Is Mills or Platner the "working class candidate"?
More statistical shenanigans from The Argument.
Lakshya Jain, in The Argument:
Despite his reputation as a working-class whisperer, Platner is actually doing far better with upscale Democratic whites than with non-college Democrats. Each primary poll with regional breakdowns has shown a common theme: Mills does far better among the blue-collar Democrats in the rural north of the state, while Platner cleans up with wealthy, coastal liberals who skew college-educated.
Both parties are fighting for control of the working class...but despite national media narratives, [Platner is] a candidate whose appeal is strongest with voters they were already going to win, and whose vulnerabilities are greatest with the voters they actually need.
I can’t say that I come into Jain’s polling analysis with much confidence these days, but even before I looked at the data this argument raised more red flags for me than usual. Why is Jain just looking at who polls better in Maine’s two Congressional districts to judge who is doing better with the working class when polls like this always have much more granular data? And if we want to know which party wins control of the working class, why are we looking at who does better in the primary rather than who does better in head-to-heads with Collins?
Let’s talk about geography first. As Jain points out, primary polls show Platner beating Mills in the south and Mills beating Platner in the north. But here’s what their numbers look like in head-to-heads against Collins:
In both cases Platner doesn’t just outperform Mills against Collins — he outperforms her by near double-digits. Whatever advantage Mills has against Platner completely evaporates here.
But what if we look at the geography a little bit closer? Here’s what UNH data shows us:
It’s true that Mills outperforms Platner in Northern Main, which has a relatively low average household income. But Platner outperforms Mills in Central Main, which has an identical average household income — but is more than twice as large. So by this measure, Platner is outperforming Mills among every income group, including the working class.
Jain writes that Platner voters “skew college-educated” — and again, I am baffled why he would use Congressional districts as a proxy for this when UNH measured it directly. Here’s what their survey says:
The real story here, of course, is that Platner is wiping the floor with Mills across the board. If you’re eager to find a trend, however, it’s the opposite of what Jain suggests: Platner actually does better with people who have very little education than he does with postgrads, who are Mills’ strongest group. Here’s how they perform against Collins:
Same trend: Platner outperforms Mills among educational group. Finally, no need for me to create a special visualization for income, because the Pan Atlantic Research poll Jain links to did it for me:
As far as I can tell, all three polls Jain sites shows Platner either tying or dramatically outperforming Mills among the working class by almost every conceivable measure. The only one that doesn’t show this is the worst one possible: a coarse comparison of districts in the primary rather than the general election. Even if he wins, and it seems like he is likely to, Platner is going to have a hard fight ahead of him if he wants to defeat Collins. But if Democrats want to win working class voters, it would be utter madness to nominate Mills instead.
Refer enough friends to this site and you can read paywalled content for free!
And if you liked this post, why not share it?