spiderstumbled
Wednesday, August 13, 2025
2024 Post Mortem (with Extra Rigor Mortis)
Monday, November 4, 2024
Four More Years But For Whoah....Wait a Second. Final 2024 General Election Forecast.
I. Four Scenarios.
The race remains extremely close. The betting odds are close to even money. The forecasts are nearly tied. But in less than 48 hours, we will probably know the winner and almost certainly the narrative will be one of these four things:
1. For a third consecutive election, the polls underestimated Trump. In this scenario the popular vote is close and Trump wins enough of the battle ground states to become president again.
2. The pollsters underestimated Harris organization and ability to turn out voters. In this scenario, the Harris campaign turns out low propensity voters and people who are simply sick of Donald Trump and all of his baggage.
3. Chaos. There has been a lot of discussion about whether 2024 will be a repeat of 2016 (Trump surges to victory) or 2020 (America turns out Trump.) But we have to consider a repeat of 2000. the polling in swing states have been so close that it's very possible we will not know who is going to be president until the courts weigh in.
4. Harris Wins Big. The Des Moines Register poll of Iowa (conducted by Ann Selzer) has been the golden child of polling the last two cycles. They found Trump up big there in 2016. Not only were they right about that, but the shift was replicated in other midwestern states and Trump took down the "blue wall" that was supposed to save Hillary Clinton. In 2020 they had Trump up in Iowa again and Trump won Iowa easily for a second time. The result was echoed in other states, and the election wound up being closer than expected.
Trump won Iowa by nine points in 2016 and by eight in 2020. Selzer had Trump winning Iowa by seven in both of those cycles. Two days ago the final Des Moines Register poll had Harris up by three. If this poll is once again extremely accurate, than Harris would flip a bunch of states that Biden lost last time.
I do not expect this poll to be that accurate for a third time in a row. But I also don't expect the wider pool of pollsters to under estimate Trump for a third straight time. In short, I think Harris is going to win. But I think it will be close.
II. Seven Swing States and Two Stretches.
The election has been narrowed to seven states for some time now. To win the election Harris needs 44 Electoral votes from these states. Trump needs 49. From most to least important, this is what I expect.
1. Pennsylvania (19 electoral votes). This is the dog fight. The candidate who wins this state is the heavy favorite to be the 47th president. Almost every recent poll has it within a point or two. But a lot of that polling has been from partisan Republican pollsters. And they only find Trump up by a point. That should be encouraging to Harris. I think she will win the state by a point or two. If we know that result by Wednesday morning, then we are probably avoiding the chaos scenario.
2. Georgia (16 EV) Harris won this state by 11,779 votes in 2020. The state also has elected two Democratic senators. It was also the place where Trump's attempts to steal the 2020 election were most blatantly corrupt. The states demographics are trending left and Harris has made a full court press to win this state. Recent polling here has been extremely close but I think Georgia is more likely than Pennsylvania to go Republican this time. I favor Trump slightly.
3. North Carolina (16 EV) Georgia and North Carolina are interchangeable for the electoral map. They each have sixteen electoral votes and they are both southern. It is the only closely contested state that Trump held on to last time. (He won by about 94K votes.) Most of the polls here have shown Trump with a slight lead but the final New York Times poll had Harris up by three.
NC was hit very hard by Hurricane Milton. The response by the federal government was extremely competent. North Carolina also have a very popular Democratic governor and the Republican candidate to replace him is a full blown loon. I think that helps on the margins.
One other piece of data worth mentioning is that inflation was relatively mild in North Carolina. Nate Silver has been talking about this a lot. It seems that the states that avoided really high inflation are going to swing more to the left than other states. (This could be significant for the House-New York and California were hit hard by inflation. That would change the presidential result but it could hurt Democrats in the House.)
So my pick in North Carolina is Kamala Harris. By about a point or two.
4. Michigan (15 EV) Michigan will be one of the most closely scrutinized states regardless of the results. Michigan is home to large Arab and Muslim populations and they are concentrated in specific cities. There are a lot of Muslim and Arab voters who feel they cannot vote for Harris because of Joe Biden's decision to support Israel's response to the attacks of October 7th. I am sympathetic to this argument but also very aware that Trump will be much worse for the people of Gaza than Kamala Harris.
I think the protest vote will be measurable and I think it will cut into Harris lead. But it won't be enough for Trump to win the state. Trump won Michigan by less than a quarter of a point in 2016 and Biden won it by three points last time. Harris will win Michigan again.
5. Arizona (11 EV) This is a weird one for me. I think AZ is a tough state to poll. Several really smart people that I follow on Twitter think this state will stay in the Dem column. It's possible but I cant' ignore that it's been awhile since Harris lead in a poll there. I think Arizona flips back to red.
6. Wisconsin (10 EV) This is where I will be on election day. I have also done a good deal of phone banking for this state. The polls are close but the Wisconsin Democrats have a great organization. I think Harris will win Wisconsin.
7. Nevada (6 EV) Nevada is the state that Dems always worry about and then they win. Polling has been a bit scattered and the Republican party has done a pretty decent job of running up a lead in early voting. But if history holds, the Reid Machine will carry Kamala over the line. Harris will win a close one, but the state will probably not be called for several days. Hopefully this time it won't hold up the call of the national race because Harris will be declared winner in Pennsylvania by Wednesday or Thursday.
The Two Stretches:
1. Kind of small beer here, but I think Harris might win Maine's 2nd congressional district. The Selzer poll of Iowa suggests that Harris is doing better with white voters than some expected. This could be a nice bonus but it won't affect the outcome anywhere.
2. The Selzer poll means we have to consider Iowa competitive. She has been accurate within a point or two the last two times and she has a reputation for letting the data speak for her. I admire that. Iowa passed an abortion ban that only took effect recently. It seems to have shifted the state left. One fun wrinkle of the Selzer poll is that it had RFK junior getting three percent of the vote. I don't think he will be much of a factor anywhere, but weird old Iowa might be the one place it matter.
I'm keeping Iowa red on my map, but I think it will be close. Let's say Trump by two.
The only changes from my last forecast are moving NC and ME-2 blue. Here is what my map looks like now:
The American economy is in exceptionally strong health. We are no longer at war. Crime is falling and we've even reduced our dependence on fossil fuels. The current administration has been a success.
Democrats 223, Republican 212
Tuesday, October 29, 2024
Four More Years But For Whoah....Wait a Second. (2024 General Election Forecast, Take Five)
In 2020 I updated my forecast sixteen times. This is just my fifth in 2024 and I don't expect to do another one.
1. The State of the Race.
The race is too close to call. Extremely so.
There are only seven states that are being contested. The polls in all seven states are well within the margin of error. The most likely outcome is that one of the two candidates will sweep all seven. There are 128 permutations of the seven swing states. According to Nate silver, there is a 25% chance that Trump will win all seven and a 15% chance that Harris will win all seven.
In this situation, complicated electoral college map is pretty useless. Flip a coin seven times and you'll be in as full command of the facts as any pollster or pundit. So I'm not going to do that today. I'm going to talk about why this race could go either way.
2. Reasons for Trump to be Optimistic.
a. The American people have not disqualified him. By now I expected enough people to have ruled out voting for for more years of that amoral aspiring tyrant. I thought he would be stuck around 45 or 46 in the polls. He isn't. He's mostly between 47 and 49. He probably wins at 49. He probably loses at 48 and will lose at 47. It's that close.
b. The forecasts and betting markets have shifted towards him in the past few weeks. The FiveThirtyEight.com model gives Trump a 53.9% chance of winning. Nate Silver gives him a 54.4% chance of winning.
The betting markets now lean in Trump's direction slightly more. Predictit has him as a four to three favorite. Polymarket is about 2 to 1 in his favor, although some of that is due to some distortion by Crypto Bros who want to shape the narrative that Trump is inevitible.
c. He is eager to cheat and a lot of his party is willing to go along with that. Whatever dirty trick they can pull, it shall be pulled. Trump has long assumed that this race will be won or lost in the courts because he has convinced himself that the courts cheated him out of victory in 2020. (They didn't. He's an idiot.)
3. Reasons for Harris to be Optimistic.
a. Women hate Donald Trump. The world has been unkind to incumbent parties since COVID-19. But the United States has one issue that makes our election different. Donald Trump appointed three justices to the Supreme Court and they all voted to give states the authority to criminalize abortion. Red states have not been shy about exercising that authority. Women have noticed.
b. Kamala Harris is more popular than Hillary Clinton was in 2016. Hillary was less hated than Trump in 2016 but they were both unpopular. Harris's personal approval ratings are roughly neutral. Trump is still unpopular generally.
c. The Democrats have raised more money and have a much better ground operation than Team Trump.
d. Trump's age is really showing. He is tired and he rambles even more than he did in his previous runs. He is old and the people have noticed. I think in the end, this will make the difference.
4. Things to Ignore.
People are naturally trying to divine the meaning of early voting patterns. There are some theories that bolster Trump (especially in Nevada) and others that bolster Harris (especially with gender differences and the numbers in Pennsylvania). But I am going to hold off on putting too much stock in any of that because 2020 was such an outlier for early and mail voting. Four years ago we were in a pandemic. One campaign acknowledged that and the other actively discouraged people from voting before election day. The situation is different now. The overall numbers will be lower but I don't think they will be as lopsided in favor Democrats either.
5. My Updated Forecast.
This race is going to be very close. For the past few weeks Trump seemed to have a slight bit of momentum behind him. I think that has petered out. In the home stretch he looks tired and frail.
Harris has run a good campaign. I don't think she's a great retail politician but she has run well. She's done the appropriate mix of interviews, rallies and advertising. She's focused on the right issues and I think she is more likely to close this deal than not.
With one week to go, I think this is where the probabilities of each swing state stand:
1. Michigan (Harris 67%)
2. Wisconsin (Harris 60%)
3. Nevada (Harris 55%)
4. Pennsylvania (Harris 53%)
5. North Carolina (Trump 51%)
6. Georgia (Trump 55%)
7. Arizona (Trump 60%)
But there are 128 permutations for how 7 coin flips can go. Next week all 128 are possible. This is the one I find most likely:
Tuesday, October 1, 2024
VP Debate Night, 2024
I. The State of the Race.
I have not had a reason to update my map since late July. For all the daily weirdness of the Trump campaign, the race has been pretty stable. Kamala Harris is a slight favorite. She will probably win the popular vote but there are enough very close states to keep the winner uncertain.
The simple version of the race is this. Harris is likely to win states worth 226 electoral votes. She has to win 44 more votes among the seven swing states. At the moment, I classify them this way:
Leaning Harris: Wisconsin, Michigan, Nevada and (just slightly) Pennsylvania.
True Toss-ups: North Carolina and Georgia.
Leaning Trump: Arizona
So Harris probably has just enough votes to win the election as it is polling now. There are five weeks to go. While there will certainly be unexpected news, there aren't many big events left on the calendar. Which leads us to tonight.
II. The Vice-Presidential Debate.
The VP debates are almost never consequential. The most memorable moment in their history was when Lloyd Bentsen insulted Dan Quayle for presuming to compare himself to John F. Kennedy. Bush-Quayle crushed Dukakis-Bentsen that November.
Tim Walz has been a great candidate so far. His favorable numbers are by far the best of all four candidates and he hasn't committed any gaffes. No dirt has come to light and his super power of normalcy has been a refreshing change of pace.
J.D. Vance has brought absolutely nothing to the Trump team. He has said a lot of dumb things and never, ever looks comfortable on the campaign trail. People do not like him and they are right to not like him.
But the previous paragraphs only set the expectatio0ns for tonight. Vance is a pretty smooth speaker when he has a script and he had the oddly misplaced confidence of a much smarter man. He could do well tonight. Well enough to exceed expectations at least. And in a pinch, he is not afraid to lie.
III. The Questions That Should Be Asked.
1. To Tim Walz, "Vice-President Harris and you have derided Trump's plan to place a tariff on all imported goods. But Joe Biden has kept many Trump-era tariffs in place. What metrics would you use to determine the difference between good and bad tariffs?"
2. To J.D. Vance: "After spreading the false rumor that immigrants from Haiti in Springfield, Ohio were eating pets, you said that you felt the need "to create stories so that the media actually pays attention to the suffering of the American people." How should voters know when you are telling the truth and when you are making stuff up?
3. To Tim Walz. "Does the United States of America admit too many, too few or just the right number of legal immigrants to this country? What criteria would you use to determine what that number should be?"
4. To J.D. Vance: "If you become president in the next four years, would you sign a national ban on abortion and if so, what would be the parameters of that ban as to length of pregnancy and exceptions for medical emergencies."
5. To Tim Walz, "Are there any specific programs or areas of the military budget that you believe should be reduced or eliminated?
6. To JD Vance, "Was the 2020 presidential election free and fair and who was the lawful winner of that election?"
7. For both candidates, "What benefits does the United States get from its relationship with Israel and are there any changes that you would make to how the United States provides aid to Israel?"
With five weeks to go the election is probably about a 60/40 tilt in favor of Kamala Harris. Tonight probably won't change that dynamic. I don't think it will much change how people feel about Tim Walz. But J.D. Vance could do a lot of good for himself by just being normal. I'm not sure he has it in him.
Tuesday, September 17, 2024
How I See It, How Trump Should See It.
Well, it has been a week. Six days ago, Kamala Harris dominated her first (and probably only) debate with Donald Trump. She was great on a lot of levels but my favorite part of her performance was how she reminded people just how tiring Donald Trump's whole shtick is. She baited him repeatedly and he walked into it every single time. Because he's a baby.
Since then he has spent his time insisting that he won the debate, doubling down on racist lies about Haitian refugees eating pets and refusing to do a second debate. (Probably his best decision in quite awhile.)
The post-debate polls have been generally good for Harris. As for my official prediction, I'm sticking with Harris 292, Trump 246. I think the popular vote result will be similar to 2020 and there will not be many changes in the electoral college either.
I. How I See It.
A. Swing Stones Leaning Blue
I am reasonably confident that Kamala Harris will win Wisconsin, Michigan and Nevada. I also think she is more likely than not to win Pennsylvania. These states will get her over the finish line with 276 votes.
B. True Toss-ups
The polls in Arizona, Georgia and Arizona have been close. I can imagine any of these states going either way. If Trump sweeps them, the election will be close and he will try like hell to steal it. Every state that Harris wins here will make it harder for Trump to do another insurrection. (It's so sad that we have to think this way, but we do.)
II. How Trump Should See It.
In my previous post, I divided the swing states into three northern states and four "southern." That was because Harris needs to sweep the three northern states (WI, MI, PA) to get to exactly 270 votes and win the election. For any northern state that she loses, she has to make up the equivalent number in the four southern states.
For Trump the math and the geography work a little differently. Trump should win at least 219 electoral votes. To get to 270, he needs 51 more. The easiest path there is to sweep Pennsylvania, North Carolina and Georgia. So for today's purposes, we'll call those three states The Eastern Swing States.
For ever electoral vote Trump loses in the east, he has to make up for them in the four "western" states.
Here are the combinations that put Trump in the White House. Because GA and NC are both worth exactly 16 votes, they are interchangeable in these scenarios.
IIII. How Forecasters And Markets See It.
I think Harris is about a two to one favorite now. Let's call it 66% chance of her winning. Harris has run a very good campaign and Trump continues to step on rakes. Harris has a fundraising advantage and unlike last time the Democrats will be out knocking on doors this fall.
Trump is going to lose the popular vote but he can still win enough swing states to get to 270. His biggest problem right now is that a lot more voters have made up their mind against voting for him than have made up their mind against voting for his opponent.
Almost all of the forecasters like Harris better today than the did a week ago. Even Nate Silver has her chances up by a full five percent from right before the debate.
Nate Silver and his former baby, 538.com are still 17 points apart because they moved in the same direction this week. But like Harris about five percent more than they did before the debate.
As for betting markets, Predictit.com favors Harris 57 to 47 and Poly Market has the race must closer with Harris leading 50% to 48%.
Tuesday, September 10, 2024
Debate Night! 2024
The second debate of the 2024 general election will be begin in under four hours. It is the first debate to include Kamala Harris. As I usually do, I will post some questions that I would like to hear the candidates be asked tonight. But first, a quick word about the state of the race.
I believe that Harris will win the popular vote by something between three and six points. I believe she will sweep the Blue Wall states of Wisconsin, Michigan and Pennsylvania. I also think she will win Nevada. I think Georgia, North Carolina and Arizona are true tossups that either candidate could win if the election were today. So my hunch is that she will win one or two of those, but not all three.
This is slightly more bullish than the average forecast, as you can see below.
I. The State of the Race.
The race is very close. Most recent national polls have been within two points. Harris has led in the majority of them, but that doesn't tell us much since she probably needs to win by a point and a half or two points to become president. The professionals and hobbyists who look at this stuff in some detail tend to put the likelihood of either candidate winning somewhere between 40 and 60 percent.
Which of course is making everyone on Politics Twitter mental. The uncertainty is grating and not everyone is taking it in stride. But I don't think there's much difference between 60/40 and 40/60 when there are eight weeks to go. Here I have round up the predictions from various models and prognosticators. And I'll include myself because well, why not?
II. Questions That Should Be Asked.
1. To Donald Trump, "Who won the 2020 election for president?"
I'd like to get even more specific and ask Trump which states he is willing to admit that he lost last time. But last week Trump seemed to make some news in admitting on a podcast that he lost "by a whisker." He probably said that to avoid the podcast host from beating him up on election denialism. I think the moderators need to push further on this today because well, shit, democracy matters.
2. To Kamala Harris, "How would you have handed the withdrawal from Afghanistan differently than Joe Biden and Donald Trump did?"
I continue to believe that the withdrawal from Afghanistan was the most successful large-scale operation carried out by the American military this century. Many disagree with me on that, although, none of them can name one more successful. But this is an issue and I want to hear both her talk about it and Trump respond to it.
3. To Donald Trump. "You are awfully fond of Vladimir Putin, the House of Saud, Xi Jinping and Kim Jung Un. Do you believe it is important for America to promote democracy abroad?"
I am recycling this from my post about the June debate, because I think it's a great question.
4. For Kamala Harris: "Why specifically did you change your views on fracking since 2019?"
5. For Donald Trump: "The first time you ran for president you promised to balance the budge and pay off the debt in four years. Your current campaign is predicated on a lot of tax cuts and a lot of new spending. The only new revenue source you have endorsed are tariffs on all imported goods. Have you crunched the numbers on all of these plans and how much do you think the national debt will be four years form now if you are elected."
6. For Kamala Harris " Do you think that any tariffs currently in place contributed to the inflation of the past few years and will you repeal or reduce any of them?
7. For Both Candidates: Three out of four of your parents were immigrants. Mr. Trump's father was born about 2 year after his parents arrived in this country from Germany. Why do you think so many voters s have hostile feelings about recently arrived Americans and what will you do to change that?
As always, I would also like to have them answer the questions on the test to become a naturalized citizen but I don't think we will get anything like that tonight.
III. If This Were a Baseball Game.
As the election proceeds, I will play around with probabilities from a different passion, baseball.
To think of think of the election as a baseball game. We'll treat the remaining weeks as innings. And we'll make Harris the home team, since we are root, root, rooting for her and if she doesn't win it will be a shame.
So with eight weeks to go, the game goes to the second inning today.
A. 538 Baseball Forecast: The game is tied in the top of the 2nd. There is one out and no runners on base.
B. Nate Silver Baseball Forecast. The visitors have a 2-0 lead in the top of the second. There are two outs and no runners on base.
Bottom Line: A lot of baseball left to be played.