As different attitudes towards the pandemic lead Rs and Ds in different directions in terms of campaign tactics, I wrote up the debate at New York:

Back in the 1990s, when disciples of James Carville dominated the ranks of Democratic campaign professionals, you could always get a big rise out of any of them by suggesting they should deploy more yard signs and billboards. Their contempt for such old-school tools was near-complete, as was their faith in a concise, poll-tested message conveyed heavily on TV. You saw an echo of this contrast in partisan preference in 2012, when Barack Obama’s data-hip team quietly mocked the conservative pundits aflutter over Mitt Romney’s yard-sign advantage.

Now even in the 1990s, Democrats didn’t completely eschew hard-core, visible campaign methods, as anyone who has ever heard a sound truck in an urban neighborhood conducting aptly named “knock and drag” get-out-the-vote operations can attest. Indeed, as partisan polarization has reduced the number of persuadable swing voters, such occasionally noisy mobilization efforts have become more important in both parties.

But this year the COVID-19 pandemic has understandably led Biden-Harris strategists to rely on more remote mobilization techniques, at least so far. And the contrast with the extremely visible advocates of MAGA, who are far less inhibited by public-health guidance, is making some Democrats nervous, as Time’s Charlotte Alter discovered in Michigan:

“This year, 83-year-old former Chrysler employee [Don Sabbe] says he’ll definitely vote for Joe Biden, but he’s getting concerned about Biden’s campaign here in Michigan.

“’I can’t even find a sign,’ Sabbe says outside a Kroger’s in Sterling Heights, where surrounding cars fly massive Donald Trump flags that say ‘No More Bullsh-t’ and fellow shoppers wear Trump T-shirts for their weekend grocery runs. ‘I’m looking for one of those storefronts. I’m looking for a campaign office for Biden. And I’m not finding one.’

“The reason Sabbe can’t find a dedicated Biden campaign field office is because there aren’t any around here. Not in Macomb County, the swing region where Sabbe lives. It’s not even clear Biden has opened any new dedicated field offices in the state; because of the pandemic, they’ve moved their field organizing effort online.”

Hilariously, when Alter asked one Biden campaign staffer how many people they had on the ground in Michigan, she was asked in turn: “What do you mean by ‘on the ground’?”

By contrast, for all its alleged social-media savvy and its heavy TV presence, the Trump-Pence campaign is very physically in-your-face, from the raucous live rallies the president loves so much to the boat parades and the huge flags and signs and every other campaign resource of the 1950s. The MAGA folk may choose to ignore polls because they allegedly miss “shy Trump voters.” But “shy” is not the word that comes to mind when you encounter bellowing red-hatted fans of the president eager to show their lack of political correctness.

In choosing a different approach, the Biden campaign is practicing what he preaches in the way of responsible behavior during a pandemic. The candidate’s representatives say they are compensating for the lack of sound and fury in other ways, according to Alter:

“Biden’s Michigan team says its campaign is significantly bigger than Clinton’s and may be the largest program in the state’s history. The campaign says it reached out to 1.4 million voters during the Democratic convention and the weekend that followed, with 500 digital-organizing events and 10,000 volunteer signups. In the week before Labor Day, the campaign sent 500,000 texts to Michigan voters — one every half-second. It has just replaced the trappings of a traditional ground game — volunteers knocking on doors, distributing literature, and so forth — with a digital field operation.”

The question is whether there’s something about the loud-and-proud Trump effort that is somehow contagious, or that helps build enthusiasm and willingness to vote in a tangible way. Looking at it conversely, does a field operation without physical voter contact forfeit something essential?

“Democrats are used to measuring their strength by their ground game, and without physical boots on the ground, the effect can be unsettling. It brings up uncomfortable questions about whether a ‘digital field’ operation can really replace a ‘traditional field’ operation without something being lost. Sure, it sounds like digital field organizing should work. But does it actually? Nobody knows, because it’s never been tried on this scale before.”

The polls look good for Biden in Michigan, but they looked good for Clinton in 2016, too — though fewer of them were being taken. If nothing else, though, perhaps the dominant physical presence of the Trump campaign and its supporters will help prevent the kind of Democratic overconfidence that may have done in HRC.

I decided to compare Biden’s poll position to those of the Democratic nominees in the last four elections, and wrote up the results at New York:

Fifty days from Election Day (November 3), Joe Biden leads Donald Trump in the RealClearPolitics polling averages nationally by 7.4 percent (50.5 to 43.1). He’s led every day of the last year, by margins ranging from 11.8 percent on September 17, 2019, to 4 percent on January 24, 2020.

Biden’s current lead is the largest a candidate has held at the 50-day mark in any of the last four presidential election cycles. In 2004 George W. Bush led John Kerry by 5.7 percent; he would ultimately win by 2.4 percent. In 2008 John McCain actually led Barack Obama by 1.3 percent; he would eventually lose by 7.3 percent. In 2012 Obama led Mitt Romney by 2.8 percent 50 days out; his actual margin was 3.9 percent. And in 2016 Clinton was up 1.3 percent 50 days out; her final popular-vote margin was 2.1 percent.

Those hoping for a late Trump surge to victory might take heart from Obama’s 8.6 percent turnaround in the last 50 days of the 2008 race. But there were some very different circumstances that year. Obama had led for most of the general election campaign before McCain got a sizable post-convention bounce. And some big bad events helped Obama turn it back around against the candidate of the party in power: Exactly 50 days from Election Day, Lehman Brothers declared bankruptcy, triggering a global financial crisis. Nine days later, the first Katie Couric interview with McCain running mate Sarah Palin was aired, which had a big negative effect on Palin’s once-formidable popularity. And the day after that, Obama ran circles around his opponent during the White House summit on the financial crisis hosted by George W. Bush.

It is possible Trump will get those kind of late breaks, but unlike Obama in 2008, he’s now the incumbent president with a consistent “very unfavorable” rating in the polls hovering at or just under 50 percent. Heavy early voting this year means that with each passing day the slice of the electorate (with an already-low undecided vote) that could be “turned around” by a Trump surge is shrinking. And Biden’s polling lead is enhanced by the fact that most national pollsters have already completed the “switchover” to a likely voter screen that often benefits Republican candidates.

While Trump partisans trash the national polls as inaccurate “like they were in 2016” (they actually weren’t), and mindlessly claim his manifest greatness will generate a landslide win, his best hope remains an Electoral College advantage that could again give him a narrow win despite a popular-vote deficit. From state polls and what we can infer from regional trends, Trump might well pull off an upset if he gets Biden’s national lead down to around three points.

But again: Trump hasn’t been within three points of Biden in the RCP national polling averages even once in the past year. So it might take something exceptional to make that happen now.

Looking back at something I wrote at the beginning of the election cycle about the strategic battleground, I revisited the key question at New York:

Soon after the 2018 midterms, I reviewed the evidence about where Democrats made gains and suggested it still wasn’t clear which strategy the party should adopt in trying to recover from the 2016 loss to Trump:

“The two most obvious regional strategies for Democrats are to win back the heartland/Rust Belt (depending on how you think about them) states that Trump narrowly carried despite a strong history of going the other way: Michigan, Pennsylvania and Wisconsin. There are two similar additional states that Obama carried twice: Iowa and Ohio.

“At the other end of the spectrum are Sun Belt states that were already quite close (Florida and North Carolina) or that have recently been trending Democratic (Arizona, Georgia, and Texas) at varying rates.”

The midterm numbers didn’t really indicate one path or the other:

“In the end, the Democratic presidential strategy for ejecting Trump in 2020 will follow the polls — hopefully better and more frequent polls than those taken by the Clinton campaign in those heartland states that ultimately cost them the election of 2016 — and perceived opportunities.”

So here we are less than two months away from Election Day, and with early voting beginning almost immediately, and the best strategic path is still unclear. According to FiveThirtyEight’s polling averages, Joe Biden leads Donald Trump nationally by 7.5 percent. If Biden maintains that sort of lead through Election Day, then the Electoral College will take care of itself and the Democrat will win very comfortably across the range of battleground states in both competitive parts of the country.

But if the national race tightens, the battleground situation gets much more complicated. Again using FiveThirtyEight’s polling averages, Biden’s currently trailing in Georgia and Texas, and has a lead under 2 percent in Florida and North Carolina. Of the realistic Sun Belt targets for Democrats, Biden has a robust lead only in Nevada (6.4 percent), which Democrats carried in 2016, and Arizona (4.7 percent).

In the competitive Rust Belt states, Biden’s national lead is matched only by his advantage in Wisconsin (7.5 percent). He has a decent cushion as well in Michigan (6.6 percent) and Minnesota (6.2 percent), and a slimmer one in Pennsylvania (4.1 percent). He’s trailing, however, in Ohio and Iowa, which means they probably become winnable only in the midst of a big Biden victory.

So all in all it looks like a Rust Belt strategy makes the most sense for Biden, right? Ron Brownstein suggests that could be the case:

“Exactly eight weeks before Election Day, Biden has strong opportunities to recapture states that President Donald Trump won in 2016 both in the Rust Belt and the Sun Belt. But public and private polls consistently show that Biden is running slightly better in the former group of battlegrounds — centered on Michigan, Pennsylvania and Wisconsin — than the latter, which include North Carolina, Florida and Arizona …

“[Biden’s] potential to improve on Hillary Clinton’s showing with older and blue-collar Whites means that even if he falls short in some or all of the Sun Belt states that many in the party see as its long-term future, he could still reach 270 Electoral College votes by recapturing Michigan, Pennsylvania and Wisconsin, the three big Rust Belt states that Trump dislodged from the Democrats’ ‘blue wall.'”

It’s easy to forget that — despite the blue trends in the Sun Belt and the red trends in the Rust Belt — Trump did better in the former in 2016, notes Brownstein:

“Trump and the Republican Party have demonstrably lost ground across both regions. But because Trump began with less margin for error in the Rust Belt states, the consequences of that erosion have been more severe for him there than in the Sun Belt.”

One big obstacle to a Biden Sun Belt strategy is the relatively high resistance of white voters — both college-educated and noncollege-educated — to Democrats in the South, which is probably the combined effect of high levels of Evangelical conservative religious affiliation and racial polarization in states with large minority populations.

Polling aside, there are reasons Team Biden might want a selective Sun Belt strategy in the home stretch. Florida, in particular, is a state Trump barely carried in 2016 and can really not afford to lose this year. But the current situation in the Sunshine State offers a reminder that it’s not just Democrats who are capable of geographic targeting. Trump is remaining competitive in Florida in no small part because he is doing extremely well among Cuban Americans and other Latinos concentrated in south Florida. These voters may have been the most important targets of all that endless and redundant socialist-bashing at the Republican National Convention.

It’s entirely possible that Trump simply cannot be reelected this year (legitimately, anyway) thanks to his poor handling of COVID-19, the collapse of his “greatest economy ever,” and accumulated public disgust with his character and personality. And it’s also possible that nothing the campaigns do will significantly affect the results in individual states. But at the margins, small things could make a big difference in a crazy-close election. The odds are high that if it is crazy-close, we’ll again be looking at the Rust Belt states that shocked the world in 2016.

With voting by mail about to begin, the first clear statistics on who will cast mail ballots are significant, so I wrote it up at New York:

Until now, the prospect of people voting by mail in unprecedented numbers in November has been mostly speculative, based on what happened in this year’s primaries, along with polling, and a general long-term trend towards this method of voting. But now as states are releasing data on general election mail ballot requests, it’s all getting very real, as the Associated Press reports:

“Mail balloting is set to begin Friday in the presidential election as North Carolina starts sending out more than 600,000 ballots to voters — responding to a massive spike in requests that has played out across the country as voters look for a safer way to cast ballots during the COVID-19 pandemic….

“In 2016, just one-quarter of the [national] electorate cast votes by mail. This time, election officials expect the majority of voters to use the method. Wisconsin has already received nearly 100,000 more requests than it did in the 2016 election. In Florida, 3,347,960 people requested ballots during the 2016 election. The state has already received 4,270,781 requests.”

And so far (from states that can and do track the party ID of mail ballot applicants) the Democrat tilt of those who plan to vote by mail is unmistakable:

“The GOP has historically dominated North Carolina mail voting, but this year the people asking for the ballots are generally not Republicans. Democrats requested more than 337,000 ballots, and independents 200,000, while only 103,000 were sought by Republicans. Voters in the state can continue to request the ballots up until Oct. 27, though that may be too close to the Nov. 3 election for them to receive the ballot and return it to their local elections office in time.

“The Democratic lead in mail ballots isn’t only in North Carolina. In Maine, 60% of requests for mail ballots have been made by Democrats and 22% by independents. In Pennsylvania, Democrats have requested nearly triple the number of absentee ballots as Republicans. In Florida, where the GOP once dominated mail voting, 47.5% of requests have come from Democrats and 32% from Republicans.”

Now normally, that would indicate higher general interest in voting among Democrats, and perhaps an impending blue tsunami. But the president’s months-long campaign against voting by mail is clearly a big, and perhaps the biggest, factor in creating this partisan tilt, so it’s unclear whether it would exist otherwise. The other implication of the early evidence on partisan splits in willingness to vote by mail is that the Red Mirage scenario – where the first votes counted on Election Night are in-person ballots that skew heavily Republican, leading to a premature Trump victory claim coupled with delegitimization of subsequently counted mail ballots as fraudulent – remains a very real threat.

Given the reaction — and arguably over-reaction — to Joseph Kennedy’s III primary loss to Ed Markey in a Massachusetts primary, I offered some ruminations about earlier setbacks to the famous political dynasty at New York:

Coverage of Joseph Kennedy III’s failed primary challenge to Senator Ed Markey has been described as signaling the end of the family dynasty he young Joe often seemed uncomfortable representing. That feels unfair, since the Kennedy clan’s reputation for political invincibility began to slowly unravel many years ago.

It’s true that Joe III’s campaign broke a long Kennedy winning streak in Massachusetts, as the Boston Globe noted:

“The powerhouse political family has dominated races in Massachusetts for decades, notching more than 30 primary and general election wins and zero losses since John F. Kennedy won a seat in Congress in 1946.”

It was also the first Kennedy loss in 13 U.S. Senate races (two by JFK, one by RFK, and nine by Ted Kennedy). But it was hardly the first big family disappointment.

Joe’s grandfather Bobby suffered the first Kennedy loss in any electoral contest over a half-century ago in 1968, when he lost an Oregon presidential primary to Eugene McCarthy a week before posting a comeback win in California — but was then assassinated after his victory speech.

Four years later, former Peace Corps director Sargent Shriver, a Kennedy by marriage, agreed to become the replacement for vice-presidential nominee Tom Eagleton (who was pushed off the ticket after the media discovered his history of drunk-driving arrests and shock treatments) on the horrifically unsuccessful presidential slate of George McGovern, which lost 49 states. McGovern had actually been the stand-in for RFK at the 1968 convention.

In 1980, Ted Kennedy’s presidential challenge to Jimmy Carter got off to a disastrous start, as he lost 12 of the first 13 primaries and caucuses, winning only Massachusetts. Though he later rallied with some big wins, it was all too little, too late, and Ted never ran for president again.

The next generation of Kennedy pols had a lot of problems. Joseph Kennedy II, the most recent candidate’s dad, chose to retire from the U.S. House in 1998 after an embarrassing scandal over his efforts to secure an annulment and marry his secretary. His sister Kathleen Kennedy Townsend served two terms as lieutenant governor of Maryland but then lost the governorship to a Republican in a huge 2002 upset. Cousin Mark Kennedy Shriver served two terms in the Maryland legislature before losing a congressional primary to (now-senator) Chris Van Hollen.

Ted Kennedy’s son Patrick won a congressional seat in Rhode Island in 1994 and made it into the House Democratic leadership, but in 2010 he gave up his seat and his electoral career after struggling with mental illness and drug addiction. Until Joe III won a Massachusetts congressional seat in 2012, there was a brief moment when no Kennedy was serving in elected office. JFK’s famous daughter, Caroline Kennedy, briefly considered a Senate bid in New York when Hillary Clinton became secretary of State, but she withdrew from consideration. Yet another RFK child, Chris Kennedy, ran for governor of Illinois in 2018 but finished third in the Democratic primary.

That’s a lot of family trial and error that should not be laid at the doorstep of Joe Kennedy III. As the Globe notes, it’s entirely possible the dynasty could make a comeback:

“There are other, younger Kennedys who could enter the political arena — Jack Schlossberg, JFK’s grandson, appeared in a video with his mother, Caroline, at the Democratic National Convention, quickly sparking memes about his resemblance to his handsome uncle, the late John Kennedy Jr. And Ted Kennedy’s grandson, Edward Kennedy III, has expressed an interest in politics. Amy Kennedy, who married Ted Kennedy’s son Patrick, recently won a Democratic House primary in New Jersey.”

And the latest Kennedy candidate is himself awfully young to retire from politics. Perhaps next time around, he can run as just “Joe.”