Here’s the Wikipedia article about Lichtman’s system. It’s nothing new, it was developed over 40 years ago and has been making predictions about presidential elections since Reagan’s reelection.
It is very subjective and has received a lot of criticism. The article gets into a lot of detail about how it works. I’m putting this in here in case people are curious about who this guy is and how his predictions work. If someone wants to have a discussion about it, it should probably be either in one of the polling threads or a new thread (though personally, I don’t think one guy’s model is interesting enough for its own thread).
Lichtman’s system was discussed at length in another P&E thread a few months back. I think it was the “Polling is broken?” thread, but would have to check.
EDIT: I misremembered – Lichtman’s system did have it’s own thread:
I take that to be a tacit admission that he is actively trying to interfere on Trump’s behalf.
The key, of course, is trying. While I fully believe that many social media trolls are Russian bots, I also feel confident that the democrats, and the U.S. government as a whole, is better prepared for cyber warfare than they were in 2016.
538’s model again has Harris down to winning just 55% of runs.
The swing states that have slipped a little for her in the last week are Michigan, Pennsylvania, North Carolina, and Arizona. Only Wisconsin, Georgia, and Nevada are holding steady.
(Click on a state to see how often it favors Harris in model runs).
Following models day to day is a recipe for stress, and doesn’t really give you any new information. The polls say it’s a close race. They’ll say that tomorrow too. Maybe this will change at some point, but there’s essentially no difference between a 55% chance, 58% chance, and even 45%. All of those are functionally coin flips.
I think we’re going to see some systemic under counting of the anti Trump vote, just like we have in elections since '22, and even in the Republican primaries. But that’s just a guess. I recommend not following the models that closely, and certainly don’t let day to day changes impact your quality of life.
Electoral College 2024 Excluding States Where the Candidates are Statistically Tied
Until either candidate is statistically ahead in states totalling 270 electoral votes or more, we’re close to a coin flip. All that the fancy models do is try to squeeze a little more precision out of “coin flip”.
I wonder if some of that cash can be diverted to important races down the ballot:
tl;dr - supporting candidates in local races and for state legislatures can help some states near tipping points like Wisconsin, Pennsylvania and Arizona, and moving the state government into Democratic control in those states (as happened in Minnesota in 2022) can make a big difference in policy.
He put out the statement @Smapti cited after Liz’s comments, but I’m not aware that he’s said he’d vote for Harris previously. Of course, I don’t know that this is an endorsement Harris will particularly want to tout.
