Is it time to worry? Why, er, no, says political scientist Brendan Nyhan, who pointed on Twitter today to this piece he wrote in 2014. Hypothetical match-ups this far out from Election Day turn out to have no predictive power. In fact, according to one study, match-ups polled just 300 days out from Election Day — a moment that won’t arrive for another six months — also have virtually no predictive power. You’re free to ignore this buzzworthy Quinnipiac survey along with every other “Trump vs. X” poll between now and January that I mine for cheap content.
Except in one respect. In an election cycle in which Democrats are paying unusual attention to electability, polling that indicates that Biden would fare best against Trump next year may become a self-fulfilling prophesy. If you’re a liberal who prefers Elizabeth Warren on the merits but whose first, second, and third priorities next year are getting Trump out of office, surveys that show Biden leading Trump by roughly twice the margin Warren leads him may be powerful reason to think hard about nominating Uncle Joe instead. I can’t imagine how Team Biden might come up with a more effective campaign ad than numbers from a respected pollster showing their guy blowing Trump out of the water.
Those numbers among independents are something else. Other Dems tested head-to-head with Trump also do well with indies but not like Biden does. (Overall, Bernie Sanders leads Trump by nine head to head while Kamala Harris leads him by eight and Elizabeth Warren by seven.) One interesting number not shown here suggests that there may be less to Biden’s lead than meets the eye, though: When Quinnipiac asked people if they’re paying a lot of attention to the campaign, some attention, or very little/none, it turned out that the more closely they’re watching the race, the tighter the margin between Biden and Trump got. Among those who are paying a lot of attention, it’s Biden by seven; among those who are paying some attention, Biden by 12; and among those paying little to none, Biden by 25(!). That trend repeated in Trump’s head-to-head polling against other Democrats like Sanders and Warren. Each one led Trump by around five points in the group that’s following the race closely and by blowout margins in the group that’s not following it at all. Presumably members of that latter group are less likely to vote next fall.
Before you dismiss Biden’s lead as fake news, note that the Times reports today that Trump has conducted his own polling and that didn’t look so hot either — to the point where POTUS told his deputies to just go ahead and lie about it if asked:
After being briefed on a devastating 17-state poll conducted by his campaign pollster, Tony Fabrizio, Mr. Trump told aides to deny that his internal polling showed him trailing Mr. Biden in many of the states he needs to win, even though he is also trailing in public polls from key states like Texas, Michigan and Pennsylvania. And when top-line details of the polling leaked, including numbers showing the president lagging in a cluster of critical Rust Belt states, Mr. Trump instructed aides to say publicly that other data showed him doing well…
Unlike nearly every recent modern president who sought re-election, Mr. Trump rarely if ever speaks to aides about what he hopes to accomplish with what would be a hard-won second term; his interest is entirely in the present, and mostly on the crisis of the moment.
It’s not just his own internal polling. A Fox News poll taken last month had Biden ahead by 11 nationally, 49/38. Part of the reason he’s dogging Biden so frequently in public is because he’s worried, which ironically may be boosting Biden’s chances among Democratic undecideds. If Biden’s the one whom Trump fears, some Dems might reason, then he’s whom we should nominate. Reading today’s Quinnipiac poll and mindful that Trump’s job approval is rarely above 42 percent for long, Josh Kraushaar goes so far as to call him an underdog for reelection. I’d say he’s no worse than 50/50, but he’s probably the most vulnerable incumbent since Bush 41 in 1992.
Never mind that, though. The real action in the Quinnipiac poll is in the Democratic primary:
Warren’s surge looks increasingly like it’s for real. This is the fourth straight Quinnipiac poll in which she’s gained ground, now tying Bernie among self-described very-liberals. It’s younger adults, men, and, weirdly, nonwhites who are keeping him a few points ahead of her but it’s hard to see why that would last. All of the intellectual energy on the left side of the campaign is coming from Warren. Voters know Sanders well from 2016 but many are still getting acquainted with her via her “I’ve got a plan” weekly policy rollout. If you’re a lefty who hasn’t yet committed to Bernie, what’s going to pull you into his camp instead of Warren’s? I think she’s going to overtake him, and once she does, some “soft” Berniebros might defect to her. A new poll of Massachusetts today shows Sanders at a measly six percent, in fact, one of the poorest showings of either of his presidential runs. Granted, that’s Warren’s home state, but she’s only at 10 percent there. The electorate simply seems cooler to him and his message this time around.
Exit question via Philip Klein: How much can we trust a poll that places the Republican nominee one point ahead of the Democrat among white voters, as Quinnipiac’s Trump/Biden match-up does? Not since 1996 has a Democrat come within single digits of winning the white vote, notes Klein. Trump and Romney each won it by around 20 points. If Biden truly is that competitive among whites, Trump’s goose is cooked. And if he isn’t — and he probably isn’t — then Quinnipiac’s poll isn’t very useful.
Update: We’re not really going to elect a man who’d plagiarize Michael Avenatti, are we?
Michael Avenatti takes a shot at Biden for saying today in Iowa “Let’s Make America America Again.”
Avenatti used the same phrase a number of times during his brief flirt with a 2020 run of his own before facing a string of legal issues. pic.twitter.com/9qcmZWunNX