Everybody loves “outsider” candidates. Until they find their Reddit posts.

It’s hard to know how voters will react to inexperienced candidates like Graham Platner. But that’s what makes them high-risk choices in must-win races.

Platner at a rally in Ogunquit, Maine yesterday. Getty images.

Even before the latest series of revelations, including a highly questionable choice of tattoos and a social media posting history that you might describe as “problematic”, I’m not sure that I’d have recommended a vote for Graham Platner on an “electability” basis. But I can understand why voters find him relatable.

About half the elected officials I’ve met have the personality of wax figures. When your every move is scrutinized, and you spend at least half your time fundraising and glad-handing constituents, the rough edges tend to get sanded off. Platner, by contrast — a 41-year-old former Marine and current oyster farmer from Sullivan, Maine — seems like a guy you’d actually want to get a beer with.

I even have a weirdly personal connection to Sullivan. My grandma, who was a brilliant architect in an era where it was hard for women architects to find work, designed a family home in Sullivan, a tiny town of 1200 people along U.S. Route 1, and I went there every summer as a kid.

But evidently, I’m not alone. A University of New Hampshire poll out this morning found Platner leading Janet Mills 58-24 among Democratic primary voters in Maine.

The internals of the poll aren’t so bad for Mills, the current two-term governor, who only recently declared her candidacy: 65 percent of primary voters have a favorable impression of her, against 16 percent unfavorable.

Platner’s splits were a stellar 60/8, by contrast. But the poll was conducted from Oct. 16 through Oct. 21. Although some of the news about Platner’s shitposting history dropped while the poll was in the field, the most serious revelations hadn’t. (Google data suggests there’s been a massive uptick in searches related to Platner only within the past 48 hours.) Instead, Platner had been getting largely favorable media coverage, not just for his compelling biography but also for his big rallies with progressive stars like Bernie Sanders.

In the past week, however, Platner was discovered to have a skull-and-crossbones tattoo, pictured in a shirtless and nearly pantsless video from his brother’s wedding (safe for work, barely). The tattoo strongly resembles the Totenkopf, a symbol used by the Nazi 3rd SS Panzer Division. Platner said he got the tattoo, along with other Marines, on a drunken trip to Croatia in 2007; he recently covered it up with another tattoo, which he showed off in a shirtless interview with a Portland TV station.

Platner’s Reddit handle, “P-Hustle”, has also been uncovered. And P-Hustle’s posting history left something for pretty much every species of Democrat from left to center to object to:

• As recently as 2021, Platner made a series of anti-gay comments.

• He cited his experience as a bartender to say how “solid this stereotype is” that Black people don’t tip well.

• Platner referred to himself as a “communist” and said all cops were bastards.

• He contradicted another commentator who said “White people aren’t as racist or stupid as Trump thinks”. “Living in white rural America, I’m afraid to tell you they actually are,” P-Hustle replied.

• And Platner downplayed concerns about sexual assault and appeared to endorse political violence. “Tell them that if they expect to fight fascism without a good semi-automatic rifle, they ought to do some reading of history,” he wrote.

At first glance, Platner might appear to be a paradigmatic example of a “Milkshake Duck” candidate, somebody whom everyone on the Internet fell in love with — until they discovered his posts:

But even after the news, Platner has had his share of defenders, and they’ve scrambled the usual Democratic coalitions. For instance, Platner tried to get ahead of the tattoo story by sharing the shirtless wedding video in a largely sympathetic Pod Save America interview.

The politics of the PSA boys are slightly complicated, but they aren’t typically what you’d associate with the anti-establishment, Bernie-endorsed wing of the party. Meanwhile, other liberals like John Ganz, the thoughtful but hardly establishment-friendly writer of the Substack Unpopular Front, have deemed the Totenkopf tattoo a line that is unacceptable to cross, even if they buy Platner’s backstory for how and why he got it.

The UNH poll aside, prediction markets see the primary as a toss-up. And I’m honestly not sure which side I’d bet. My instinct was that these intraparty frictions would be hard for Platner to navigate, especially given that some (especially older) voters would prefer Mills to begin with. But I also expected Mills to start with a higher floor of support than the UNH poll found her at — the first and, so far, only nonpartisan poll of the race.

A few words about candidate “authenticity”

I don’t personally take offense easily, and I’m not sure that I have any interest in weighing in on precisely which of Platner’s past actions you ought to find unforgivably offensive given his self-description as a “retired shitposter”. Indeed, I find myself with some contradictory impulses.

On the one hand, I don’t think we should expect people to self-police their social media postings in case they might one day run for political office.

On the other hand, this isn’t really a matter of cancellation: nobody is saying that, for instance, Platner’s oyster farm should be taken away from him. Rather, Democrats have to choose exactly one from among the roughly 1 million Mainers who are at least 30 years old and U.S. citizens, and therefore eligible to run against Susan Collins in a race that could hardly be more vital to their chances of retaking the U.S. Senate in 2026. (Or more likely, 2028.)

I’m also not sure what to make of Platner’s redemption arc. He’s been more “authentic”-seeming than your typical politician in interviews, but “authenticity” itself can be performative. Was there a little too much tryhard when Platner told The Advocate that his views of gay people began to change when he “attended showtunes night at JR’s,” a Washington, D.C. gay bar? That itself is a little stereotypical; I’m a gay guy who would be happy enough never to hear another showtune again. Or when Platner wrote in a Reddit AMA that he’d “stand right in the fucking way of anyone who’s going to try to come after the freedoms of the LGBTQIA+ community”, is that authentic, or is that overcompensating?

And Platner already had a complicated history. What to make of the fact that he served in the Iraq War even after protesting against it?

On the third hand — we’re quickly running out of hands — as I said in this week’s SBSQ, I’m not sure I really care what’s in a politician’s heart anyway. You’re not voting for a senator to get a beer with him. You’re doing so because you hope that he’ll reliably represent your point of view should he be elected. Or if not, that he’ll make some tactful choices to win elections against other candidates you’d find even worse.

The latter consideration is especially important when a candidate is running for federal office. I’m not sure that I want to get into how I plan to vote in New York’s mayoral election next month. But that race has strictly local implications, so I might be more willing to gamble on a higher-variance candidate that sent the right message. In the Senate, by contrast, more and more voting takes place strictly across party lines. Either Mills or Platner would probably reliably oppose Trump and support any Democrat who won the White House in 2028.

Should he win the primary, Platner is almost certainly a higher-risk, if perhaps also higher-reward choice than Mills. As Eli concluded last month, if you think Collins is more likely than not to lose against a “generic Democrat”, there’s no reason to take chances. And for what it’s worth, Mills, if an unexciting choice, would rate considerably higher than a “generic Democrat” when we turn on our midterms model next year because of her status as an elected governor. As I’ll discuss in a moment, experience in elected office is a positive statistical indicator.

Then again, if you think Mills is another Sara Gideon in the making, then maybe you want to take a risk.

Experience in elected office is usually a positive sign

Most of all, though, I do not expect the views expressed by national political pundits — whether it’s me, Ganz, or the Pod Save America guys — to have any correlation whatsoever with how voters in Maine will feel next November. And really, that’s precisely why I think it’s worth paying attention to the statistical regularities that our model cares about.

One of those is that more moderate candidates usually do better, holding other factors constant. For some very detailed thoughts on this, see this newsletter from August. (Some of the recent academic work denying this is complete crap.)

At the same time, “usually” and “holding other factors” constant are crucial qualifiers. The Democratic Party is extremely unpopular at the moment, particularly establishment leaders like Chuck Schumer. So Platner’s lack of partisan loyalty could be an advantage. He’s at least had the sense to acknowledge, for instance, what overwhelming majorities of voters did: that Joe Biden’s decision to run for reelection last year was a terrible idea. (Though this could also be interpreted as an implicit shot against Mills, age 77, who would be the oldest first-term senator.)

Moreover, while Platner might be an economic populist, he obviously hasn’t trafficked in progressive/woke cultural tropes — quite the opposite, until recently, at least if you’re reading from “P-Hustle”. Platner also holds some more centrist positions that he hasn’t disavowed, such as his stance on guns. It’s the culture stuff that’s probably Democrats’ bigger problem at the moment, especially in a working-class state like Maine. Having a candidate who “codes” as a regular guy whom you’d like to get a beer with might actually count for something.¹

But another longstanding trend we’ve observed in our Congressional model is that more experience in elected office is predictive of better general election performance, holding other factors constant. And this is easier to quantify than ideology; we’ve typically used a 4-point scale:

Tier 4: Senator or governor.

Tier 3: U.S. representative, or statewide elected office (e.g., secretary of state), or mayor of a city with a population larger than that of an average-sized congressional district.

Tier 2: Any other nontrivial elected office (e.g., state senator).

Tier 1: Has never held any nontrivial elected office.

By this measure, Mills is a Tier 4 candidate as an elected governor, while Platner is a Tier 1; he’s never held elected office before or even run for it. And this isn’t necessarily true of other “outsider” candidates. Bernie Sanders, for instance, ascended the ranks from mayor of Burlington, Vermont, to U.S. Representative to U.S. Senator, and he often won his races by impressive margins even considering the blueness of Vermont.

Even Zohran Mandani has won three terms to the New York State Assembly, including defeating an incumbent Democrat in his first race in 2020. That helps to explain why he’s consistently hit his marks during the campaign.

Why this advantage for experienced candidates? Don’t voters basically hate all politicians? Well, no. They feel more warmly toward their own officeholders than toward politicians in the abstract — by definition, warmly enough to usually re-elect them.

There are also a few hypotheses that we can rule out. While incumbents and other experienced candidates often have stronger fundraising networks, Platner has had no trouble at all raising money. But our model finds this experience advantage even after controlling for fundraising.

Meanwhile, inexperienced candidates will have been less well-vetted, as we’re discovering with Platner. That might explain some of the difference. Still, our model actually does attempt to control for political scandals, and we still find this effect despite that.

Side note: Competitive primaries are good, actually. I don’t know whether Mills’s opposition researchers are behind the recent revelations about Platner. But imagine the predicament Democrats would face if the news had come out next October, not this one.

Instead, the reason experience is a positive indicator is probably just that we pundits have a great deal of trouble relating to ordinary voters. So the empirical evidence of having actually won elections, especially for more competitive offices, is far stronger than our opinions.

It can be hard to come up with reliable rules for which candidates will resonate with voters. At the moment, for instance, Schumer trails Alexandria Ocasio-Cortez in polls of a prospective New York U.S. Senate primary in 2028. But even at a time when moderate candidates are out of vogue, Pennsylvania Gov. Josh Shapiro is extremely popular in his home state. Meanwhile, some candidates, like the current U.S. president, are impervious to virtually any scandal, while others can’t survive allegations that would barely register for somebody like Trump.

I can imagine Platner’s Reddit shitposts being anything from a substantial liability with general election voters next November to a nothingburger to even a slight advantage if voters think they make him seem more authentic. It’s more complicated still because various of the comments can be critiqued from either a liberal or a conservative direction. The anti-gay comments will probably be more of a problem in the primary — but the posts about white, rural voters could be the bigger issue next year in one of the nation’s whitest states.

And I’m not even sure how to assess the electoral impact of the tattoo. These days, it’s usually Democrats who accuse Republicans of being “secret Nazis” and not the other way around.

Voters often forgive a controversial past. Our research into scandals has also found that if a candidate has been elected or reelected despite a scandal, it basically has no further impact from that point forward. Only newly-revealed scandals matter, in other words. After that point, it gets priced into the equation by voters — and it’s old news.

But Platner hasn’t been elected to anything, ever — and the news is fresh. How well he navigates the revelations over the next few weeks will tell us something. So might prospective general-election polls against Collins, though polling in Maine tends to be both sparse and rough. (The polls considerably underestimated Collins in 2020, for example.)

Mills, moreover, is precisely the sort of candidate who might be appreciated more by voters in Maine — incidentally, the oldest state in the country — than by younger national pundits who are jockeying for online influence. So as much as I usually criticize Democrats for being too risk-averse, I’m just not sure that Platner is a smart risk to take.

1. Furthermore, non-incumbent candidates may be more flexible in defining their ideology to voters. In our model, our indicator of ideology/partisanship is based on roll-call voting in the U.S. Congress, so it applies only to candidates who have already served in Congress.