Letter: Allan Lichtman’s Response
On Wednesday, we published a piece concerning Allan Lichtman’s “13 Keys” model and the 2016 election. In the spirit of open discussion, we’ve decided to publish a response Professor Lichtman sent to us shortly after our piece was published:
Thank you for the opportunity to respond to your post about the Keys system and my 2016 prediction. I do not intend to respond point-by-point to what I believe is a misleading post. However, I will make a few salient points that should help you as aspiring forecasters.
First, things change. When I first developed the Keys, there was no daylight for nearly 100 years between the popular vote and the winner. However, the popular vote has become irrelevant in recent years because of the many millions of Democratic votes from N.Y. and C.A. with no comparable votes from Red states. That is why, like other modelers (e.g., Helmut Norpoth), I only called the winner in 2016.
Second, critically, you do not quote what I said in the actual Washington Post interview on September 23, 2016, where I called Trump’s win. I said nothing about the popular vote but made it crystal clear that I called Trump to win. Rather, I said, “Based on the 13 keys, it would predict a Donald Trump victory… But I would say, more to the point, they point to a generic Republican victory.” Also, in the full video, at the same time, I predicted that Trump would be impeached, which made no sense if I were talking about the popular vote.
Third, you do not review the response to my 2016 call by authorities in the field, who recognized my call for a Trump win against all the conventional wisdom:
- Gerald M. Pomper, arguably the Dean of American political science: “In 2016, nine of eleven major studies predicted Clinton’s lead in the national popular vote. However, by neglecting the Electoral College and variations among the state votes, they generally failed to predict Trump’s victory. One scholar did continue his perfect record of election predictions, using simpler evaluations of the historical setting (Lichtman 2016).” Pomper, “The Presidential Election: A Troubled Democracy,” in William J. Crotty, Winning The Presidency 2016, p. 76.
- Dennis W. Johnson, Professor Emeritus of Political Management, George Washington University: “Historian Allan J. Lichtman, author of Predicting the Next President: The Keys to the White House 2016, has correctly predicted the presidential election during the last thirty years, including this one.” Johnson, THE ELECTION OF 2016, in Johnson and Lara M. Brown, CAMPAIGNING FOR PRESIDENT 2016, fn. 107.
- Patrick Horst, Institute for Political Science and Sociology, Bonn, Lichtman “was in a few to forecast Trump’s victory over Hillary Clinton.” Horst, The Politics Of Removal: The Impeachment of President in Michael T. Oswald, Mobilization, Representation, and Responsiveness in the American Democracy, P. 90
- John A. Tures, Professor of Political Science, LaGrange College, “Allan Lichtman was another star political forecaster, who called the 2016 election for Trump in September 2016,” https://theconversation.com/political-forecast-models-arent-necessarily-more- accurate-than-polls-or-the-weather-144266.
- Shankar Vedantam NPR’s resident science authority: “On Sept. 23, Lichtman, a historian at American University, declared that Trump would win, and he stuck by that call through the tumultuous final weeks of the campaign.” https://www.npr.org/2016/11/15/502074201/why-polls-predicted-a-hillary-clinton- win-and-were-so-wrong-about-the-election
I was also co-winner with Professor Norpoth of the Stekler Award for Courage in Forecasting.
Fourth, you misunderstood the Keys. I never said that Clinton’s failure to gain 2/3 or more of the delegate vote, the definition of the Contest Key, was canceled by Sanders’ endorsement. As I noted in my 2016 book, Clinton would have to “win two-thirds of the delegates and secure Contest Key 2.” The turning of this Key against Clinton and other relevant matters are fully explicated in my latest book, which I referred you to and which you did not consult before writing your post. And I have never hedged on the Keys in 2016 or any other contest.
Fifth, it is arguable that if I was right about 2016, I was wrong about 2000. However, I did not just assert that 2000 was a stolen election. I proved this in my 2001 report to the U.S. Commission on Civil Rights, which is still available on their website. A subsequent independent analysis by Professor Walter Mebane of Cornell University verified my findings. His study proved that Gore should have won Florida and the presidency by tens of thousands of votes (“The Wrong Man is President!” Symposium: U.S. Elections, September 2004).
Sixth, you disparage my calls relative to the polls. Yet your own data shows that I made calls contrary to the polls in at least two elections. In addition, while Reagan may have been a popular president in 1984, he was not in April 1982 when I made my call; his approval rating was 43%. Likewise, when I called for Obama’s reelection in March 2010, his approval rating was 48%.
Seventh, You misunderstand academic freedom. American University does not supervise my work. I alone am responsible for my work.
We thank Allan Lichtman for his response. We stand by our reporting and the following Editor’s Note.
Though some of our piece criticizes the “13 Keys” system substantively and Professor Lichtman has responded with counterarguments to some of these concerns, the thrust of our piece is based on two major claims which he has not responded to, despite our inquiries:
2016’s Popular Vote vs. Electoral Vote Dispute
The core of our piece focuses on Lichtman’s 2016 prediction. We submit that his 13 Keys predicted a popular vote victory for Trump. We reached this conclusion based on his 2016 book and his October 2016 article in Social Education, the latter of which contains the following quote: “As a national system, the Keys predict the popular vote, not the state-by-state tally of Electoral College votes.” It goes on to say, “When six or more [keys] are false, the challenging party wins.” This clearly connects the Keys to a win in the popular vote, given the explicit caveat that they do not predict the Electoral College and do predict the popular vote.
While Lichtman does say in the September 23, 2016 Washington Post interview and article that Trump would “win,” and while he does not mention the popular vote (the article itself does: “Nobody knows for certain who will win on Nov. 8 — but one man is pretty sure: Professor Allan Lichtman, who has correctly predicted the winner of the popular vote in every presidential election since 1984.”), this alone is not evidence that the Keys were no longer meant to predict the popular vote. He uses the word “win” similarly to describe a Keys victory in his 2004, 2008, 2012, and 2016 papers.
When he sent us the above response, we asked again about the two core points of our piece. He did not respond substantively to this first point. Because we were unable to find any evidence that Lichtman claimed the Keys were no longer meant to predict the popular vote before November 8, 2016, and he has not satisfactorily responded to our inquiry concerning his paper or book’s clear statements, we stand by our reporting regarding his 2016 prediction.
Claiming to Have Ceased Predicting the Popular Vote “Since 2000”
The second substantive claim is that Lichtman has been inconsistent about when his switch to predicting the popular vote occurred. Lichtman stated in a May 2, 2024, livestream that, “Ever since 2000 you know I’ve just been picking the winner.” This is debunked by the aforementioned 2016 sources and many other sources, among them: his 2004 paper, his 2008 book, his 2008 paper, his 2012 paper, and a 2020 article he wrote which directly contradicts any change “since 2000” and notably does not address when in 2016 (whether before or after the 2016 election) that his modification was made.
We asked Lichtman about this inconsistency as well. He did not provide a substantive response. Because we could not find a source published between November 7, 2000, and November 8, 2016, that demonstrates that Lichtman had redesigned the Keys to only predict the Electoral College winner, we stand by our reporting on this point as well.