National Democratic primary polling is beginning to turn against Bernie Sanders, and through that lens, it really does look like he is now a measurable third to Elizabeth Warren’s second place charge against Joe Biden. However, national polls are the wrong way to think about this race, given that the primary is determined on a state by state basis. This misfocus is emblematic of the major blind spot the Democratic Party demonstrated this past decade by losing a thousand seats over the course of the Obama presidency: national politics may give us our favorite TV characters, but all politics is local. The Republicans get this—which is why they have most of the power in this country—and if you want to properly understand how any presidential primary will unfold, you must look at state by state polls.
So to test this narrative of decline that’s beginning to take root against Bernie (which is compounded by genuinely bad polling for Sanders in the first state to vote in Iowa), I put together a very basic spreadsheet using the most recent state-level polls from FiveThirtyEight, and the RealClearPolitics average for Iowa, New Hampshire, Nevada, South Carolina, California, Massachusetts and Texas, then averaged it all out.
This is an imperfect way to obtain a specific figure, as not all polls are created equal nor taken at the same time, but it is a quick and dirty snapshot of where current polling stands. The race is young enough that no polling should be taken as gospel yet, and not enough polls have been taken to get a full 50 state sample. These figures only include 32 states, with 25 of the 32 slated to vote before April 4th, 2020—about midway through the primary schedule.
With all those qualifiers, here is where our best polling currently stands via this crude state-by-state average:
I do not believe that this proves that Bernie and Warren are in a near-tie as this imperfect average demonstrates, and I do think that she has quite possibly edged ahead of him given that most newer polling is generally more positive for Warren. However, this notion that Elizabeth Warren has clearly defeated Bernie Sanders to win the battle for the soul of the left, and is ready to turn this primary into a primetime showdown with Biden, is only supported by a handful of polls that ultimately do not matter. The ones that do matter don’t prove anything demonstrative other than a semi-comfortable lead for Joe Biden.
Given the frontrunner’s obvious shortcomings and the fact that Bernie Sanders is about as popular among the Democratic over-50 crowd as Donald Trump is, Elizabeth Warren does seem to be in the best position to win the primary going forward.
However, one of the threads missing from the current narrative is that while Warren has charged up the leaderboard in a bunch of key states, whether that has translated to the vitally important south remains to be seen (five of the first fifteen primaries take place in the south, and that excludes Texas and Oklahoma where she and Bernie both struggle as well). A recent poll from High Point (which 538 rates as B+) has Sanders with a five-point lead over Warren for second in North Carolina. Two July polls (albeit from D- rated pollsters) place Warren in the single digits in Mississippi and Alabama (her national RCP average at the same time was a little less than 15%), and in ten polls used in the RCP South Carolina rolling average this year, she has surpassed Sanders in four (with the caveat that two of the four have come in the last two months).
Three polls out of these 32 place Elizabeth Warren in the single digits, while just one has Sanders in the danger zone: a 538 C rated pollster from Missouri which places him behind Biden, Warren, Kamala and Buttigieg at 4%. I don’t bring all this up to present a case against Warren’s candidacy in these numbers, but to muck up the narrative taking root around her supposed certainty as Biden’s only challenger. Nothing should be clear right now, other than that Biden’s numbers are consistently bigger than both.
All three main contenders have glaring weaknesses, and as far as trying to project current polling into a future narrative, it is far likelier that there is no clear frontrunner than there is one clear frontrunner—let alone an obvious 2-3 order to the Warren-Bernie dynamic. Anyone pushing the narrative that Warren is clearly ahead right now is either placing too much emphasis on the polling out of Iowa or the national polls, or is trying to will that narrative into existence. We have a long way to go until the first votes are cast, and voters should think like voters, not like pundits (because voters know more about politics than most pundits). Vote for who you think best represents your interests, not who the pundits say has the best chance to beat Trump, or who you think is trending up or down—especially since the only data that matters is telling us that the dynamic which has defined most of this race continues uninterrupted.
Biden > Bernie & Warren
Bernie + Warren > Biden
Until this reality changes, we’re stuck with Biden.
Jacob Weindling is a writer for Paste politics. Follow him on Twitter at @Jakeweindling.