The good news is that there are two polls out of Pennsylvania today, one of which is reasonably encouraging for Trump. Change Research, a Democratic firm, has the race within four points, 50/46. That’s the same margin they found when they polled the state two weeks ago. Despite all the hype about Trump’s debate performance and the fallout from his COVID diagnosis, he hasn’t lost ground there.
The second poll out today, from Monmouth, isn’t encouraging. Monmouth also had the race within four points when it polled Pennsylvania at the end of August. Today: Biden 54, Trump 42 among registered voters.
I agree with Ed that this morning’s CNN numbers showing Biden up 16 nationally was a “w-w-w-what?” outlier. I think today’s Monmouth poll is an outlier. I also believe that the NBC poll I wrote about yesterday morning showing Biden up nationally by 14 was an outlier.
But at some point enough outliers pile up that they begin to look like a trend.
We don’t need to obsess over whether Biden’s actually up 12, or nine, or six, though. (A YouGov poll of the state published Sunday had him up seven.) It should suffice at the moment to say (1) he’s leading by enough in the states he needs to win that he would win, probably easily, if the vote were held today. And (2) to the extent that there’s a trend one way or another right now, the trend is towards a Biden landslide, not a narrow Trump upset.
Interestingly, Monmouth’s pollster believes that the debate did more damage to Trump in Pennsylvania than his COVID infection did. The numbers scarcely moved after he announced that he’d tested positive.
“If any recent event moved the needle it was more likely last week’s debate than the president’s Covid diagnosis. What seems to be more important than either event, though, is voters’ focus on which candidate they trust more on the issues that keep them up at night,” said Patrick Murray, director of the independent Monmouth University Polling Institute.
The poll found little differentiation in vote preferences before and after the president revealed he has Covid-19. Half of the poll’s interviews were conducted on each side of that breaking news. After adjusting for the sample’s party registration balance in each half, Biden held a 54% to 42% lead among registered voters in interviews conducted Wednesday and Thursday and a similar 55% to 42% lead in the Friday through Sunday period…
Of particular note, [Biden] now leads Trump by 54% to 45% among middle-aged voters (50 to 64 years old). This is a group that Trump led by 54% to 45% in Monmouth’s August Pennsylvania poll and has generally led in most states throughout the campaign. Biden maintains leads he has held with younger voters (54% to 40% among those aged 18 to 49) and senior voters (55% to 42% among those aged 65 or older).
Four years ago Trump won voters aged 45-64 by seven points and seniors by 10 points. He now trails in the latter group by double digits and nearly double digits in the former. There’s not much more to say about the race than that, really.
One thing that jumps out both in Monmouth’s numbers and in other recent polls of Pennsylvania, though, is how consistently Biden polls at 49 percent or better. Whether his margin over Trump is smallish (four points) or yuge (12 points), Sleepy Joe is always either on the cusp of a majority or enjoys a majority outright. In the last 16 polls of the state tracked by RCP, he’s been at 49 percent or higher in 14 of them. In nine polls he’s hit 50 or better. These two tables from Monmouth capture Trump’s dilemma:
Forty-eight percent say they’re locked in for Biden and another nine percent are considering voting for him. Only 39 percent say no way. For Trump the numbers are the opposite. Fifty-one percent say no way; just 44 percent are either locked in or thinking about backing him. The guy who’s already on the brink of a majority has more room to grow than the guy who’s trailing.
Here’s one way to think about the big picture right now:
If the polls were exactly as wrong as they were over the final three weeks in 2016, Biden would still win with over 300 electoral voteshttps://t.co/FuxTub58ar
— Nate Cohn (@Nate_Cohn) October 6, 2020
Here’s another via Phil Klein:
Fundamental point I keep coming back to: All Biden needs to do to win is do 1 point better than Clinton in Michigan, Wisconsin, and Pennsylvania.
— Philip Klein (@philipaklein) October 6, 2020
Biden’s momentarily up five points or better in all three of those states in the RCP average. In fact, thanks partly to Monmouth, Pennsylvania is his strongest state today: He leads by 6.5 points there versus 5.8 in Michigan and 5.6 in Wisconsin. PA is the state he’s working to flip because, until recently, it was considered a *better* opportunity for him than the other two. (And it may still be. The Change Research poll that has Biden up four in Pennsylvania has him up seven in Wisconsin and eight in Michigan.)
Here’s a third way to think about the race. In a good year for a Republican, Texas is a state where the winner wins comfortably and Pennsylvania is a state where things get tight. Right now it seems likely that the opposite will be true this year. Trump leads by just 3.2 points in Texas in RCP today, almost exactly half the margin Biden enjoys in PA. And Biden’s campaign thinks he has enough of a chance to pull the upset in Texas that they’re about to drop $6 million there in ads. There’s a small chance right now that Texas turns blue and a large chance that Biden loses the state only narrowly. That’s not the makings of successful reelection bid for Trump.
I’ll leave you with one more detail from Monmouth. Their poll imagined two different scenarios for the election, one with low turnout and one with high turnout. Biden leads by eight in the former and by 11 in the latter. (In both cases he notches 53 percent or better.) If you’re wondering which of those scenarios is more likely, low turnout or high, read this Reuters story about early voting. Exit quotation: “With four weeks to go before Election Day, more than 4 million Americans already have voted, more than 50 times the 75,000 at this time in 2016, according to the United States Elections Project, which compiles early voting data.”