Lichtman's 13 Keys to the 2024 Election

[Peterkin]

50 minutes ago, Airbrush said:

Allen Lichtman has addressed your concerns and believes his model is not yet outdated.

Then we must disagree, but only one of us free to change his method according to prevailing trend and still claim its efficacy.

[Airbrush]

22 hours ago, iNow said:

Even if the model is garbage and is based on the number of eyes on a potato spud or on how many ears of corn Tonya can fit into a record player box, it STILL has a 50/50 chance of being correct 

Suppose he had a 50% probability in one year.  For 9 years: 50% to the power of 9 = A very tiny probability his model is random.

[swansont]

1 hour ago, Airbrush said:

Could you quote that from his published paper of Oct 2016?  I can't find it.

I did quote that.

https://www.scienceforums.net/topic/134213-lichtmans-13-keys-to-the-2024-election/#comment-1274528

Social Education 80(5) p256, second paragraph 

“the Keys predict the popular vote, not the state-by-state tally of Electoral College votes”

 

1 hour ago, Airbrush said:

What octopus did you mention?  I searched above and could not find it.  Is there an octopus that correctly predicted 9 of the last 10 elections?  No octopus lives 40 years.  They live between 1 year and up to 5 years in captivity.

https://www.scienceforums.net/topic/134213-lichtmans-13-keys-to-the-2024-election/#comment-1274357

You’re 0-for-2 in finding things posted in this thread.

[iNow]

2 hours ago, Airbrush said:

50% to the power of 9 = A very tiny probability his model is random.

That’s not how this works. Chance alone using a coin flip would be exactly as good as him about 3x out of every 100 tries. Get a couple thousand people flipping coins and something like 15% of them would have equally good predictions.

Forgive me, but I find his prediction “abilities” far less impressive than you do. He’s capitalizing on people who are bad at statistics. 

Edited September 6, 2024 by iNow

[Peterkin]

5 hours ago, Airbrush said:

For 9 years: 50% to the power of 9 = A very tiny probability his model is random.

How, exactly, does this equation work? I'm no mathematician, but ....

3 hours ago, iNow said:

He’s capitalizing on people who are bad at statistics. 

I suspect he actually believes that those key factors determine the outcome of elections. Based on a set of assumptions about US politics that have prevailed in the 20th century, they would be strongly indicative of the outcome. But he didn't add in unprecedented factors - probably because they're extremely difficult to quantify and any attempt would have complicated the model too much. 

[Airbrush]

When you examine Allan Lichtman’s 13 Keys, the keys were correct, and Allan simply interpreted them slightly off.  I agree with Allan’s analysis of the first 4 keys, but for short-term and long-term economy, those were perceived as FALSE by the public because of massive disinformation.  I agree with Allan on keys 7, 8, and 9, but I disagree on 10 and 11.  Allan believed that Biden got the benefit of uniting countries against Russia’s invasion of Ukraine, but the public did not perceive that.  I agree with Allan on key 12, but I differ on key 13, because although you cannot call Trump “charismatic” he gets partial credit for having cult leader popularity, like rarely seen, except for truly charismatic candidates.  So, when I count the keys that fall against the Whitehouse party, I get 6 and a half strikes against the Whitehouse party.  Which indicates a Trump victory.  Here is how I analyze the 13 keys.  (1) Biden loses the 2 economy keys, (2) loses another Foreign/Military success key, and (3) gets half credit for challenger being “semi-charismatic.”  Here is my analysis:

1 Party Mandate – FALSE

2 No Primary Contest – TRUE

3 Incumbent seeking re-election – FALSE

4 No Third Party – TRUE

5 Strong Short-Term Economy – FALSE (PERCEIVED AS FALSE BECAUSE OF MASSIVE DISINFORMATION

6 Strong Long-Term Economy – FALSE (PERCEIVED AS FALSE BECAUSE OF MASSIVE DISINFORMATION

7 Major Policy Changes – TRUE

8 No Social Unrest – TRUE

9 No Scandal – TRUE

10 No Foreign or Military Failure – FALSE (PERCEIVED AS FALSE)

11 Major Foreign or Military Success – FALSE (PERCEIVED AS FALSE)

12 Charismatic Incumbent – FALSE

13 Uncharismatic Challenger – PARTIALLY FALSE BECAUSE TRUMP WITH HIS CULT FOLLOWING GETS PARTIAL CREDIT FOR CHARISMA

 

 

Edited November 24, 2024 by Airbrush

[iNow]

On 8/29/2024 at 1:26 PM, Airbrush said:

Lichtman has correctly predicted the results of ten consecutive elections, from 1984 to 2020. 

On 9/1/2024 at 2:10 PM, Airbrush said:

Lichtman believes a lot would have to go wrong for Harris, for Harris to lose. 

On 9/6/2024 at 11:09 AM, Airbrush said:

last night Lichtman pulled the trigger, before the debate, and said Harris would win.

 

[swansont]

6 hours ago, Airbrush said:

the keys were correct, and Allan simply interpreted them slightly off.

Thank you for confirming that this is utterly unscientific.

It’s more of a cult, like when someone keeps predicting the aliens will show up, and then comes up with an excuse when the target date comes and goes with no aliens. But keep believing! All that’s missing is a plea for more money.

[Peterkin]

12 hours ago, Airbrush said:

When you examine Allan Lichtman’s 13 Keys, the keys were correct, and Allan simply interpreted them slightly off.

To what do either the method or its prediction make any difference whatsoever?

Seems to me like a lump of grey, non-functional goo for people to hand off one to another and chatter about. 

[Airbrush]

On 11/24/2024 at 11:26 AM, Peterkin said:

To what do either the method or its prediction make any difference whatsoever?

Seems to me like a lump of grey, non-functional goo for people to hand off one to another and chatter about. 

What is the probability that someone could correctly GUESS 9 out of the past 11 elections?  Lichtman is now 82% accurate.

Edited 12 hours ago by Airbrush

[KJW]

12 minutes ago, Airbrush said:

What is the probability that someone could correctly GUESS 9 out of the past 11 elections?  Lichtman is now 82% accurate.

The flaw with this is that elections are not random and therefore applying probability is not valid. Some elections are easy to predict the outcome, others not so easy. If some methodology correctly predicts easy elections but gets the hard ones wrong, then that methodology is really quite useless, regardless of how many easy elections are correctly predicted.

 

[swansont]

22 minutes ago, Airbrush said:

What is the probability that someone could correctly GUESS 9 out of the past 11 elections?  Lichtman is now 82% accurate.

That he’s a tad better than a coin flip is hardly a ringing endorsement. A coin flip gets you 6 correct out of 12, and all you need are three elections where the winner is pretty obvious where there’s no real guesswork (Reagan beating Carter, Bush beating Dukakis are two candidates)

[Airbrush]

4 minutes ago, KJW said:

The flaw with this is that elections are not random and therefore applying probability is not valid. Some elections are easy to predict the outcome, others not so easy. If some methodology correctly predicts easy elections but gets the hard ones wrong, then that methodology is really quite useless, regardless of how many easy elections are correctly predicted.

 

Excellent observation.  I hadn't thought of that.  Elections are not purely random.  In your estimation, how many of the past 11 elections were "easy" to predict?  I should consult with some Historians.

Here is a list of the major candidates for U.S. President from the two leading parties over the past 11 elections, spanning from 1984 to 2024:

Election Year	Democratic Candidate	Republican Candidate
2024	Kamala Harris	Donald Trump
2020	Joe Biden	Donald Trump
2016	Hillary Clinton	Donald Trump
2012	Barack Obama	Mitt Romney
2008	Barack Obama	John McCain
2004	John Kerry	George W. Bush
2000	Al Gore	George W. Bush
1996	Bill Clinton	Bob Dole
1992	Bill Clinton	George H. W. Bush
1988	Michael Dukakis	George H. W. Bush
1984	Walter Mondale	Ronald Reagan

Wikipedia

7 minutes ago, swansont said:

That he’s a tad better than a coin flip is hardly a ringing endorsement. A coin flip gets you 6 correct out of 12, and all you need are three elections where the winner is pretty obvious where there’s no real guesswork (Reagan beating Carter, Bush beating Dukakis are two candidates)

Interesting to note that.  I should post your reply below several of Lichtman's Youtubes. ☺️

[KJW]

55 minutes ago, Airbrush said:

In your estimation, how many of the past 11 elections were "easy" to predict?

I'm not an American, so I can't answer this question. However, I did have a rather strong interest in this US election, more than any previous US election (and even more than the recent state election at home). Based on the information I was receiving, it seemed to me that Trump was going to win. Although the political commentators were saying the polls were within the margin of error, the polls were nevertheless pointing to a Trump victory. And many of the "ordinary folk" that I saw interviewed were saying that they were voting for Trump. But I'm no expert and certainly wouldn't have locked in a prediction. Watching the election results come in, it did seem like a foregone conclusion from the start.
 

 

Edited 10 hours ago by KJW