Ahead of the U.S. presidential election in November, numerous public opinion polls show presumptive Democratic presidential nominee Joe Biden leading President Donald Trump.
While polls can reflect how popular a candidate is at a point in time, they don’t always accurately predict the election result.
VOA spoke to experts Allan Lichtman and Helmut Norpoth, both of whom have called the outcome of the election based on their own prediction models. Here are their verdicts, which both scholars declared as “final.”
Professor of History at American University, Washington, DC.
Lichtman has correctly predicted all presidential election results since 1984. In 2000, he forecast that Al Gore would win the election, and stands by that prediction. Gore won the popular vote but lost the presidency to George W. Bush after the Supreme Court ruled to stop the recount after a long dispute over inconclusive ballots cast in Florida. Lichtman has since adjusted his metrics to call the candidate with the most electoral votes, not the candidate with the most popular votes. Lichtman also predicted Trump’s impeachment.
Prediction: Biden wins.
Methodology: Keys to the White House
Lichtman uses a series of 13 “keys” in the form of true or false questions. A "true" answer earns a point for the incumbent, while a "false" answer earns a point for the challenger. The keys predict that the candidate with the most points will win the election.
Those keys and their answers for the 2020 race, according to Lichtman are:
1. The incumbent’s party gained house seats between midterm elections – FALSE
2. There is no primary contest for the incumbent’s party – TRUE
3. The incumbent is running for reelection – TRUE
4. There is no third-party challenger – TRUE
5. The short-term economy is strong – FALSE
6. The long-term economic growth during the incumbent’s term has been as good as the past two terms – FALSE
7. The incumbent has made major changes to national policy – TRUE
8. There is no social unrest during the incumbent’s term – FALSE
9. The incumbent is untainted by scandal – FALSE
10. The incumbent has no major foreign or military failures abroad – TRUE
11. The incumbent has a major foreign or military success abroad – FALSE
12. The incumbent is charismatic – FALSE
13. The challenger is uncharismatic – TRUE
Total: Incumbent 6 points, Challenger 7 points.
Lichtman said his prediction has changed after the pandemic and the widespread social unrest following the death of George Floyd while in police custody in May.
“In just a matter of a few months, Donald Trump and the Republicans went from what looked like a sure win with just four keys against them, to a predicted loss with 7 keys—one more than needed to predict their defeat,” he said.
Lichtman said he will not change his prediction again but there are two factors that lie outside the realm of the keys: voter suppression and election meddling.
“The Republican base is old white guys like me—that is the most shrinking part of the electorate,” said Lichtman, a registered Democrat. “The GOP cannot manufacture new old white guys but what they can try to do is suppress the vote of the rising Democratic base of minorities and young people. That has me worried.”
Another concern for Lichtman is election intervention by foreign actors.
“We know the Russians will be back, and maybe back in more force because they've learned a lot since 2016,” Lichtman said. “And we know for certain that Donald Trump will again welcome and exploit any Russian intervention that he thinks will help him win.”
Professor of Political Science at State University of New York at Stony Brook.
Norpoth correctly predicted five of the past six presidential elections since developing his model in 1992. When applied to previous elections, Norpoth’s model correctly predicted the last 27 elections except for the 2000 election in which George W. Bush defeated Al Gore and the 1960 election in which John F. Kennedy defeated Richard Nixon.
In March of 2016, Norpoth predicted Trump having an 87 percent chance of winning.
Methodology: The Primary Model
Norpoth’s Primary Model uses statistical representation of U.S. presidential races with one key metric—the importance of early presidential primaries.
A state-level election, a primary is usually held in February of a presidential election year, where voters choose who would be the political party’s nominee to run in the November presidential election. New Hampshire and South Carolina hold the first primaries for both the Democratic and Republican parties.
The model uses data going back to 1912 when presidential primaries were first introduced and concludes that the candidate with the better primary vote tends to win the general election.
Joe Biden won 8.4 percent of votes in the New Hampshire primary and 48.4 percent in South Carolina. Trump won the Republican primary in New Hampshire with 85.6 percent votes. There was no primary election for Republicans in South Carolina this year.
“South Carolina canceled the Republican primary so I have no number for that. But they canceled it for lack of competition, so it probably would be 100 percent or something close to that,” said Norpoth, a registered Independent. “Either way, in those two primaries Donald Trump gets a vastly higher score than Joe Biden, so that puts him in the driver's seat as far as the primary part of the model is concerned.”
Norpoth’s model also factors in what he calls the “swing of the electoral pendulum,” the theory that control of the White House swings from one party to the other in presidential elections, on average after two to three terms.
Trump wins. Norpoth concluded the president has a 91 percent chance of reelection and Biden has a 9 percent chance of winning.
Norpoth is not predicting whether Trump will win or lose the popular vote this year but projected that Trump will gain 363 electoral votes while Biden will gain 175 electoral votes.
No caveat. Despite most polls showing Biden in the lead, Norpoth declared that his forecast is “unconditional and final.”
“We're living in an age of cancel culture, woke politics, etc.," he said. "Maybe some people are reluctant to admit even to a pollster that they're supporting Donald Trump because it doesn't sound right, it doesn't sit right with a lot of people.”
Norpoth said neither the pandemic nor Black Lives Matter protests had any bearing on his projection and insisted that a sitting president with a superior performance in the primaries compared to the opponent has never lost.
“It’s written in stone,” Norpoth said. “It cannot bend, but it may break. In the end, there's a chance—9 percent—that it's going to come out wrong so that’s a chance I’m taking.”