I got a flood of responses to yesterday’s post on Joe Biden’s poll numbers. As always, I learn from all of them. I should say that I know these posts provoke people. But it’s not mischievous provocation. It’s with a purpose. To that end, most of the replies were some form of making two points. First, we’re 18 months from the election. Polls can and do change and they don’t mean that much now. (TPM Reader AG sent me the link and text to this CNN write-up of an April 2004 Gallup poll showing John Kerry beating George W. Bush.) Second, Joe Biden sucks! By which they mean, as they explain, Biden has any number of already apparent weaknesses and liabilities – he’s old; he’s got a history running failed presidential campaigns; he’s on the line for various policies of the past which Democrats now repudiate; he’s old school and establishment; he’s out of touch with the aspirations and beliefs of younger and more diverse Democrats, etc etc. The point being that both within the Democratic primaries and in a hypothetical general election these liabilities will inevitably weigh him down and sink him regardless of what polls today may say.
I actually agree with a lot of these criticisms and said as much a few months ago. More than that, I worry they will weigh him down over time.
But all of these responses strike me as beside the point or missing the point. To the degree there’s any confusion I 100% agree that polls 18 months out from a presidential election are not locked in stone. And if anyone is saying I think these polls mean Trump can’t beat Biden I absolutely do not think that. The possibility that he can and may worries me a great deal.
It frankly worries me just as much is that most of the Democratic field is polling as relatively weakly as they are vis a vis Trump. Do they have time to grow as voters become more familiar with them? Absolutely. And I anticipate this happening, though I’m somewhat less confident of it now than I was a few months ago.
For me it is overwhelmingly important that a Democrat win the White House in 2020 and that is far more important to me than which of the candidates the nominee ends up being. I’m actually quite happy seeing the nominee be almost any of the current list of nominees – and to the extent it’s one of the ones I’d be less happy about I’d enthusiastically support that person. My first, second and third priority is their ability to beat Trump. Of all the top tier candidates the most logical nominee to me is Kamala Harris. But if by next winter Biden looks like he has a significantly better shot at beating President Trump I’d greatly prefer him or whoever else seems to have a significantly better shot.
To put the point more crisply, if all the candidates look like they are likely to beat Trump but one a bit more likely than others that isn’t that important to me. But if one looks like they have a solid shot and the rest seem like 50/50 propositions, I’m going to be for the one with a solid shot even if it’s not my first choice.
Here’s why I keep pressing this point about the current general election horse race numbers. The response I see to these numbers usually ends up being some form of this argument: Yes, he’s doing best now but he has various liabilities that will eventually pull his numbers down. That may well be true. In fact, I suspect it may be true myself for the reasons listed above. But we always need to submit our assumptions and theories to outside evidence or metrics that are not controlled by assumptions and preferences. Polls are far from perfect but general election polls are a pretty good bit of outside data with which we can challenge our assumptions – assumptions which are by definition really just our theories and often our desires.
I think these polls are important because they suggest that the assumptions about the electorate that many of us have may simply be wrong, that for all his liabilities (which I fully recognize) Biden may draw the biggest numbers into the Democratic column.
Data not controlled by our assumptions and desires is always critical to watch.
Now, let me clear up a final point. Some readers are clearly reading this as: Biden’s the strongest candidate. So if you’re supporting someone who’s not Biden, see the writing on the wall, get your head out of your behind and get on board. That is absolutely not what I’m saying. It is really not.
We have a six-month plus primary process in front of us. I don’t subscribe to the idea that a vicious primary process helps the eventual nominee. It can definitely weaken the nominee. But given the electoral questions about all the top candidates (Biden and all the rest) we need to see how they cope in the wild, against attacks, showing their ability to think and pivot on their toes. No one has to commit until early next year. And speaking for myself I very much want to see what they come up with, how they prove their ability to mount the kind of campaign that can defeat President Trump. If I were king and had to chose the nominee myself now I’d be quite nervous about choosing Biden precisely because I worry about how he’ll survive in the electoral wilds. I’d be even more worried about choosing one of the candidates who currently are tied with Trump.
It always worries me when people dismiss outside, empirical data because of their assumptions. Our assumptions and premises can never be durable if they can’t be challenged by data or information that is not in turn controlled or shaped by those assumptions and premises. Indeed, when our assumptions don’t line up with outside data we should be skeptical and look for holes or blindspots in our own thinking. We all live in bubbles, feedback loops of our own assumptions and desires. As I wrote yesterday, “you simply cannot make this claim about Biden being a weak general election candidate without grappling with the fact that basically every poll for months shows that he is significantly stronger than every other Democrat up against Trump.” When I hear back the same set of assumptions that Biden is simply a bad candidate and therefore the polls are either wrong or will perforce change that seems like a refusal to consider the possibility that the assumptions are wrong – assumptions which in the abstract I agree with!
So that’s my point. I’ll follow up today with some emails which I think are good counters to my take on this, ones that made me reconsider my own take.