Allan Lichtman vs polls: Indicators for US presidential elections

The media loves to try to predict who will be the next US president.

WHY NOT JUST WAIT AND SEE?

The main question is: Why? Why not just wait and see? The outcome will reveal itself when it’s time.

That would be a sane and sober approach, and it doesn’t fit the needs of most media.

Most media thrive on news and drama. That’s how they get readers and viewers. One way to generate news and drama is by trying to predict who the next president will be.

Many of us buy into it. If the media is focused on something, we do too. The media largely sets the topics for our public, and sometimes private, discourse.

GROUNDED IN SOCIAL & ECONOMIC FACTORS VS POLLS

So let’s look at indicators for who may get elected.

There are two main types of indicators: One is grounded in what’s happening in society and the world, and the other is polling. The first tends to be more stable over time. The second changes between pollsters and from week to week. The first may be more reliable, and the second needs a lot of analysis and interpretation to give useful results. (538 and Nate Silver were usually very good at it.1)

ALLAN LICTHMAN’S KEYS TO THE WHITE HOUSE

Among those in the first category, Allan Lichtman seems to have a pretty reliable system.

He is a historian who developed and tested his model – Keys to the White House – on historical data. The indicators are based on social and economic factors and events that influence or reflect who people actually vote for. He uses a mix of objective and somewhat subjective factors, which I think provides a good balance. Who people vote for is largely based on impressions, and subjective factors can catch some of that. How is the economy going? Who is the incumbent? How are the wars going? What was the tendency in the recent mid-term election? How charismatic is the candidate? And so on.

The model also has predictive value. Since he started applying it in the early 1980s, it’s been accurate2,3.

So why don’t the media go with his system? Why are they so focused on the always-changing polls?

The answer is likely the same: They need news and drama, and the polls provide that.

Allan Lichtman’s system may be more accurate, but it’s not news. It’s enough to interview him once or twice and that’s it. It’s not so good for the media, apart from spicing it up a little now and then.

WHAT DOES HIS SYSTEM SAY ABOUT THE 2024 ELECTION

His system shows that if the presidential election was today, Biden would likely win.

Between now and the election this fall, a lot has to go wrong for him to lose the election according to the Keys to the White House. What can go wrong includes wars abroad, the economy, and so on.

A WORD ABOUT PREDICTIONS IN GENERAL

These models are based on indicators and they just need to be largely accurate to be useful, and Lichtman’s system clearly is.

It’s natural to be focused on the current prediction and whether or not it turns out to be accurate. To me, the model itself is more interesting, along with how accurate it is over time and many instances.

WHAT THE INDICATORS SAY ABOUT SOCIETY & APPLYING THE SYSTEM TO OTHER SITUATIONS

I also find it interesting what Lichtman’s indicators say about society. It tells us something about what’s actually important to people, and how we arrive at collective decisions.

As far as I know, he hasn’t applied his system to other types of elections or other types or collective decision-making. I am also not aware of his system being adopted and applied to other countries. All of that would be very interesting and tell us more about how societies and collective decision-making work.

NOTES

(1) In 2016, Allan Lichtman predicted Trump as the winner based on his system. 538 and Nate Silver said there was a one-in-four chance Trump would win. (Whenever the polls looked the way they did, Trump would win one of four times.) I followed both before the election, which is why I wasn’t too surprised when Trump won. (I happened to be in Rockridge at the time, at the edge of Oakland, and heard the riots in the distance!)

(2) One exception was Al Gore, although the outcome of that election was ultimately decided by a court.

(3) The predictions may have been accurate not just because of his system. It may also be because of him and what he brings to it when he evaluates which direction each of the keys is going. Someone else would likely view the keys differently and their predictions would be different. That’s why I used his name in the title and not the name of his system.

Here are two recent interviews with Allan Lichtman from Times Radio: Biden can absolutely win the election | Biden to defeat Trump in election 2024

Image by me and Midjourney


INITIAL DRAFT

Allan Lichtman is a historian who has developed several indicators of who is likely to be elected president in the US. His system is based on history, and since he started applying it in the early 1980s, it’s never been wrong1. (Apart from Al Gore, although the outcome of that election was ultimately decided by a court.)

His system shows that if the presidential election was today, Biden would win, and a lot has to go wrong for him to lose the actual election.

To me, these indicators make a lot more sense than polls. They are based on real factors and events that influence who people actually vote for.

So why don’t the media go with his system? Why are they so focused on the always-changing polls?

The answer is simple: They need news and drama, and the polls provide that.

Allan Lichtman’s system may be more accurate, but it’s not news. It’s enough to interview him once or twice and that’s it. It’s not so good for the media, apart from spicing it up a little now and then.

(1) These kinds of systems are based on indicators and the outcome is an indicator as well. It just needs to be largely accurate to be useful.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.