The Vice-Presidential debate took place four days ago. In many ways it was like a time machine took us back to the politics we used to have: A debate that was primarily about policy rather than personal attacks, and where the candidates often had complimentary things to say about one another.

But in other ways it was a reminder of what our politics have become. Vanity Fair’s headline perhaps put it best: “JD Vance Won The Debate–If You Ignore The Facts”. Vance was, as one might expect from a Yale Law grad, a very accomplished debater. The problem is, he painted a world in which Trump peacefully transferred power on January 20th without fuss; in which the real threat to democracy was the Biden Administration’s efforts to stop the spread of misinformation on social media about COVID-19; in which Trump saved Obamacare; and in which illegal immigration is a major driver of the housing crisis. (As somebody tweeted: “Love the idea of immigrants crossing the border and immediately outbidding a white couple for a 3 bed/2 bath in suburban Virginia.”) Perhaps the most memorable line of the debate was Vance’s retort to a rare interjection from the moderator: “The rules were that you guys weren’t going to fact check.”

Vance may have won the debate, but narrowly if so; and life rapidly moved on.

The next morning, Judge Chutkan released a redacted version of Special Counsel Smith’s 165-page brief on how the SCOTUS presidential immunity framework from Trump v. U.S. ought to, in the government’s view, impact proceedings in U.S. v. Trump (Jan 6th edition). The redactions were surprisingly modest, basically just replacing names with numbered codes for co-conspirators (CC), other persons (P), and companies (C); however in most cases it is pretty clear who or what the codes represent.

The day after that, former co-chair of the Jan 6th Committee Liz Cheney campaigned with Vice-President Harris in the Wisconsin town of Ripon (in which the Republican Party was first formed in the early 1850s), under signs reading “Country Before Party”.

Earlier in the week Hurricane Helene wrecked devastation in Florida, Georgia, and especially western North Carolina. As of today the death toll is up to 227, making Helene the deadliest storm in the mainland U.S. since Katrina in 2005. And just today it appears that another storm, Milton, could strike Florida on Wednesday as a Category 3 hurricane. The political ramifications of these storms in swing states remain unclear.

Port workers struck in midweek, and there were concerns this strike could create severe havoc leading up to the election; however, the strike was settled fairly quickly, and without President Biden having to invoke the Taft-Hartley Act.

Today Trump returned to Butler, Pennsylvania, the site of the July assassination attempt, without incident. And in other assassination attempt news that I had missed last week: The federal case against the man who was hiding in the bushes outside Trump’s golf course allegedly seeking to kill him was randomly assigned to, of all people, Judge Cannon. I wonder how fast she’ll move that case along.

Nothing has changed this week in the polls’ assessment of the race; today Silver’s model is at 56.1% for Harris, on a national vote share of +2.7%, virtually unchanged from last week.

It’s been a relatively quiet week.

Perhaps the most important event of the week transpired in Nebraska, where State Senator McDonnell confirmed his continued opposition to a winner-take-all electoral vote allocation in that state, prompting Governor Pullen to rule out calling a special legislative session to try and change the law prior to the 2024 election. This preserves a Democratic path to a 270-268 victory that involves retaining MI/PA/WI while losing AZ/GA/NV, with the 270th vote coming from Nebraska’s 2nd district, the ‘blue dot’ of Omaha in the midwestern sea of red.

The importance of NE-2 was underscored by a new CNN poll this week, showing Harris with a 53-42 lead in that district. The same poll also showed that district’s incumbent representative, Republican Don Bacon, trailing 44-50 in his race against Tony Vargas; this is the second recent poll to show Bacon, previously considered the favorite, 6 points behind in what could be a key race for Democrats in their bid to retake the House.

As of this morning Silver’s model has Harris at 57.3% to win, on a national popular vote margin of +2.7%. Silver currently has NV, PA, WI and MI all in the 59-66% range for Harris, and NC, GA, and AZ all in the 58-65% range for Trump.

We are now only three days away from the Vice-Presidential debate, which seems increasingly likely to be the last debate of this electoral cycle. Harris had recently accepted a CNN invite to debate Trump on October 23rd, 13 days before the election; however Trump refused, claiming it is too late to hold another debate. (In 2012, the third and final Obama-Romney debate took place on October 22nd; the third and final Clinton-Trump debate in 2016 was on October 19th.)

On the legal front, this week a New York appellate court held oral arguments on Trump’s appeal of Judge Engoron’s verdict in the Trump Org fraud case. It is unclear from the arguments how the court might rule, and whether or not that ruling will come before the election. And Special Counsel Smith just filed with Judge Chutkan, under seal, an supersized brief on how the SCOTUS presidential immunity ruling in Trump v. U.S. impacts the indictment in U.S. v. Trump (D.C. edition); however as of this writing it is unclear if/when portions of the brief will be redacted and unsealed.

In political corruption news this week, federal charges were filed against the sitting major of New York City, Eric Adams, a moderate Democrat. Interestingly, he is currently on the slate of Democratic electors in New York, raising one observer to raise the possibility of Harris winning 270-268 but Adams acting as a faithless elector to swing the election to Trump (since a Republican house would get to break a 269-269 tie).

Finally, just like any Presidential candidate would do forty days out from the election, Trump this week announced a new line of Trump watches, including a limited edition watch priced at $100,000. Ah, the cognitive dissonance of Vance incorrectly complaining about the high cost of eggs while Trump is trying to get people to pay $100,000 for a watch that likely costs $20,000 to produce.

Some hijinks may be afoot in Nebraska, raising the possibility that a small number of state legislators could tip the scales on the 2024 presidential election before the ballots are cast.

Nebraska is of course one of only two states, the other being Maine, who award 1 electoral vote to the winner of each congressional district, plus 2 to the state-wide winner. Nebraska’s current practice started in 1992, and there have since been two occasions – 2008 and 2020 – where the so-called “Blue Dot” of Omaha awarded its single vote to the Democratic candidate. There was discussion back in April that Republicans sought to change that practice and revert to the more common practice of awarding all electoral votes on a statewide basis; however, that would have required a two-thirds majority of the (unicameral) Nebraska legislature, and that level of support for the change did not exist.

Well, according to new reporting from the Washington Post and New York Times, Trump is making a renewed push to get Nebraska to change its rules in time for this fall’s election. As I recently discussed, there is a very plausible electoral map where Harris wins 270-268 if Nebraska allocates its votes the way it currently does, but where Trump would win the election (after the House votes state-by-state to resolve a 269-269 tie) if Nebraska changes its approach. The key legislative holdout at this point is State Senator Mike McDonnell, a former Democrat who switched parties earlier this year over reproductive and transgender rights; he is term-limited but may be seeking to run for mayor of Omaha next year. If McDonnell can be convinced to switch his position, two other holdouts may follow in which case the Republicans may have the votes; and they may be able to make this change, via a special legislative session, any time up until the day before election day.

When Nebraska was considering this in the spring, the conventional wisdom was that if it were to make the change, then the Democrats controlling the Maine state government would enact the same change. This would preserve balance in close-fought elections, trading Omaha’s Democratic-leaning electoral vote with the Republican-leaning electoral vote in Northern Maine. However, by delaying the Republicans may have gained the upper hand. While Maine could (unlike Nebraska) apparently pass this change with simple majorities in both houses, such a bill would not take effect until 90 days after passage; in order to enact a bill that would take immediate effect, there would need to be two-thirds majorities in both houses, and the Maine Democrats are not believed to have the necessary votes for that.

But this week’s potential electoral hijinks are not limited to Nebraska. There has been a lot of concern in certain circles lately about the Georgia Election Board, the membership of which has seen significant turnover this year with the new majority seen as pro-Trump. In the latest of several potentially disruptive rules changes made by the Board, this week they voted to require an immediate hand count of the number of ballots, to verify that the total is consistent with the results of the machine count and resolve any inconsistencies that may be found. Critics of the proposal say that it is impractical in all but the smallest counties, as well as unlawful.

Silver’s model has Harris’s odds of winning up to 51% for the first time since the DNC ended, with an expected national popular vote margin of 50.6% to 48.3%. Right now NV (52.6%), PA (53.7%), WI (55.4%), and MI (61.8%) are all leaning slightly to Harris, while Trump is slightly favored in NC (61.7%), GA (64.1%), and AZ (64.5%).

This week two different groups filed amicus briefs with the 11th Circuit in support of the government’s appeal of Judge Cannon’s dismissal of the charges in U.S. v. Trump (Florida edition). One brief came from CREW, who was responsible for bringing the Colorado case that led to this spring’s SCOTUS decision in Trump v. Anderson; the other came from a newer group called State Democracy Defenders Action, whose board members include George Conway. Both the CREW and SDDA briefs urge the 11th Circuit, in addition to overturning Cannon’s ruling, to exercise its supervisory authority to re-assign the case to a judge other than Cannon. Special Counsel Smith did not take that position in the government’s brief, although many pundits had suggested he ought to.

Finally, yesterday the lock-up period on Trump’s majority holdings of TMTG expired, although there’s no indication that he sold any stock and he has said he won’t. DJT’s stock price was down 26% for the week, closing yesterday at a new low of $13.55; it has been the worst-performing stock in the Russell 1000 since Aug 5th, losing about half its value in that time.

Over the weekend, there was what is commonly being called an “apparent assassination attempt” on Trump, at his golf course in West Palm Beach FL. The would-be assassin allegedly spent 12 hours hidden in shrubbery on the perimeter of Trump’s course, only for his gun barrel to be spotted by the Secret Service when Trump was one hole away from reaching the gunman’s position. The alleged individual involved is a white male in his late 50s with a lengthy criminal record, and who in 2020 had posted on social media that he’d voted for Trump in 2016 but had become disillusioned.

Prior to this event, it was starting to look as through Harris was seeing a post-debate bump in the polls. Having said that, Nate Silver noted today that there have not been many high-quality, post-debate, state-level polls released yet. As a result his model, which at this writing gives Harris a 44% chance of winning the electoral college, may not yet fully reflect her strong debate performance. I’ll wait another week or two before I start paying too much attention again to Silver’s model.

Silver did point out recently, in the wake of an Alaska poll showing Trump with only a 47-42 lead, that one can easily imagine an electoral map in which Harris holds a 268-267 lead without Alaska, making the 49th state the unexpected key to the election. (This involves Trump flipping both GA and PA relative to the 2020 map, while Harris holds on to AZ, MI, NV, and WI and does not flip NC.) Intriguingly, Alaska will be using ranked choice in the presidential election for the first time this year, and there are 8 names on its ballot, including RFK Jr.

Today Trump’s attorneys asked the 11th Circuit to extend, from September 25th to October 25th, the deadline for their reply brief in the government’s appeal of Judge Cannon’s dismissal of U.S. v. Trump (Florida edition). Special Counsel Smith is not expected to oppose the request. Even with the delay, it is plausible that oral arguments will be scheduled between the election and the inauguration.

Last night was the Harris-Trump debate, moderated by ABC News with no audience and with the candidates’ microphones generally muted when it wasn’t their turn to speak, conditions that had been negotiated by the Biden and Trump campaigns back in May.

The general consensus, at least outside MAGA world, is that Harris demolished Trump. Headlines today from relatively nonpartisan media outlets include “How Harris Roped a Dope” (David Frum for The Atlantic), “Kamala Harris Broke Donald Trump” (Peter Wehner for The Atlantic, including the line “I’ve never seen a candidate execute a debate strategy as well as Harris did”), “The Narcissist Lost the Debate” (Will Saletan for The Bulwark), “The Beginning of the End for Trump” (Bill Kristol for The Bulwark), “It Was Donald Trump’s Night–In the Worst Possible Way” (Jeff Greenfield for Politico), and “Harris Won the Debate–And It Wasn’t Close” (unbylined in Politico). At a minimum, this debate was as good for Harris as June’s debate was bad for Biden. Naturally Trump claims that he won (bigly, one presumes).

To their credit (but to the consternation of many right-wing pundits), the ABC moderators actually did some real-time fact-checking of Trump on a handful of occasions; not that Trump acknowledged his errors. The Washington Post fact-check of the debate is very lengthy, and predominantly indicts Trump rather than Harris. Many of the lies he told last night are golden oldies in the Trump lexicon, but among the newer and also most ludicrous things Trump said last night was to repeat a completely unfounded right-wing rumor that Haitian immigrants in Springfield, Ohio have been eating neighborhood cats and dogs. Trump also accused Walz, without any support, of advocating ninth-month abortions and “execution after birth”. Gerald Ford must be turning in his grave about the things Trump gets away with saying at a debate.

What a difference two months has made, and it remains difficult to imagine how these two months could have gone any better for the Democrats: a selfless decision by Biden to fall on his sword; a painless transition to Harris; a surprise VP pick that appears to have been well-received; a decided uptick in the level of optimism and hope among Democratic supporters; an excellent convention and acceptance speech; very strong fundraising; a successful effort to rebrand an incumbent Vice-President as the change candidate; and now an excellent debate performance against the most experienced Presidential debater of all time. Even so, nobody is under any illusion that the election will be anything but close.

Oh – and maybe ten minutes after the end of the debate, Taylor Swift publicly endorsed Harris, responding to Trump having recently shared AI-generated images ostensibly showing that the pop singer was endorsing Trump. Hundreds of thousands of people have already clicked on Swift’s link to a national voter registration site in the wake of her endorsement.

Yesterday was also the last day of primary elections for this cycle, with three states finally selecting their Congressional candidates. One of yesterday’s Democratic primary victors, Delaware State Senator Sarah McBride, is expected to become the first openly transgender House member this fall, in a seat that is open because the incumbent Democrat is running to replace retiring Democratic Senator Tom Carper.

In other electoral news, this week there were State Supreme Court decisions in both Michigan and North Carolina in regards to RFK Jr.’s effort to remove his name from the ballot, now that he has ended his campaign and endorsed Trump. He will no longer be on the North Carolina ballot, but he will still be on it in Michigan.

Finally, DJT stock was down 17% in early trading today after last night’s debate, but only closed down 10% by day’s end, in the mid-$16s.

It’s been a moderately eventful week, as we are now only 2 days away from the (first? only?) Harris-Trump debate.

We’d heard little from excommunicated Republican Liz Cheney in recent months; whereas other Republicans, including her fellow Jan 6th Committee member Adam Kinzinger, had participated in the DNC, she did not but was apparently keeping her powder dry for later use. This week she gave a speech in North Carolina in which she stated she would be voting for Harris. Two days later her father, the now-83-year-old former Vice President, released a statement that he too would be voting for Harris. (I’d like to imagine 2008 Dick Cheney’s reaction to being told that, 16 years later, he would be endorsing for President a black woman Democrat who was currently serving as District Attorney of San Francisco.) Liz Cheney also endorsed Rep. Colin Allred in his Senate race in Texas against Ted Cruz.

Three judicial developments this week, which I’ll address in ascending order of importance.

First: Trump made an appearance in court at the 2nd Circuit oral arguments for his appeal of the Carroll II verdict from May 2023, the $5 million defamation judgment against Trump (as opposed to the more recent, and larger, Carroll I defamation judgment). The appellate ruling is not expected to occur before the election, and news reports suggest the hearing didn’t go particularly well for Trump.

Second: Judge Chutkan held a hearing to discuss next steps in U.S. vs. Trump (D.C. edition). She was unsympathetic to Trump’s lawyers’ arguments that the existence of the 2024 Presidential campaign was relevant to the timing of how things should proceed in this trial. At the same time, she observed that it would be futile to attempt to schedule a trial date, given that there will almost certainly be a second interlocutory appeal no matter how she were to rule regarding the impact of Trump vs. U.S. on this case. In the end she has set a deadline of September 26th for the filing of the government’s brief on the presidential immunity issue, with Trump’s reply brief due October 17th. However, some portions of those briefs may be filed under seal, so it is unclear whether the public portions of those briefs will contain anything newsworthy. Chutkan also noted that Judge Cannon’s ruling that Special Counsel Smith wasn’t properly appointed isn’t relevant to this case at this time, since there is binding precedent in the D.C. Circuit that is contrary to Cannon’s ruling down in the 11th Circuit.

Third: Somewhat surprisingly, Judge Merchan decided to take the temperature down a notch during the election campaign. His ruling on Trump’s post-trial motion to vacate the guilty verdicts in New York vs. Trump in light of Trump v. U.S., previously scheduled for this week, has now been delayed until November 12th. If Trump still needs to be sentenced after Merchan issues that ruling, the sentencing has been rescheduled to November 26th.

Turning back to politics… This week there were two particularly interesting Senate polls, with mixed results. Bad news for the Democrats came from Montana, where a poll shows three-term Senator Tester trailing Tim Sheehy 45-51 (with the same poll showing Trump comfortably ahead of Harris 56-41). On the other hand there was a glimmer of hope for Democrats from a poll in Florida, where Debbie Mucarsel-Powell only trails Senator Scott 45-46 (with Trump ahead 50-45).

The latest Senate poll in Texas was less hopeful for the Democrats, showing Allred behind Cruz 44-48 (with Trump ahead 50-45), while one in Ohio showed Senator Brown holding on at 46-44 over Bernie Moreno (with Trump ahead 53-43). There was however an intriguing poll from Nebraska last week, which once again showed Harris ahead of Trump in the 2nd district, 47-42. This poll also showed 2nd district Representative Bacon significantly behind his Democratic challenger, Tony Vargas, 46-40. It also showed a surprisingly strong statewide performance by Independent candidate Dan Osborn in his race against incumbent Republican Senator Fischer (a race the Democrats are sitting out), with Fischer only ahead 39-38. Osborn, a veteran and union leader in his late 40s who was a registered Democrat until 2016, has no prior political experience.

Turning back to the campaign itself: Many observers have noted this week that Trump’s stump speeches are becoming increasingly unhinged, even by his standards. In addition to routinely talking about non-existent “post-birth abortions”, Trump has now started claiming inaccurately that schools are performing gender-affirming surgeries on trans youths. He also attracted a lot of ire this week for responding to a question about how he would address voters’ concerns about the high cost of child care with a rambling response that focused on his tariff plans. I don’t quite understand why this man is taken seriously as a candidate for President.

Finally, with Trump’s lock-up period for his DJT stock due to expire later this month, it is worth noting that DJT’s stock price has been steadily declining over the last several weeks. After briefly spiking over $40 in the immediate wake of the assassination attempt, the stock closed this week at a new post-merger low of around $17.

It’s Labor Day today.

The DNC ended a week and a half ago now, and things have been fairly quiet in recent days on the political front. It is hard to imagine that the DNC could have gone better for Harris than it did, but yet she does not appear to have benefited from a post-convention polling bump. Perhaps that reflects the fact that there are fewer swayable voters now than in past generations. As a consequence, Silver’s model currently has Harris’ chances down to 44%, but I think to some extent that reflects a methodological belief that there ought to be a post-convention bump, and hence the absence of such a bump is interpreted as a bearish sign for Harris.

The first, and as of yet only, Harris-Trump debate is now 8 days away. Trump seemingly flirted with dropping out of the debate before recommitting to it. We shall see what happens on the debate stage. Notably, Harris will be meeting Trump in person for the first time at the debate.

On the day after the DNC ended, mercurial independent candidate RFK Jr. dropped out of the race and endorsed Trump. While at times earlier this year he was polling in the low to mid teens, that was largely seen as a “pox on both your houses” vote; after Biden’s replacement by Harris, RFK’s support had dwindled down to the low to mid single digits. And with his support having been ideologically diffuse, most observers feel like his withdrawal from the race is unlikely to matter much. There may be some states where his name remains on the ballot, in fact, given the lateness of his decision to leave.

Developments on the legal front have been somewhat more interesting in recent days, although we appear to be rapidly reaching the end of the “Trump’s legal jeopardy” phase of the 2024 election.

First, Special Counsel Smith had secretly empaneled a new grand jury which has now returned a superseding indictment in U.S. v. Trump (D.C. edition). Contrary to the guess of some pundits, the superseding indictment has not been expanded to include other potential defendants. Instead, the new version of the indictment removes all references to conduct that clearly represents “official acts” under the SCOTUS decision in Trump v. U.S., such as Trump’s efforts to install Jeffrey Clark as Acting Attorney General in December 2020; indeed, Clark is no longer an unindicted co-conspirator in the revised indictment. The new version still charges Trump with the same four felonies, but does so in a way that Smith believes ought to survive judicial scrutiny in light of Trump v. U.S. Judge Chutkan will hold a hearing in three days’ time, but it is conceivable that said hearing will be the last pre-election activity in this case.

Second, Smith has also recently filed his appeal with the 11th Circuit in U.S. vs. Trump (Florida edition), arguing that Judge Cannon’s dismissal of the case (on the grounds that Smith was improperly appointed) was contrary to binding SCOTUS precedent in U.S. v. Nixon, in addition to being an incorrect reading of existing law. Trump’s reply brief will be due later in September.

Finally, we are now two weeks away from when Judge Merchan had previously indicated he would rule on Trump’s post-trial motion to set aside the verdict in New York v. Trump in light of the SCOTUS ruling in Trump v. U.S., with sentencing to follow two days later if it remains necessary. There is a wide consensus that even if Merchan does sentence Trump to jail time, that sentence will be stayed pending appeal.

With the Georgia case on hold due to interlocutory appeal, it now appears that between September 18th and the election there may be no further developments of any sort in Trump’s legal woes, thus clarifying that one of the many issues on the ballot this fall is whether or not Trump’s actions around the 2020 election and his transition away from the White House will continue to be the subject of legal scrutiny.

Tonight is day 2 of the 4-day DNC, from the United Center in Chicago. I have quite a bit of business travel this week, so I’m not going to be able to watch as much of the convention as I might want to, left to my druthers. I was able to hear most of Biden’s “One Last Time” speech, which closed out yesterday’s opening day, while sitting on an airplane; both Obamas are scheduled for tonight, then Walz tomorrow, and Harris to wrap things up on the final night.

A couple of minor political developments this week apart from the DNC. First, as expected Senator Menendez resigned today in the wake of his fraud conviction, and also has dropped out of this fall’s election where he had threatened to run as an independent. New Jersey’s governor is appointing a former aide to fill the seat temporarily, keeping the Senate at 51-49, and has signaled his intention to immediately appoint the winner of November’s election so as to give that winner (presumably, Representative Kim) seniority over other members of the entering Senate class. Menendez’s sentencing is scheduled for late October. Second, former Representative Santos has pled guilty to felony charges, avoiding a trial that was set to start in a couple of weeks. His sentencing is scheduled for early February and he is expected to serve jail time; indeed, as part of his plea deal he agreed to not appeal any sentence that is less than 8 years.

I wanted to make a few further observations related to my most recent post about the state of the race, pre-DNC.

In that post I discussed a highly plausible scenario in which Harris wins 270-268. That scenario starts with the Biden-Trump map, but with the 3 “Sun Belt” states of Arizona, Georgia, and Nevada flipping back to Trump, while Harris retains the 3 “Rust Belt” states of Michigan, Pennsylvania, and Wisconsin. If that scenario had happened in 2020, it would have been a 273-265 win for Biden. As such, in a sense population shifts between the 2010 and 2020 censuses have transferred 3 electoral votes to Trump, relative to four years ago.

That 3-vote transfer has elevated the potential importance of Nebraska-2, the electoral district containing Omaha, in the 2024 election. Four years ago, it was possible to imagine a 270-268 scenario in which NE-2 was decisive, namely if Trump had prevailed in both Georgia and Pennsylvania while everything else stayed as it was. In today’s map, that particular outcome (Trump winning Pennsylvania while losing Arizona and Nevada and Omaha) seems much less likely, but if it were to occur then it is Trump, not Harris, who would be a 270-268 winner.

Walz rallied in Omaha over the weekend, and today a new poll came out showing Harris ahead 50-42 in NE-2. On the strength of that new poll, Silver’s model now has Harris at 78% in NE-2, up from 67% last week. Without Omaha’s single electoral vote, losing the 3 Sun Belt states would put Harris into a 269-269 tie, which would throw the election to the House of Representatives where the Republicans currently have a majority of state delegations.

Indeed, prior to Biden’s departure from the ticket, this particular Omaha-driven 270-268 Biden victory was the most likely path to victory for Biden, with the 3 Sun Belt states seemingly having fallen out of reach. But with the Harris coalition running stronger than the Biden coalition in not only those 3 Sun Belt states but also North Carolina — which in today’s version of Silver’s model has moved ahead of Georgia on Harris’ list of targets, with Trump at 63% to retake Georgia but only 60% to keep North Carolina — the Democrats are now looking at multiple potential paths to electoral victory.

Nate Silver decided against trying to implement a formal Senate electoral model for the 2024 cycle, but the math appears difficult for the Democrats to hold on to the Senate. Today things sit at 51-49, but Manchin’s seat in West Virginia is a lost cause for the Democrats. Additionally, the Democrats have two vulnerable incumbents in red states: Brown in Ohio, and Tester in Montana. The good news is that with Harris on the top of the ticket, other Senate incumbents like Rosen (NV) and Baldwin (WI) no longer appear to be as vulnerable as it once seemed, and the open races in AZ and MI seem much more winnable for the Democrats. The bad news is that if either Brown or Tester loses, it is very unclear from where the Democrats could possibly pick up a 50th seat. The only two possibilities on the map, and they both feel like real longshots, are defeating Scott in Florida or Cruz in Texas. I think an interesting possibility would be whether either Murkowski of Alaska or Collins of Maine, both of whom seem increasingly detached from the modern Republican party (particularly on reproductive freedom), could be enticed to switch parties and provide a 50th vote.

It’s even harder to know what’s really going on in the House, but if you believe the Cook Political Report the most likely result is a 221-214 Republican edge. If the Democrats are to prevail in the House, they are going to need to find a way to pick up most if not all of 6 Republican-held swing districts in California and New York, as well as perhaps take the CO-3 district that Boebert almost lost last time and has now abandoned, and/or knock off moderate Republican Don Bacon in NE-2, who is still ahead 46-44 in the same new poll that had Harris leading 50-42.

As such, despite the wind at the Democrats’ backs over the past three weeks, it is still somewhat easier to imagine a Republican sweep in November than it is a Democratic sweep.

No surprises in either of the primary elections in which I voted yesterday: Senator Klobuchar won 94% of the vote, and Representative Craig (MN-2) won 91%. Elsewhere in Minnesota, Representative Omar (MN-5) managed to stave off Don Samuels for the second straight primary election, expanding her margin from 50-48 in 2022 to 56-43 in 2024. Interestingly, AIPAC appears to have sat the Omar-Samuels primary out.

I thought this would be a good point in the campaign — long enough after the assassination attempt and the RNC and the Democrats’ change of candidate, but still before the DNC — to take stock of the state of the race on a state-by-state basis, using Nate Silver’s model as a guide.

Let’s start by counting the “very safe” electoral votes for both sides, which for this purpose I’ll define as a 95.5% or greater chance (i.e., two-sigma) of victory in Silver’s model:

Very Safe Democrat (170): California (99.6%), Connecticut (97.9%), District of Columbia (100.0%), Hawaii (95.6%), Maryland (99.7%), Illinois (98.7%), Maine-1 (97.7%), Massachusetts (99.3%), New Jersey (96.0%), New York (98.2%), Rhode Island (96.8%), Vermont (98.7%), Washington (98.4%)

Very Safe Republican (103): Alabama (99.1%), Arkansas (99.2%), Idaho (99.2%), Indiana (97.7%), Kansas (97.8%), Kentucky (99.4%), Louisiana (97.7%), Missouri (98.8%), Nebraska at large (99.0%), Nebraska-1 (97.4%), Nebraska-3 (99.7%), North Dakota (98.9%), Oklahoma (99.7%), South Dakota (98.3%), Tennessee (99.5%), Utah (96.2%), West Virginia (99.3%), Wyoming (99.6%)

All of those feel like sure things in practice, although I am a little surprised that Trump’s odds in Hawaii and New Jersey are as good as 25-to-1. I also have fantasies that Walz’s presence on the ticket can increase the Democratic chances in Nebraska’s 1st district, in which he was born.

Next, let’s look in descending order at the remaining states that each party expects to win, which for this purpose I’ll define as an 80% chance of victory in Silver’s model.

Expected Democrat (51): Delaware (94.0%), Oregon (92.7%), Minnesota (90.9%), Colorado (89.8%), New Mexico (85.3%), Virginia (81.7%), Maine at large (80.3%)

Expected Republican (82): Montana (94.9%), South Carolina (94.4%), Mississippi (92.9%), Ohio (89.6%), Iowa (86.3%), Texas (84.1%)

While some of the states in each column used to be swing states, given our current politics it would be surprising for any of these to defect to the other side, although both Virginia and Texas bear monitoring.

At this point, if everything were to have proceeded according to expectations then the Democrats would be ahead 221-185.

In descending order, here are the top Democratic targets to get from 221 to 270, per Silver’s model:

• New Hampshire (4) = 76.4%
• Nebraska-2 (1) = 67.1%
• Michigan (15) = 66.5%
• Wisconsin (10) = 63.8%
• Pennsylvania (19) = 59.1%

If Harris wins those and loses all the other swing states, she wins 270-268.

Conversely from the Trump perspective, in descending order here are the top Republican targets per Silver’s model to get beyond 185:

• Alaska (3) = 78.3%
• Florida (30) = 77.2%
• Maine-2 (1) = 74.0%
• North Carolina (16) = 60.7%
• Georgia (16) = 55.4%
• Arizona (11) = 51.6%
• Nevada (6) = 46.3%

Winning those states is how Trump gets to 268, which means he would need to add, say, Pennsylvania (40.9%) to get over the hump. Hence, the conventional wisdom that Pennsylvania is 2024’s tipping point state.

Three closing notes. First, as these probabilities suggest, right now you’d rather be in Harris’ shoes than Trump’s; today Silver’s model gives Harris a 56.7% chance of an electoral college victory, on an average nationwide popular vote advantage of 2.5%. Second, with increased swing-state polling in the past couple of weeks it no longer appears to be true that Trump’s “electoral college edge” in popular vote terms has widened from 2.0% to 2.5% with the replacement of Biden by Harris. Finally, it’s interesting to see Trump potentially needing to play defense in not only North Carolina, but perhaps also Florida, particularly in light of the abortion issue being on the ballot there.