when five or fewer of these statements prove false, the voters side with the incumbent. When six or more are false, the incumbent party gets tossed out. This analytical matrix seeks to apply to politics a set of “pattern recognition’’ algorithms designed to illuminate the politics of today by discerning patterns of circumstance that have guided the country’s political path through history.

In touting this analytical concept as a distillation of referendum politics in presidential elections, I also have noted its possible shortcomings as a predictive tool. Historical patterns, however finely calculated, could be broken at some point by intervening developments or events. And some of the so-called keys call for subjective judgments that inevitably generate disagreement and debate. For example, one of the keys seeks to assess whether the incumbent president has been beset by a major scandal. In their discussion of Ronald Reagan’s second term, the authors conclude that the Iran-Contra scandal was not of sufficient magnitude to turn that particular key against the Republicans. This is difficult to credit, given the long agony of congressional hearings, the career destruction of so many people near the president, and the persistent talk of impeachment or resignation. Finally, life generally, and certainly political life in particular, is too messy, chaotic, and ultimately unpredictable to be captured with utter consistency in any set of algorithms, however sophisticated.

But the Lichtman-DeCell thesis offers a far more accurate view of any campaign than we get from politicians, pundits, and campaign operatives preoccupied with “horse race’’ analyses focused on strategic and tactical maneuvers, clever debate ripostes, or positioning on issues. These are seen as the fundamental factors propelling one horse ahead in the race or retarding the progress of another. The index for assessing all this during the campaign is the public-opinion polls, which determine who’s up and who’s down at any given moment. Lichtman and DeCell reject this approach, particularly the view that early polls have any predictive value at all.

With all that in mind, the Lichtman-DeCell thesis, if applied as a general measure of presidential performance and the possible ballot-box consequences, offers a good starting point for assessing whether the incumbent party is likely to retain the White House or lose it in this election year. And the arbitrariness of some answers, by inviting debate, adds to the intrigue of the exercise. It’s something of a parlor game, so let’s play. But first let’s note that not all the statements (presented for true-or-false answers) are strictly about success levels. Some relate simply to the circumstances surrounding the election and don’t necessarily reflect on incumbent performance.

Yet, according to the thesis, they carry significant weight. So here goes:

Key 1 (Incumbent-Party Mandate): After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than it did after the previous midterm elections. This reflects the general political standing of the incumbent party. The answer for this year is false—unfavorable to the incumbent.

Key 2 (Nomination Contest): There is no serious contest for the incumbent-party nomination. Bernie Sanders, of course, has rendered this unfavorable to incumbent.

Key 3 (Incumbency): The incumbent-party candidate is the sitting president. This reflects a modest benefit for presidents seeking reelection. Unfavorable to incumbent.

Key 4 (Third Party): There is no significant third-party or independent campaign. With both major-party candidates logging unfavorable ratings exceeding 50 percent, Libertarian Party candidate Gary Johnson could emerge as a significant factor, and even Green Party candidate Jill Stein could have a material impact on the outcome. Recent RealClearPolitics poll averages put Johnson at 8 percent, while Stein polls at around 4 percent. Don’t discount the possibility that this could tilt against the incumbent party, but for now I’m scoring it as indeterminate.

Key 5 (Short-Term Economy): The economy is not in recession during the election campaign. It seems inconceivable that this could turn false. Favorable to incumbent.

Key 6 (Long-Term Economy): Real per-capita economic growth during the term equals or exceeds mean growth during the previous two terms. Favorable to incumbent.

Key 7 (Policy Change): The incumbent administration effects major changes in national policy. Obama has been stymied by the opposition Congress this term, but voters don’t care about the reasons. It’s difficult to attribute to him any significant policy change. I chalk it up as unfavorable to incumbent.

Key 8 (Social Unrest): There is no sustained social unrest during the term. This key denotes the voters’ collective inclination to hold government accountable when there is a lack of domestic tranquility. We’re talking here about blood in the streets, such as what emerged from the racial turmoil of the 1960s or the labor strife of the late 19th century and early 20th century. We haven’t had such riots in recent years (the Rodney King riots in Los Angeles are probably the latest). But the American people, in their collective judgment, may fault the incumbent party for the significant numbers of deaths from the Islamist terrorist attacks in San Bernardino and Orlando or the recent assassination attacks on police officers. This remains indeterminate, in part because we don’t have much precedent for such violence of this magnitude, and hence its political significance is difficult to assess. But it could tilt against the incumbent.

Key 9 (Scandal): The incumbent administration is untainted by major scandal. I’m putting this down as unfavorable to the incumbent on the thesis that FBI Director James Comey essentially elevated the Hillary Clinton email-server controversy to major-scandal status. The scandalous behavior occurred during the first Obama term, but the story broke—and hence the scandal emerged—during the second term. This may be debatable, but most political analysts perceive it as a net negative for the incumbent-party nominee, particularly since the scandal attaches to Hillary Clinton directly.

Key 10 (Foreign or Military Failure): The incumbent administration suffers no major failure in foreign or military affairs. The emergence of ISIS as an Islamist terrorist force that actually commands strategic Mideast territory constitutes such a foreign/military failure. Unfavorable to incumbent.

Key 11 (Foreign or Military Success): The incumbent administration achieves a major success in foreign or military affairs. I view the Iranian nuclear deal as a significant success, and I suspect the voters do too, notwithstanding the controversy that still surrounds it. Favorable to incumbent.

Key 12 (Incumbent Charisma): The incumbent-party candidate is charismatic or a national hero. No debate here: unfavorable to incumbent.

Key 13 (Challenger Charisma): The challenging-party candidate is not charismatic or a national hero. Whatever one says about Trump, it’s difficult to argue he lacks charisma. Unfavorable to incumbent. 

♦♦♦

Based on my scoring, the incumbent Obama administration carries the burden of eight unfavorable keys, with another two that could turn against it before Election Day. If there is any merit in the Lichtman-DeCell framework, then the Democratic Party stands today as ineligible for rehire by the American electorate. There will be no third consecutive Democratic term under Hillary Clinton.

Of course the keys could be wrong in this particular election year, or I could be wrong in my scoring of some that call for subjective assessments (most likely, keys 7, 9, and 10—policy change, scandal, and foreign/military failure). The fact that the keys track with every presidential outcome since 1860 doesn’t guarantee that they will track with this election—particularly given that the GOP nominee is unlike any previous nominee and carries plenty of political baggage of his own.

But, even when we pull back from applying the keys as a pure predictive matrix to be taken literally, they represent a prism on presidential politics through history that shows clearly that Democrats carry a significant burden this year. It is the burden of referendum politics mixed with a generally unimpressive incumbent record and other unfavorable political circumstances. The burden can be seen also in the RealClearPolitics poll average on the direction of the country—the so-called right-track/wrong-track question. Fully 63 percent of respondents now say the country is on the wrong track, while only 31 percent believe it is heading in the right direction. These are remarkable numbers, and they correspond precisely with the Lichtman-DeCell keys.

Or consider the 2012 presidential outcome. Much has been written about the supreme confidence of the Mitt Romney campaign, and Romney himself, that the Republican candidate would upend the Obama incumbency that year. The GOP political operative Karl Rove refused to believe it when Fox News projected Obama the winner on Election Night; as an on-air Fox News commentator that evening, he resisted with unyielding stubbornness the notion that the network was on solid ground in making its call. The network was right, and Rove was wrong.

But anyone following the Lichtman-DeCell keys harbored no doubt that Obama would be reelected. Only two or three keys, depending on the scoring, turned against him. It could be argued that some of his shortcomings as a leader hadn’t yet caught up with him by election time, and so he retained a relatively high standing in the eyes of voters. But since then, it seems, those shortcomings have manifested themselves in numerous problems, including a stalled domestic program, Mideast chaos, the ISIS threat, growing Islamist terrorism at home, intraparty frictions, and a lingering scandal.

The same happened to George W. Bush, whose leadership weaknesses didn’t yield the kinds of first-term problems that precluded reelection. But by the end of Bush’s second term the country could see that his cumulative leadership over eight years had been something of a disaster, with an Iraq quagmire and the country’s most severe economic downturn since the Great Depression. Following Bush’s second term, no Republican was going to beat any Democrat in 2008, irrespective of how the campaign unfolded. The referendum politics of presidential elections simply precluded that possibility.

Are we in a similar situation today? Difficult to say. And it would be a mistake to suggest that referendum politics is the end all and be all of every presidential election. Other considerations often come into play—the character of the candidates, the record of the challenger, the relative likability of the combatants. But incumbency performance is by far the most compelling factor. And, for those clinging to the notion that Donald Trump can’t possibly become president because of his rough personal traits, indecorous campaign rhetoric, lack of political experience, and outlandish political positions, a word of caution may be in order: beware the Harriman syndrome.

Robert W. Merry is author of books on American history and foreign policy, including Where They Stand: The American Presidents in the Eyes of Voters and Historians.