With weeks to go, Trump needs a big change

[Bloomberg]( Follow Us //link.mail.bloombergbusiness.com/click/21532088.16245/aHR0cHM6Ly90d2l0dGVyLmNvbS9ib3Bpbmlvbg/57d8157f3f92a424d68d6450B21d81761 [Get the newsletter](hash=b9b2681361bede0e1069ca238efb1ec2) Get Jonathan Bernstein’s newsletter every morning in your inbox. [Click here to subscribe](hash=b9b2681361bede0e1069ca238efb1ec2). With fewer than seven weeks to go before Election Day, a few points: - People are voting. Only about 60,000 have so far returned ballots by mail, most of those in North Carolina, but the totals are going to increase dramatically once we hit October. Political scientist Michael McDonald has [all the details]( at his early voting site. I should add that there’s nothing at all suspicious or fraudulent about absentee and other early voting, which has been used in many states for several cycles now without any sign of trouble. It’s true that with more people than usual planning to vote by mail this year, there may be some logistical glitches and vote counts could be slow in some states. But that’s okay, too.  - We’re still seeing plenty of state and national polls, and there’s very little evidence of significant change. Former Vice President Joe Biden is up by about [7 percentage points nationally]( and by a smaller margin in the [most likely tipping point states](, suggesting that he needs to win by about 3 points nationally in order to win the Electoral College and, therefore, the election.  - That’s probably a large enough lead to withstand normal polling errors even if the polls happen to be overestimating Biden’s support — which isn’t necessarily the case. By the way, I highly recommend the [new Upshot polling averages](, which show what would happen with a polling error similar to those in 2016 or 2012, but with two caveats. One is that the chart assumes any error would be in the same direction, which it may not be; I’d rather they demonstrated the size of the error, which could favor either candidate. The other thing is that the “polling error” here includes the final three weeks of polls in each of those previous elections, but there’s good reason to believe that the race shifted in Donald Trump’s direction in the final days in 2016, so part of what’s being called an error here is really polls taken before voters changed their minds.  - Just to be clear: In 2016, Trump needed both that late change and some polling error in order to win — narrowly! — despite trailing in the polls. He’s currently doing somewhat worse than he was going into those final weeks in 2016, so he’ll need a similar combination of late change and errors in his favor to be able to repeat what happened. The problem? Biden’s lead isn’t just larger than Hillary Clinton’s was; it’s also been extremely stable. That doesn’t mean it can’t suddenly change, but it’s getting harder to come up with reasons why that would happen.  - The most dramatic poll released on Wednesday was from Maine. It showed Biden with a huge lead statewide and a solid lead in the closely contested Second Congressional District, which has its own electoral vote. Pundits were quick to label this survey an outlier, which is surely correct, and we can expect some goofy poll results just from dumb luck. But that doesn’t mean we should pretend the poll never happened; the right way to treat unusual and unlikely results is to include them in the polling averages, not to ignore them.  - Meanwhile, that Maine poll also was very good news for Democrats looking at the Senate. The Maine seat is probably the fourth-most-likely pickup for them in this cycle. Assuming they lose a seat in Alabama, they’ll need four pickups to reach 50 seats, and thus a majority if Biden wins (with Senator Kamala Harris becoming vice president and breaking the tie). Democrats have a few other pickup opportunities, so reaching 52 or more isn’t out of the question. But Maine and perhaps North Carolina are still close enough that they certainly could fall one or two short. One interesting shift: A few months ago, it was hard to picture Maine’s somewhat moderate Senator Susan Collins surviving a bad Republican year, but now it may be more likely. 1. Andrea Pena-Vasquez and Maryann H. Kwakwa at the Monkey Cage on [Barack Obama, Kamala Harris and ethnicity](. 2. Brian Stelter and Oliver Darcy on how to [get it right on election night (and before and after)](. 3. Derek Thompson on [Trump’s reluctance over the summer to endorse fiscal stimulus](. I’m not sure about Trump, who seemed to change his mind about this on Wednesday, but I do think the most likely answer for many Republicans is that they just don’t believe that spending money boosts the economy. 4. My Bloomberg Opinion colleague Noah Smith on [the 1970s economy](. 5. And Jonathan Chait on Trump’s misinformation about [stock ownership](. Get Early Returns every morning in your inbox. [Click here to subscribe](hash=b9b2681361bede0e1069ca238efb1ec2). Also subscribe to [Bloomberg All Access]( and get much, much more. You’ll receive our unmatched global news coverage and two in-depth daily newsletters, the Bloomberg Open and the Bloomberg Close.  Before it’s here, it’s on the Bloomberg Terminal. Find out more about how the Terminal delivers information and analysis that financial professionals can’t find anywhere else. [Learn more](.  You received this message because you are subscribed to Bloomberg's Early Returns newsletter. [Unsubscribe]( | [Bloomberg.com]( | [Contact Us]( Bloomberg L.P. 731 Lexington, New York, NY, 10022