lmost two years removed from the biggest blown call of his long life as one of America’s most famous presidential-election prognosticators, Allan Lichtman is still looking ahead. It’s late July and Lichtman, an American University professor, is sitting in his house talking to a YouTube au dience about the upcoming midterms. Lichtman’s son, Sam, cohosts the show. He mentions a map that has Texas leaning red. “I think that’s wrong, ” Lichtman says. When Lichtman prognosticates, which is often, his arms move. A lot. The ideas that start in his brain travel to his face and mouth, then end up in his fingers, which punctuate whatever he’s saying—sometimes hovering by his head, jabbing forward to phrases like very important or tight race, and other times pointing directly at the camera for unsettlingly long stretches. If he has doubts, they are deep down, somewhere else. Sam pulls up the Kalshi odds on the Texas Senate race between Republican Ken Paxton and Democrat James Talarico. At the time, Paxton leads, 53 percent to 47 percent. Lichtman is unfazed. “Dead heat,” he says. “Yeah, but that still leans red, right?” Sam asks. “I wouldn’t say so,” Lichtman says. Maybe in an ordinary election year, he explains. And maybe in Texas, where Democrats haven’t won statewide office in decades. But not now. During Donald Trump’s second presidential term, Lichtman continues, Democrats have been way outperforming the polls. “You cannot judge Democratic performance by the polls,” he says. Lichtman’s self-assurance isn’t surprising. At age 79, he’s written a stack of books—multiple histories of American conservatism, arguments for impeaching Trump and repealing the Second Amendment—and been an expert witness in more than a hundred civil-rights cases. He once had a 17-day winning streak on the TV quiz show Tic-Tac-Dough. By his own account, Lichtman has made more than 1,000 media appearances, mostly to discuss “the Keys to the White House,” an election-forecasting system he co-created in 1981 and used to correctly predict the winner of nearly every presidential race for the next 40 years. Then came 2024. Early that September, Lichtman starred in a splashy New York Times op-ed video, topped by a headline dubbing him “the prophet of presidential elections.” He predicted that Kamala Harris would win. She did not. This was no small deal for Lichtman, who for years had publicly sparred with other election forecasters over the legitimacy of his methodology. Until Harris lost, the prophet could always point to the scoreboard. Afterward, when Washingtonian reached out to ask what happened, Lichtman did something out of character: He declined an interview, saying he was going to take time away to figure out what had gone awry. Today, Lichtman says, he has some rough ideas. But nothing definitive. Meanwhile, he hasn’t exactly gone into self-imposed exile, instead making episodes of “Lichtman Live” for his YouTube channel, which has 182,000 subscribers. Many of his videos have modest, four-digit view counts—but when the word “prediction” appears in the title or thumbnail, interest rises. The top five most-watched videos on his channel are about his predictions. Number one, which has more than 1 million views, is titled “Allan Lichtman’s Predictions for 2025: You’ll never guess number 8!!” Maybe this strikes you as sheer chutzpah—the weatherman who missed last week’s hurricane telling you about next week’s sunny skies. And maybe it is. Still, Lichtman has an audience: 15,000 followers on Bluesky and 97,000 on X. More than 50,000 viewers watched him pontificate on Paxton versus Talarico. And he’s hardly alone. Everywhere you look in American politics, there are oracles just like him: pollsters and modelers, strategists and pundits, yard-sign counters and oversize-touchscreen jabbers, all suggesting that the future is knowable. Predictions are what everybody wants, even when they turn out wrong. Lichtman has always been brimming with ideas. He was a bright kid from Brooklyn, a bookworm his siblings called the family’s “golden child.” His parents wanted him to become a doctor. In the early 1960s, he rode the New York City subway from Brownsville an hour each way to attend Stuyvesant, the prestigious and selective STEM-focused high school in Manhattan. One day, Lichtman recalls, a guest lecturer discussed an experiment in which DNA from fuzzy bacteria was injected into smooth bacteria, and the smooth bacteria turned fuzzy. A thought began making its way to Lichtman’s fingertips: Why not take DNA from normal cells and inject it into cancerous cells to cure cancer? “Of course, nobody would listen to me, a 14-year-old kid,” he says. “I tried to tell everyone about it. Even when I got to college, I tried to tell people, and no one would listen. “And of course, decades later, gene therapy.” Skipping two grades, Lichtman went to college at Brandeis at age 16. Realizing he didn’t have the stomach for hospital work, he switched his major from premed to history. He then got his PhD at Harvard, where he coached the school’s debate team. In 1981, Lichtman was a visiting scholar at CalTech when Volodia Kellis-Borok, a Soviet geophysicist, asked him to collaborate on an unusual idea: applying the logic of earthquake prediction to elections. Kellis-Borok needed an expert in American history and politics to help him flesh things out. Together, they created the Keys, which conceptualized presidential races not as Candidate A versus Candidate B but as stability versus quake—the status quo versus change. Every election, the system posits, is essentially a referendum on the strength and performance of the incumbent candidate or party. Lichtman argues that most of what we focus on in the lead-up to elections—“the ads, the speeches, the debates, the dirty tricks”—is irrelevant horse-race nonsense. What matters, he says, are the fundamentals of governance. Lichtman’s forecasting system is rooted in historical study—and, he freely admits, his judgment. The Keys reflect this. Drawn from studying every presidential race between 1860 and 1980, they consist of 13 true-or-false items such as whether the economy is in recession and whether the incumbent has been tainted by a major scandal or military failure. If five or fewer of the Keys favor the challenger, the incumbent is predicted to retain power. Otherwise, it’s throw the bum out. Some of the Keys are objective—recessions are a measurable, yes/no proposition—but many are not. For instance, two Keys deal with charisma, asking if either candidate is “charismatic or a national hero.” Would Trump qualify? It depends on who you ask. (For the record, Lichtman does not consider Trump sufficiently charismatic, because his appeal doesn’t broadly translate across party lines.) In his books, Lichtman does a lot of work to define the situations in which the Keys favor or disfavor a challenger or incumbent. But he’s also quite open about the fact that the whole system relies on discretion. His discretion. “When I first developed the system, I was blasted—and I still am, remarkably—by the conventional political scientists and modelers,” Lichtman says. “I had committed the ultimate sin of forecasting, the sin of subjectivity. You call it subjectivity, I call it judgment.” Lichtman and Kellis-Borok published an academic paper on the Keys. They figured maybe five people would read it—but one day, Lichtman says, he opened a newspaper to find a story about their system. “It turns out six people read our article,” he says. This gave Lichtman another idea: My gosh, I can go pop. He sent the paper to Washingtonian and got a call back from an editor named Ken DeCell. DeCell had a simple question. Could Lichtman predict who would win the next presidential race? The election was three years away. “I knew if I gave him the proper academic answer—probability, caveat, not yet—my pop career would be over in ten seconds,” Lichtman says. Lichtman swallowed hard. “Yes, I can,” he said. As The Keys Turn Allan Lichtman uses 13 true/false “Keys” to forecast presidential races. If six or more are false, the incumbent is predicted to lose. Party Mandate The incumbent party holds more House seats following the midterms than after the previous midterms. Contest There’s no serious contest for the incumbent-party nomination. Contest There’s no serious contest for the incumbent-party nomination. Incumbency The incumbent-party candidate is the sitting President. Third Party There’s no significant third-party or independent campaign. Short-Term Economy The economy is not in recession during the campaign. Long-Term Economy Real per capita growth during the incumbent’s term equals or exceeds mean growth during the previous two terms. Policy Change The incumbent effects major changes in national policy. Social Unrest There’s no sustained unrest during the incumbent’s term. Scandal The incumbent is untainted by major scandal. Foreign/Military Failure The incumbent suffers no major failure in either. Incumbent Charisma The incumbent is charismatic or a national hero. Foreign/Military Success The incumbent achieves a major success in either. Challenger Charisma The challenger is not charismatic or a national hero. The next spring, Lichtman used the Keys to make his first major public prediction, right here in the pages of Washingtonian: According to the article “How to Bet in ’84,” President Ronald Reagan would remain in office. At the time, a second term was no sure thing. The economy was in a recession, the worst since the Great Depression. Unemployment was high. Reagan’s approval ratings were falling, and his age—almost 70 when he was inaugurated—was a real concern among voters. Still, Lichtman called his shot. It brought him attention from places he hadn’t expected, most notably the White House. Lichtman says he was summoned there by Reagan adviser and Republican strategist Lee Atwater, who wanted to know what would happen if the President declined to run for reelection. Lichtman broke it down. The GOP already had lost three Keys—if they replaced Reagan at the top of the ticket with Vice President George H.W. Bush, they would lose three more, including charisma. Game over. Atwater, Lichtman says, breathed a sigh of relief. Thank you, Professor. Reagan stayed in. In the 1984 election, he won 49 states. Lichtman’s star began to rise—and his predictions kept hitting. In the spring of 1988, he predicted that Bush would maintain his party’s hold on the Oval Office, even though polls and pundits at that moment had him losing to challenger Michael Dukakis. Three years later, Lichtman said that Bush would be a one-and-done President, despite his sky-high approval ratings in the wake of the Gulf War. That prediction, Lichtman says, prompted a phone call from Kay Goss, an assistant to then–Arkansas governor Bill Clinton. Other Democrats were reluctant to challenge a popular incumbent. Goss wanted to know: Was Bush actually beatable? Yes, Lichtman said. He then sent Clinton “a memo and a copy” of the book he wrote with DeCell, The 13 Keys to the Presidency: Prediction Without Risk. “And the rest is history,” Lichtman says. Election forecasting has long been a part of American politics—presidential straw polls date back to 1824. The modern era of system-based prognostication owes much to Louis Bean, an economist whose 1948 book, How to Predict Elections, defied prevailing vibes by correctly forecasting that President Harry Truman would defeat challenger Thomas Dewey. (Like Lichtman, Bean eschewed polling in favor of factors such as the state of the economy and the strength of third-party candidates. Also like Lichtman, he was dubbed a “prophet,” in this case by Life magazine.) In the decades since Lichtman helped create the Keys, an entire cottage industry has sprung up around political predictions, feeding our seemingly insatiable appetite to know tomorrow’s results today. That hunger has made predicting a regular feature on cable news—sometimes, specific predictions are the news. It’s made mini-celebrities out of analysts such as Steve Kornacki and Nate Silver. It’s been good to Lichtman, too. He’s published two books on the Keys, spoken at forecasting conferences and to the CIA, and done a Reddit AMA. Every election season, he pops up in the press like Punxsutawney Phil on Groundhog Day. Even his second date with his now-wife of 35 years, Karyn Strickler, was at a party for his first book, held at Politics and Prose. Strickler, who calls Lichtman “one of the smartest people that’s ever lived,” says her husband has always wanted to be famous. Sam, their son, concurs. Lichtman doesn’t exactly disagree—“I guess,” he says—but clarifies that he never wanted to be “Kim Kardashian famous. I want to be famous for doing the right thing.” An unabashed progressive, Lichtman is proud of his extensive expert-witness work in civil-rights cases, most involving voting, and especially proud of a 2001 report he authored for the US Commission on Civil Rights, arguing that Black voters in Florida disproportionately had their ballots discarded in the 2000 presidential election—something that Lichtman later argued allowed George W. Bush to win a razor-tight election against Al Gore. “Yes, I like the fact that the Keys attract attention,” he says. “But I don’t consider that anywhere close to my full legacy.” This doesn’t mean Lichtman isn’t proud of the Keys—or of being right. At one point during an interview for this story, he held up a framed printout of a Washington Post article from 2016, which included his prediction for that year’s election. The printout is marked up with black Sharpie ink: PROFESSOR—CONGRATS—GOOD CALL Below that is Trump’s distinctive signature. Yep, it’s true. Six weeks before the 2016 election, when most polls and forecasts projected a Hillary Clinton victory, Lichtman said she would lose. Lichtman wasn’t happy about his prediction. He considered Trump a “history-smashing” candidate, and not in a good way: scandal-ridden, lacking political experience, possessing questionable character, a man who, as he told NPR, brought out “the worst elements in our society—the Ku Klux Klan, the neo-Nazis, the white supremacists.” (In the same Post interview containing Lichtman’s forecast, he also predicted Trump would be impeached, making him two-for-two.) During this time, Lichtman had trouble sleeping. He says that good friends and liberal academic colleagues alike pressured him to change his mind. In the end, he held fast. The Keys were the Keys, and too many had turned against the Democratic nominee. “Until he does his prediction, he’s not always sure which way it’s going to go,” says Max Friedman, one of Lichtman’s colleagues in AU’s history department. “And then he applies his method, and then he has confidence in the system.” Lichtman had taken big swings before. Two years before the 2008 election—before even knowing who the Democratic and Republican nominees would be—he correctly forecast that the GOP would fail to hold the White House. Similarly, he predicted in January 2010 that Barack Obama would win reelection. But this one was different. Trump’s victory was a national, even global shock. “Some of [Lichtman’s] friends weren’t talking to him anymore,” Friedman says. “They said it was his fault.” Lichtman’s clairvoyant zag brought a new level of attention: a mention in a Maureen Dowd column in the Times, blog after blog after blog on the Post’s website, and—before the 2020 election—an appearance on Bill Maher’s Real Time. It also brought delayed scrutiny. Two summers ago, AU alums Michael Lovito and Lars Emerson published an article on the website the Postrider that accused Lichtman of not being “completely honest” about his 2016 prediction. How so? The Keys, they wrote, had long forecast the winner of the popular vote, not the Electoral College. Trump won only the latter, which meant Lichtman was actually wrong. A war of words ensued—claims and counterclaims, debunkings and rebuttals, some of it excruciatingly pedantic—and Lichtman at one point called the article “defamatory” in an email that he copied to both a lawyer friend he often works with on civil-rights cases and the school’s general counsel. (Lichtman did not end up suing the duo.) Today, the whole episode still seems to sting. When I brought it up with Lichtman, he got upset for the first and only time in our conversations, telling me not to write about it. Lichtman has other critics. Contemporary electoral forecasting has largely moved away from systems like the Keys and toward the kind of data-driven work made famous by Silver, which aggregates and weights various public-opinion polls (and sometimes other numbers, such as economic indicators) to determine the probability that a particular candidate will win. Proponents of this approach can be unkind to the Keys. Silver, who did not respond to a Washingtonian interview request, once wrote of Lichtman that “it’s less that he has discovered the right set of keys than that he’s a locksmith and can keep minting new keys until he happens to open all 38 doors.” Data journalist G. Elliot Morris, a former editor of the FiveThirtyEight website created by Silver, responded to an interview request for this story with a one-line email: “If Allan Lichtman is an election forecaster, I’m a toaster.” The disdain cuts both ways. Lichtman has called Silver a “compiler of polls, a clerk” who lacks historical knowledge. He rankles at hedging, probabilistic pre-dictions— Donald Trump has a 29-percent chance of winning —dismissing them as intellectual cowardice. Silver, Lichtman says, “can never be wrong, and he can never be right, because no matter what happens, he’s always got this escape hatch, right? I don’t do that. I go out on a limb, and that’s what makes my predictions so gut-wrenching.” Lichtman likes to say that the Keys are unbiased and that his own political preferences play no part in his predictions: If he was just a wish-casting progressive, he’d have picked Clinton in 2016, or John Kerry in 2004. Still, the system is intensely personal. The Keys turn, Lichtman says, when they turn—sometimes years before an election, sometimes just weeks. To know when that’s happened, you have to wait for Lichtman to tell you. Small wonder, then, that he seems to process attacks on the Keys as attacks on him. If the system is wrong, if his predictions are wrong, then he’s wrong, at a deep and discomforting level. “Being right and being smartest has been the foundation of all of his success,” says Sam, Lichtman’s son. “So maybe there’s a fear around that. If that [comes into] jeopardy, he’ll lose his success. So he needs to try to cling to that in some way.” Sam tells a story. Years ago, his uncle Steve, Lichtman’s brother, brought up the 2000 election. Back then, Lichtman had predicted a Gore victory, and the Democratic candidate won the popular vote—but lost to Bush in the Electoral College. Lichtman insisted he’d made the right call, because the Keys were designed to forecast popular-vote winners. In 2016, by contrast, Lichtman picked Trump—who, like Bush, won the Electoral College but lost the popular vote. When, Steve asked, did the Keys stop being about the popular vote? And why? “I remember my dad got really angry,” Sam says. When I later mention this to Lichtman, he’s uncharacteristically demure, saying he doesn’t really recall the conversation with his brother. Maybe, Strickler suggests, it had just been about the 2000 election. Lichtman’s son says that public anger at his father’s prediction that Kamala Harris would win in 2024 left Lichtman with a sense of “I’m just a professor, and I’m just making predictions. Why do people want to kill me?” Lichtman has always gotten butterflies in his stomach before announcing his presidential picks—but in 2024, he says, he had “crows.” From his perspective as a historian, the race was unprecedented. The incumbent had dropped out just a few months before Election Day. Still, he had to make a prediction. That’s what everyone expected. So he did. Proclaiming the Keys to be the “constant north star of political prediction,” he chose Harris. And he was wrong. Lichtman admits he was wrong. You have to give him credit for that. He admitted it on Election Night, during a livestream with his son, just before 1 am. It didn’t feel good, being wrong, and some viewers gleefully piled on, brigading into the comments of the YouTube show. Bro thinks he’s nostradamus, turned out he’s just a clown The dude lost his keys and dignity in the same night It’s pretty sad to see your life’s work being proven as just junk science Allan Lichtman YOU’RE FIRED Only that wasn’t the worst of it. The worst came earlier, right after Lichtman appeared in the Times video announcing his prediction. He started getting hate mail. Death threats. He’s been doing this for years. He’d never seen that before. Then he was doxxed. People started showing up at his house, banging on the door late at night. What did they want? Lichtman and Strickler became well acquainted with local police. Their phones started ringing all the time—somehow displaying the caller IDs of family members, even when everybody was in the same room. Strickler began having nightmares about being in an emergency and not being able to dial 911. One night, she says, that’s exactly what happened. Someone was at the door, and when she took Lichtman’s phone and tried to call for help, “someone had taken control of Allan’s phone and I couldn’t get out to 911.” For decades, the family had left their doors unlocked. When I visited their house in early September, right as I walked in, they noted the security system they had installed after the election. Lichtman, says his son Sam, “didn’t even know this hate existed. And I think that made him sad and kind of scared. It was kind of sad to watch, honestly. It was kind of a blow to the gut to him. I think he also had a sense of ‘I’m just a professor, and I’m just making predictions. Why do people want to kill me?’ ” People mocked Lichtman for being wrong. They were furious with him for picking Harris in the first place. If what we want from election prophets is to simply get it right—the way we want weather forecasters to let us know if we need to reschedule tomorrow’s tee time—then wishing bodily harm on an elderly history professor for the unforgivable sin of having an informed opinion, in public, seems like a bit much. But what if we really want something else? The Prophet’s Picks Since the 1984 campaign, Allan Lichtman has predicted the winner of every White House race. Here are his selections—and the results. 1984 Ronald Reagan 1988 George H.W. Bush 1992 Bill Clinton 1996 Bill Clinton 2000 Al Gore* 2004 George W. Bush 2008 Barack Obama 2012 Barack Obama 2016 Donald Trump* 2020 Joe Biden 2024 Kamala Harris *In 2000, Gore won the popular vote but lost to George W. Bush in the Electoral College. In 2016, Trump lost the popular vote but beat Hillary Clinton in the Electoral College. Elections feel increasingly consequential. Politicians tell us that each one—the next one—is the most important of our lifetime, with salvation or devastation just around the corner. And the next election is always undecided, always uncertain, right until the moment it isn’t. People hate that. They hate the waiting, the not knowing what’s next. In psychological studies, researchers have found that participants would rather receive an electric shock immediately than be told they may receive one at an unspecified time later. Forecasters offer a salve, a comforting illusion that the world works according to rules, with a future that can be foreseen. Just get the shock right now. During the 2016 race, Strickler recalls, an eight-year-old neighbor would repeatedly come over, asking Lichtman and Strickler what would happen. The girl was consumed with fear. “She wanted Allan to tell her that Trump was not going to win,” Strickler says. Lichtman couldn’t do that. Ever since Harris lost, Lichtman has been reflecting on the Keys. What did they get wrong? What did he get wrong? Maybe, Lichtman thinks, it’s disinformation. Maybe elections remain referendums on incumbent performance—on competent governance, not clever slogans—but voters are so flooded by lies and half-truths and propaganda that they can’t make sound judgments. He cites a pair of election-related polls on abortion. One shows a 30-point gap between Americans who identify as pro-choice and those who are pro-life. The other, on the question of which candidate would do a better job with the issue, shows Harris leading Trump—whose handpicked Supreme Court justices overturned Roe v. Wade —by just three points. Sam doesn’t buy his father’s reasoning: “I felt like his response to 2024 wasn’t good, and I’ve tried to be like, ‘Stop blaming the voters and stop the misinformation thing.’ It doesn’t really seem to penetrate.” Lichtman keeps moving. He published his latest book in September, a historical analysis of what makes great Presidents, well, great. (Yes, Keys are involved.) He still teaches—one of his former grad students, Rebecca DeWolfe, says he’s a brilliant lecturer, as good off the top of his head as he is with prepared notes. He does his live show. A longtime runner, he plans to compete in next year’s Senior Olympics and works out regularly, despite arthritis in both knees and a recent shoulder replacement. (His doctor, he says, told him to stop running ten years ago.) During one of our interviews, Lichtman answers questions while dressed in workout gear: a Senior Olympics tech shirt and a wrist brace on his left hand. After we finish, he says he’s going out for a two-mile run. I ask him why he’s still doing all this. “Sheer terror,” he says. Lichtman is afraid. Afraid of losing his edge. Of being irrelevant. Of aging, he says, “in a way that I can’t function.” He’s afraid of the future none of us can see but all of us know is coming. He figures if he just keeps teaching and running, doing what he’s always done, then perhaps he can “keep the terror off.” He’s thinking about tweaking the Keys and still expects to make a presidential prediction for 2028. I ask him what happens beyond that. “Does it survive without me?” he says. “I’m not sure.” In late September, Lichtman is back on YouTube, once again talking about the Texas Senate race. Unlike in July, some polls now show Talarico, the Democratic candidate, as a slight favorite. Lichtman’s hand leaps from his lap, jolting toward the camera. “We always have to be cautious about making predictions for a Democratic win in Texas,” he says with a smile. “No matter what it may look like.” I know that he’s not always right. I know that nobody is always right. There are political scientists who argue—quite convincingly, with math and data—that there’s no way to know if the entire field of election forecasting is any more prescient than flipping a coin. I check the polls and scan the news and read about the history and know that Talarico’s chances are good, relatively speaking, and that’s about all I can be sure of. Still, against my better instincts, I find myself hanging on Lichtman’s every word. I just want to know what’s going to happen. Now. Because I know the terror, too. Lichtman keeps talking. He won’t predict a Talarico win. But, he says, “it is more likely than not.” That’s not a bold statement. It’s basically what the polls and models and prediction markets are saying. It’s what guys like Silver would say. What’s striking, though, is the certainty in Lichtman’s voice. The note of control. The confidence about what the future holds. At that moment, it’s hard to determine who, exactly, he’s helping with this—his audience or himself. If I had to guess, I’d say both. This article appears in the November 2026 issue of Washingtonian.The post The Election Prophet Who Got It Wrong first appeared on Washingtonian. ...read more read less