But then, self-appointed augurs that we are, we start looking more closely, digging through the remains of the contest and holding things up to the light. In fairness, many observers, including many in the media, were doing this before the polls closed, announcing loudly what they expected to find in Michigan’s slimy entrails. (This is an augury reference, not a function of my having gone to Ohio State.) That’s what I mean when I say that it’s hard to differentiate between finding something important and the importance of finding something.
One thing that people are seizing upon in Michigan is the fact that 1 in 8 Democratic primary voters opted to vote for “uncommitted” rather than the incumbent president. The organizers of the effort to send a message to Biden — generally about his approach to the war in Gaza — stated that they hoped to get 10,000 people to vote in that way. They ended up getting more than 100,000.
This was a good example of expectation-setting (there were nearly 20,000 “uncommitted” votes in the 2016 Democratic contest, for example) and overperforming anyway. But it doesn’t really answer the question of whether this matters for November.
(This is a bit of a cop-out, certainly, since nearly anything that occurs now can safely be assumed to not offer much insight into the general election. But then, that also works the other way: It’s easy to overinterpret things after an election as having offered insights into the results! I digress.)
Biden fared worse across the state than the last incumbent Democratic president had: Barack Obama in 2012. You can see that below, and that he did significantly worse than Obama in counties that voted for him in 2020. (The further a dot from the diagonal line, the bigger the gap between Biden and Obama.)
But the percentage of the vote that was “uncommitted” in each contest was similar. A lot more people came out to vote “uncommitted” in 2024 (101,000 as of writing with just shy of 99 percent of the votes reported vs. about 21,000 in 2012) but a lot more people also came out to vote for Biden than Obama.
The counties where the “uncommitted” percentage was significantly higher for Biden than Obama? Largely those same counties that he won in the 2020 general election.
There were two groups expected to be predisposed to cast “uncommitted” votes in the primary: young people and Arab Americans, both groups that have expressed frustration about Biden’s handling of the war in Gaza.
Washington Post analysis of precincts by L2, a political data company, shows that precincts in Wayne County with more Muslim Americans (largely determined using computer models) were substantially more likely to vote uncommitted.
Age can be more precisely evaluated and, in that same county, precincts with larger populations of young people cast higher percentages of their votes for “uncommitted.”
The fear among Democrats is less that those voters will support Trump in November than that they simply stay home. Many may — though of course, general elections tend to solidify wavering voters and lots can change before then. It is useful to consider that Biden has more breathing room in Michigan than other swing states; the 101,000 “uncommitted” votes are still tens of thousands of votes fewer than his victory margin in 2020. There are also undoubtedly a lot of Biden skeptics who didn’t turn out for the primary but cast a ballot for him in 2020.
The focus on Biden is heavily a function of this public, state-specific effort to send him a message. Meanwhile, on the Republican side, there’s an ongoing protest vote, mostly manifested in support for former U.N. ambassador Nikki Haley’s technically existent challenge to Trump’s bid for the nomination.
In Michigan, that meant that Trump underperformed Biden in nearly every county.
There are probably scores of useful caveats here, including that Trump has an actual organized, well-funded opponent. Biden is also the incumbent president while Trump is not.
Compare Biden’s performance this year with the more recent incumbent primary seeking renomination: Trump in 2020. By that standard, Biden did far worse.
But Republicans circled the wagons around Trump in 2020 (institutionally and generally) in a way that hasn’t happened for Biden this year.
What’s the lesson? Take your pick. You have six charts to pick from.
The most important lesson, certainly, is the most boring one, the one for which no chart is needed: Biden and Trump both won easily and will almost certainly be the two major-party candidates on the ballot in November.