I wasn’t a believer either. The 2016 version of Bernie Sanders followed an insurgent model that doesn’t play well as an encore. If you’ve ever seen Hangover 2, you know a sequel with the same plot as the original falls flat. Sure, he had a base of true believers, but many of his primary voters were protesting Hillary Clinton.
There were so many more choices this time. Then Elizabeth Warren took off. An early July national poll from The Economist/YouGov had Bernie at 9%. It was a bit of an outlier, but still. We’re talking about the runner-up from last time, with almost 100% name recognition.
His remaining supporters were paying less attention to the campaign than any other candidate’s. Many surmised all but the most dedicated Berners would abandon him as soon as they woke up from their political hibernation. Conversation was more about how much of a spoiler he would be if he stayed in after losing early primaries. Would this cost Warren her nomination?
A funny thing happened on the way to oblivion. Check out his numbers in the new early state polls from CBS/YouGov:
Iowa: 26% (2nd, -3)
New Hampshire: 25% (3rd, -2)
Nevada: 29% (1st, +2)
We know better than to leap to conclusions based on a single pollster. This is a credible one, used by the DNC as qualifiying data for debate participation.
Though CBS/YouGov surveys are usually somewhat favorable to Sanders, they aren’t outliers. And his trend is favorable. Bernie is up 5 points since they last polled New Hampshire. He’s 7 ahead of their last Iowa reading.
The conventional narrative and previous numerical evidence told a story where Sanders was in the low-mid 20s at the end of last year and through the winter. Not great shape for the second place finisher last time, but decent. Then Joe Biden officially entered.
The narrow-to-medium gap between them widened instantly. Sanders dropped into the high teens. Then Warren began regularly surpassing him. Whether she actually pulled ahead depends on who you ask. The Economist thinks she’s still a few points ahead. Real Clear Politics shows them in a knot for most of the past 60 days.
Maybe most importantly, Bernie’s early state numbers were frequently worse than his national numbers. That’s bad two ways. Especially in Iowa and New Hampshire, voters start paying attention earlier than the rest of us. If they were preferring Warren, or looking toward lower-profile candidates, it could portend a national problem once those voters are more awake.
And early results always impact how voters in later states feel. It pushes media coverage and donations. Voters like backing a winner. If he was already stuck in the mid-upper teens nationally, and then struck out in the first few states, not a good combination.
His floor is around ten percent. That’s where his worst poll results clustered. In better surveys, if you pull out the voters who are still considering other candidates, again, that’s about where he is. That’s better than it sounds. Only Biden has a higher floor at the moment. But it only works in conjunction with a lot of other voters being willing to strongly consider him.
CBS/YouGov was kind enough to help us measure this. They asked voters to list all the candidates they were considering, as well as their first and second choices.
On average, half of the early state Democratic primary electorate is at least kicking the tires on Bernie. That’s more than the percentage that voted for him last time. If you assume all of his previous voters are still willing to think about it, there are at least some Hillary Clinton supporters who are open.
Iowa: 52% (41% 1st/2nd)
New Hampshire: 50% (42%)
Nevada: 55% (48%)
South Carolina: 44% (35%)
His numbers as a top two choice are even more encouraging. Biden is still controlling the field in South Carolina. He’s at 43% as a first choice, with 64% of primary voters considering him.
Sanders is running second. A full 35% have him in their top two. Remember, he only earned 26% in a two-way race with Hillary in 2016. Biden appears as strong with African American voters now as she was then. Plus Kamala Harris, Cory Booker, and a pile of other choices exist.
Bernie has clearly made strides with the black community in this state. He already had a foothold among Millennial voters of color, but this indicates widening acceptance.
As long as Biden is viable, Sanders is no threat to start winning southern primaries. But he got clobbered there in 2016. At the moment, he’s running ahead of Warren, while Harris and Booker have failed to effectively launch. Keeping the deficit in the Old Confederacy manageable is crucial to his chances.
You can argue familiarity is helping him with southern black voters. The Clintons had a history going back decades. Biden has extended ties to important community leaders on top of the obvious Obama connection. Sanders was a newbie four years ago.
That doesn’t explain why he stabilized and now might be surging in the other early voting states. Electability does. CBS/YouGov also asked voters whether a candidate was likely to beat President Trump, had a shot to, or whether it would be a stretch.
As per usual, Biden was perceived as most electable. Consistently, three quarters of voters think he’d likely beat Trump. The other quarter thinks he’d at least scare him. The percentage thinking Biden would likely lose is a rounding error.
Feelings are more mixed about Warren. About half of voters think she’d be a favorite. A little less than a tenth think she’d really struggle. The remaining 40+ percent figure Warren would put up a good fight.
Sanders falls between Biden and Warren. Voters assume his Democratic Socialist agenda is harder to digest than Biden’s comparative moderation. But whether they’re more worried about a female nominee, or are comforted by Sanders having gone through the process before, or #Pocahontas, confidence in him is higher than her. In Iowa and New Hampshire, two thirds of Democrats think Trump would get Berned.
Here’s his advantage over Warren in each of the first four states:
New Hampshire: +14
South Carolina: +6
Whether voters are closely tracking the polls, or just have good Spidey sense, these impressions closely track actual national poll results when Trump is matched against individual Democratic contenders. Biden does best. Sanders usually gives away a couple/few points, but regularly defeats Trump. Warren’s results are much closer to 50/50.
We can say all we want about how head-to-head polling more than a year out from the election isn’t predictive. That doesn’t mean it’s not influential or revealing.
It’s too early for Bernie to start working on his convention nomination speech. Biden and Warren are still more likely nominees. If no candidate wins a majority of delegates ahead of the convention, which is statistically very, very, very possible, he’s at a disadvantage compared to his opponents.
Unless or until Harris, Pete Buttigieg, or one of the others has a breakthrough, this is a three person race. And Bernie Sanders is very much one of the three. Precedent, age, and a hundred other things be damned, he’s a strong candidate.