Biden 356
image description
Ties 18
Trump 164
image description
Click for Senate
Dem 54
image description
   
GOP 46
image description
  • Strongly Dem (217)
  • Likely Dem (62)
  • Barely Dem (77)
  • Exactly tied (18)
  • Barely GOP (38)
  • Likely GOP (53)
  • Strongly GOP (73)
270 Electoral votes needed to win This date in 2016 2012 2008
New polls: CT FL MI NC PA WY
Dem pickups vs. 2016: AZ FL GA IA MI NC PA WI
GOP pickups vs. 2016: (None)
Political Wire logo Trump Praises Supporters Who Surrounded Biden Bus
New Battleground State Polls
Fauci Offers Blunt Assessment of Trump
Republicans Seek to Void 100K Votes In Texas
The Poll Would Have to Be Way Off for Trump to Win
Democrats Grow More Anxious About Pennsylvania

TODAY'S HEADLINES (click to jump there; use your browser's "Back" button to return here)
      •  Saturday Q&A
      •  Today's Presidential Polls
      •  Today's Senate Polls

Happy Halloween! Let's hope that today is the scariest day of the year as opposed to, oh...say...November 3.

Saturday Q&A

For obvious reasons, we got close to a thousand questions this week. We picked out a considerably larger number than usual to answer (and one to not answer, at least not yet); hope you enjoy.

Q: We know what the polls say, and we also know that the polls don't account for judicial meddling, foreign interference, vote suppression, or any of those other pesky known unknowns. But as a trusting and loyal reader I'd love to hear your guesses. According to the crystal ball that beats in your chest, how is this election most likely to play out? A.W., Brooklyn, NY

A: This is the one question we're not going to answer...yet. On Tuesday morning, however, we will give our answers. And if readers would care to write in with their guesses (in 100 words or so), we'll run some of those, too.



Q: Can you provide an estimate of the EVs Joe Biden and Donald Trump will officially have by the end of day on election night? J.E., Pennington, NJ

A: Because of absentee votes, military ballots, etc., there probably won't be any official results by the end of the evening. That said, it is likely that the Associated Press and other outlets will call a fair number of states, and that many states will also release unofficial results. A reasonable estimate is that Joe Biden will have 200+ EVs effectively won by the end of the night, and Donald Trump will have around 100.



Q: In addition to fast-counting Florida, what are some of the other fast-counting States? Hopefully some are tipping-point states (this is coming from a moderate Democrat who has chewed his fingernails to the bone). A.K.P., Huntsville, AL

A: The New York Times is running a tracker with the latest statements from election officials in all 50 states and Washington, DC. And if you read it over, you will see that there are two answers that pop up a lot: (1) We hope to have results by noon on Wednesday, and (2) We'll have results whenever we have results. In other words, not that many states aspire to report by the end of the day on Tuesday. The ones that do hope to pull it off: Alabama, Florida, possibly Maine, Montana, North Carolina, possibly South Carolina, Vermont, and possibly Wisconsin. That does not mean that the media cannot project winners, however, based on exit polls and other data. Further, if Florida and North Carolina actually do come through and report on election night, that may tell us everything we need to know. If Biden wins both, it's over; otherwise, we're going to have to wait for the "Midwest."



Q: So, Pennsylvania is a make-or-break state for both campaigns, as the winner becomes favored to win it all, Joe Biden with a 97% probability and Donald Trump at 69%. I get that. But what about Florida? What does a win there do for either candidate? And what if these two states each go to a different candidate? C.S., Wantagh, NY

A: Note that we are getting those numbers from FiveThirtyEight's new "scenario generator." According to them, if Biden wins Florida, his odds of winning the whole election are greater than 99%. If Trump wins Florida, his odds only rise from 10% to 29%. If Florida goes to Biden and Pennsylvania goes to Trump, then FiveThirtyEight says Biden is still 95% to win the election. And if Florida goes to Trump and Pennsylvania goes to Biden, then Biden is again 95% to win the election. The message would seem to be that if Biden wins either state, it's pretty much over. And you didn't ask, but if Biden wins North Carolina, his odds also jump to greater than 99%. So it would actually seem that there are three states that, if Trump loses them, are fatal to his reelection hopes.



Q: Are there any counties that match the presidential election outcome every time? Or rather, which counties have consistently gotten the election winner right the longest? Could we use these counties as a bellwether to tell us what to expect this time around? M.M., Corvallis, OR

A: We are going to begin this answer with an old joke/story about a stockbroker. One day, she calls half her clients and tells them to buy Stock X and calls the other half to sell Stock X. Then, a few months later, she calls half of the group that ended up getting the good advice and tells them to buy Stock Y and calls the other half that got good advice and tells them to sell Stock Y. Repeat the maneuver a few times, and it means that for 99.9% of her clients, the last advice she gave them was bad. But for 0.1%, she's a genius who is right every time.

We note this, obviously, because one wants to be a little cautious about bellwether cities, or states, or trends, or sports outcomes, or whatever else "seems" predictive. After all, there are a lot of coin flip events in the world, and if you look hard enough, you can find just about any pattern you want. There are 3,141 counties in the U.S. and so, even if each of them chose their candidate completely at random, we would expect between 1 and 2 counties to get 10 elections correct in a row.

And now, we direct your attention to this list of counties that have gotten the presidential election right at a way-above-average clip. The longest streak in the U.S. right now belongs to Valencia County, NM, which has been correct in every election since 1952. That's 17 in a row, so obviously considerably better than random chance. That said, there were quite a few elections in there where pretty much every county was right, so if we consider only closely contested elections, Valencia County isn't too far away from being the country's "genius stockbroker" county. Meanwhile, Vigo County, IN, has voted for the winning candidate in every election since 1956, and another four counties—Westmoreland County, VA; Ottawa County, OH; Juneau County, WI; and Sawyer County, WI—have voted for the winning candidate in every election since 1964.



Q: Is there factual (non-anecdotal) evidence that visits by a presidential candidate really help in this day and age? Hillary Clinton has been beaten over her head for not visiting Wisconsin, but people neglect the fact that she visited Pennsylvania many, many times and the Democrats held their convention in Philadelphia and she still lost the state. I feel that people bring up Wisconsin only because it was so close. They selectively pick the one piece of data simply because it is convenient and not necessarily correlative. T.J.R., Metuchen, NJ

Q: You frequently mention that if Hillary Clinton had visited the Rust Belt more often she may have won in 2016. Given that most rallies have a few thousand attendees or so, what good are they really? At best, the candidate would draw in some undecideds who are leaning their way already, while at worst you just get full attendance of die-hard supporters who just want to see you. Is there any data that shows rallies can really move the needle or are they just an antiquated tradition? A.G., Santa Clarita, CA

A: There is some evidence that it matters. However, there is a difference between a general political event (like a speech at a college campus) and a rally. The former has a much greater chance of reaching undecided voters than the latter. Also, it may be that the primary benefit of paying a visit to Wisconsin or Michigan or Arizona may be that the candidate avoids the negative response of not showing up. That is to say, making an appearance in those places may not excite voters there all that much, but ignoring them may piss those same voters off. This dynamic was particularly on display in 2000, when Tennesseeans were irritated that Al Gore seemed to be taking them for granted (and, of course, gave their decisive EVs to George W. Bush).

Also, a visit to a state produces a non unsubstantial amount of free media coverage. It may be difficult to quantify the value of that precisely, but it's surely higher than "zero."



Q: You have noted multiple times that Joe Biden has been crushing Donald Trump on TV in terms of advertising. However, you have also mentioned that Trump has a stronger presence on social media and YouTube. I am concerned about this, since I believe that a large majority of people have more time and opportunities to spend online as opposed to watching TV, and that this may be a strategic mistake on the part of the Biden campaign. Is there anything you could say to assuage this concern? V.G.D., Perrysburg, OH

A: We think your assumption about social media is incorrect. It is, first of all, an extremely fragmented landscape. Beyond a few central hubs like Facebook and maybe Twitter, the folks using Twitch are different from the ones reading Breitbart are different from the ones who like 4chan are different from the ones who spend hours watching YouTube clips. Second, there is overwhelming evidence that the users of various social media sites (particularly Twitter) are not representative of the general electorate or of either party's base. And so, Trump's social media-oriented campaign is undoubtedly missing a lot of voters, and is almost certainly attuned to the desires of a certain loud, angry segment of the electorate (that is already with him) and not to the voters he needs to persuade.

Oh, and the Biden campaign has spent plenty on social media, just in case. Just not quite as much as the Trump campaign.



Q: Does the opposition nominee (Joe Biden) receive the Presidential Daily Briefing (PDB)? I recall that he gets secret service protection after being the nominee and in many cases after becoming the presumptive nominee. But does he also begin receiving the highest-level intelligence at some point? I ask because while some people might not be reading that document, it surely contains valuable information about any potential threats to the election and if Biden had access to it he would be in a better position to counteract it. So does he? M.S., Milwaukee, WI

A: Absolutely. That decision was made by Harry S. Truman, who was kept entirely out of the loop by Franklin D. Roosevelt, and so was totally unprepared to assume the presidency when FDR died shortly after being inaugurated for a fourth time. Most famously, Truman did not learn about the Manhattan Project until he'd been president for almost two weeks. Harry S. didn't want any of his successors (or the country) to go through that again, so he issued an executive order calling for major-party nominees to be given the PDB. It's also worth noting that former presidents are given the PDB, which means that even if Biden didn't have the info, Barack Obama does. And we suspect those two gentlemen have each other's phone numbers.

And you are right that if there is stuff in there about election threats, Biden would be in a position to do something about it.



Q: Do you have any idea why "Can I Change My Vote" was trending on Google? I have some difficulty believing, as some sites suggest, that it is because of anything to do with Hunter Biden. P.R., Arvada, CO

A: There was a meme on Facebook (which Donald Trump retweeted, of course) claiming that searches for "Can I Change My Vote" were way up after the final presidential debate and/or after the alleged Hunter Biden e-mails were reported by The New York Post. However, it is easy enough to look up Google trends, and it turns out that this is not true. In fact, the increase in searches for "Can I Change My Vote" happened not after the debate or the Post stories, but instead after the meme was circulated widely on Facebook. In other words, it strongly suggests that people weren't interested in changing their vote but instead were, like you, trying to figure out what the meme was all about.



Q: I received a postcard in the mail that showed my name, party affiliation, and years that I have voted. Apparently this information is public record? In the guise of offering a service to me, this postcard informed me that if I vote this year, it will mail this information to all of my neighbors (in the name of "voter information transparency").

Whoa! Little did they know I've already sent in my ballot—and I do not mind if my neighbors see my party affiliation. But it felt invasive and there are plenty of people who would be quite intimidated by that, depending on their neighbor situations. Was this likely the result of a voter suppression campaign? I am a registered Democrat in Orange County, CA. I imagine a good number of my neighbors are registered Republicans. Is it legal for someone to mail your voting record to your neighbors?
S.H., Orange County, CA

A: In many states, this information is indeed publicly available. And anything that's publicly available can be mailed to anyone the sender sees fit.

That said, this is quite obviously a sleazy trick designed to shame you into not voting Democratic, for fear that you will be outed to your Republican neighbors. That implied threat, however, is illegal, as the California elections code (section 2194) says that voter information "Shall not be used for...the harassment of any voter or voter’s household". Several Republican politicians have gotten into hot water in recent years for similar tricks, most obviously Sen. Ted Cruz (R-TX).



Q: It looks like turnout everywhere is very high. Is high turnout always good for Democrats? D.K., Iowa City, IA

A: "Always" is strong but, as a general rule, yes. There are more Democrats than there are Republicans, and higher turnout reduces the chance of a mathematically unlikely result where the minority wins. Imagine you have a six-sided die, and your side needs 50% of the rolls to be 5 or 6 to "win." Well, you're more likely to pull that off if you only have to roll 20 times than if you have to roll 40.

It helps the Republicans that their voters are, in general, more reliable voters than Democrats. That means their floor is pretty high, even if their ceiling is lower. This is the whole purpose of voter ID laws and other forms of suppression—to limit the extent by which Democrats exceed their floor.

Note that this applies specifically to national elections. Obviously, in places where Democrats are in the minority (like much of the South), then high turnout favors the Republicans and not the Democrats.



Q: Considering how much attention has been paid to the question about when mailed ballots must arrive/be postmarked for them to be counted, I wonder if absentee ballot arrivals at county recorders' offices are tapering off as we approach Election Day? If they have already peaked and fewer are coming in each day, that might indicate that those voters who requested ballots have either gotten their ballot in already or are worried about their vote not being counted and now plan to vote in person (early or on Election Day). I'd feel a whole lot less concerned about large numbers of votes getting thrown out if we were seeing a reduction in ballots coming in these last few days compared to whatever the numbers were last week. A.F., Iowa City, IA

A: It's a little too early to know that and, indeed, we may never know it. There are lots of ballots still arriving from folks who sent them with a "safe" amount of time left. Further, given the way weekend mail processing works, Monday could well see more mail-in ballots received than any day so far (since it will essentially be three days' worth). If we do learn the answer to this question, it will be late next week, when we see how the post-Nov. 3 arrivals compare to the pre-Nov. 3 arrivals.



Q: I live in Colorado, a safe state, so I'm trying to figure out how I can be most impactful. I'm thinking texting and/or phone calls to Maine voters, because without the Senate a Biden administration would be completely hamstrung, and given Maine's small population, you can get more "bang for the buck." One voter in Maine matters more than in a more populous state. But I'm curious about your thoughts—what's the most impactful thing that individuals in safe states can do, on the last few days of the election? J.H., Denver, CO

A: For those who are in swing states, and who have a car, undoubtedly the most valuable thing you can do is volunteer to help transport voters to their polling places. This, of course, assumes that you follow appropriate COVID-19 protocols, and that you're not in a high-risk profession (like healthcare worker). Volunteering to pick up people's absentee ballots, in states where ballot harvesting is legal, is also a valuable contribution.

Beyond that, the campaigns/parties are going to have far and away the best information about how volunteers can do the most good. They will have micro-targeted data about who has and has not voted, and who is and is not potentially persuadable/capable of being motivated. So, we would suggest contacting the local Democratic Party or Republican Party office or the local Biden or Trump office, and offering to do whatever they think is most useful.



Q: Would an election official be within their right to order someone who is COVID positive, or clearly showing symptoms of COVID, to leave a polling place before casting their vote? P.F., Fairbanks, AK

A: Nope. It is not the province of election workers to disqualify voters, much less to create, in effect, a new condition for being allowed to vote. If that was possible, can you imagine how many Black folks in Texas or Latinos in Arizona might suddenly be "diagnosed" with COVID-19?

What states are doing is instructing poll workers how to deal with people who won't wear masks and/or who appear to be manifesting symptoms of COVID-19. You can read California's version here, if you would like to see the example. The basic commands to poll workers are: (1) keep your distance, and (2) try to stop the situation from escalating. If a COVID-19 sufferer were to become belligerent, however, or were to do something obviously dangerous like deliberately cough on people, then the poll workers could call and have them arrested.



Q: On one hand, you cite the Democratic anxiety about voter turnout in Miami potentially being problematic for a Biden win; on the other hand, you have five polls today that have Biden ahead of Trump. Can you make me feel better about the dynamic here? I'm enthused about the polls, but terrified about the voter turnout issue and how that could reverberate across multiple swing states. C.S., Olympia, WA

A: When (Z) was an undergraduate, a professor once explained how evidence works by noting that when Charles Manson went on trial, there were over 400 pieces of evidence that suggested he was innocent of the charges against him. However, there were also 2,000 pieces of evidence against him, which is why he ultimately went up the river and spent the rest of his life in the crowbar hotel.

When it comes to this election, or any election, there are always indicators pointing in every direction. And when all is said and done, people will look back and point to the consistent national polling for Joe Biden, or late polling in Wisconsin, or sluggish Latino turnout in Miami, or Donald Trump's unwavering base, or FiveThirtyEight's voter models, or any of a dozen other things and will say "See! The signs were there!" The problem is that we don't have the benefit of hindsight right now, and can only judge based on the evidence before us, which is vast, if sometimes contradictory. And the lion's share of the evidence says that Joe Biden and the Democrats are in excellent shape and should have a very good day on Tuesday.



Q: Do we need to talk about John Kerry? In early 2001, a Republican president was sworn in despite losing the popular vote. That president's popularity spiked after the September 11 attacks, but ebbed as his re-election bid approached. Certainly, a great number of Americans were disenchanted or angry with this president, despite his 2001 and 2003 tax cuts. In John Kerry, we had a nominee who didn't inspire much animus, but was hardly every Democrat's top choice. Turnout from the election was up over 5% over 2000—supposedly a good thing, but Kerry lost both the popular vote and electoral college. Could an analogue be brewing between 2004 and now? M.P., Fort Worth, TX

Q: I've said before that this year's election is very similar to 1980. An incumbent president facing a lingering national crisis and is not getting the job done. A challenger who has run for the presidency before, is fairly well liked by many, but still is seen by some as not capable due to his advanced age.

It's been said that just days before the 1980 election, the polls were somewhat tight between Carter and Reagan. It was only on Election Day that we saw the Reagan landslide take shape.

My question is, do you think we might be seeing another landslide coming even though it's considered relatively tight right now?
R.H.D., Webster, NY

A: We put these two questions together because they both illustrate the same thing: It is easy enough to make a compelling comparison between any two elections that are reasonably close in time to each other. That doesn't mean the comparison is illuminating, however.

In the case of 1980, for example, there was an x-factor that pollsters struggled to account for, namely the presence of a somewhat serious third-party candidate in John Anderson, a Republican-turned-independent. The assumption was that he would steal votes from Ronald Reagan (since both were Republicans), but he actually took more votes from Jimmy Carter. That election also featured a bona fide ongoing foreign policy disaster in the form of the Iran hostage crisis, which really hurt the incumbent.

As to 2004, this site was in existence at that time. And so, you can see how we had it on Oct. 30, 2004: Kerry 243, Bush 280, with New Jersey tied. The final tally was Kerry 252, Bush 286. Point being, the Kerry campaign's polling was nowhere near as strong as the Biden campaign's polling is right now.

As with our answer above, in about a week, it will be easy to know which past election was sorta similar to this one. But it will only be sorta similar, because there is no historical analogue to going through a pandemic in the midst of a presidential election, or for all the early/mail voting, or for the likelihood of the greatest turnout in American history.



Q: Did the Senate Majority Leader Mitch McConnell (R-KY) make a tactical mistake by confirming Amy Coney Barrett before the election? Inasmuch as they would have deferred to after the election, that may have kept the pressure up on voters in the evangelical camp to vote for various Senate incumbents and the current President in order to get Coney Barrett confirmed. At this point there is no significant motivation in that specific camp of voters to vote for otherwise ostensibly morally/ethically challenged politicians. R.J., San Francisco, CA

A: McConnell is going to be re-elected, regardless of what happens with the Supreme Court (or when it happens). And he has now achieved his fondest dream, namely tipping the Supreme Court in a conservative direction for a generation. So, he's a winner all around. The person who might have benefited from waiting is Trump. And so, we find persuasive John Gruber's argument that McConnell played Trump like a fiddle, manipulating the President into doing something that harmed his own political fortunes to the benefit of McConnell's political fortunes.



Q: Should I be worried about a prolonged Senate free-for-all caused by seats being vacated for slots in Joe Biden's cabinet, or even a competitive race for Kamala Harris' seat? Or in this case, are the blue states bulletproof? G.S., Queens, NY

A: This depends on a couple of known unknowns, namely: (1) What is the ultimate composition of the Senate? and (2) Which senators depart for the Biden cabinet? If the Senate is 50-50, that's very different from if it's 54-46 for the Democrats. Similarly, adding Sen. Sherrod Brown (D-OH) to the cabinet (not that his name has been mentioned) would be a whole different kettle of fish from adding Sen. Chris Coons (D-DE).

With that said, we would say there is little for Democrats to be concerned about. In California, the Republican bench is thin, thin, thin. The last half-dozen candidates the GOP has put forward for the U.S. Senate have been chosen primarily because of their ability to self-fund. And since the state adopted a jungle primary, only one Republican has even advanced to the final round of voting (Elizabeth Emken in 2012). If Arnold Schwarzenegger, at the age of 73 and coming off heart surgery, were to toss his hat into the ring, he might make a race of it. Beyond that, however, the Democrats could nominate a talking donkey to replace Harris and the donkey would win.

As to the other states where a seat might come open, like Vermont and Massachusetts, they are quite blue, and the Democrats would lay some very careful groundwork before letting a seat come open. In the latter, for example, the state legislature would likely step in and tweak the rules to favor the Democrats (say, by shortening the period between vacancy and special election). That would make it all the easier for Joe Kennedy III, with his name recognition to win. In Vermont, one can imagine some horse trading. "Gov. Phil Scott, why fight the inevitability that a Democrat will replace Sen. Sanders? If you tap a Sanders protégé to replace him, we bet that protégé will find a way to get, say, $5 billion in infrastructure funding for your state."



Q: My question is about the 2022 Senate map. I know this year was, initially, going to be a challenging year for the Democrats. As long as the blue team doesn't take their foot off the gas, it's looking like there's a strong chance of taking back the Senate this go round, and maybe even getting all the way up to 57, as your most recent map suggests. Looking ahead to 2022, what is the outlook there? Is it going to be another challenging democratic election? Especially considering the notion that team blue turns out worse in off-cycle elections? A.B., Columbia, SC

A: Be careful of giving the Democrats most or all of this year's toss-ups; you don't want to be disappointed if they claim a 52-48 majority (which would be an excellent result for them). In any event, 2022 is actually a poor map for the Republicans. They will be defending 20-22 seats compared to 12-14 for the Democrats. And the GOP will also be defending a larger number of vulnerable seats. Republican-held seats in Florida, Iowa, North Carolina, Ohio, Pennsylvania and Wisconsin are up; by contrast there are only three Democratic-held seats that might plausibly be competitive, in Colorado, Nevada and New Hampshire. There are also two seats currently held by Republicans but that might be held by Democrats by then, and that will be competitive either way—Arizona and Georgia.

As you can see, the Republicans will have a minimum of 6 competitive seats, compared to a maximum of 5 for the Democrats, and the Republicans' seats are probably, on the average, in greater danger (e.g., Pennsylvania is much more likely to flip than Colorado; Wisconsin is much more likely to flip than New Hampshire). Further, because the Republicans will be defending more seats overall, there's more potential for a surprise competitive race. For example, it's not impossible that by 2022, Lisa Murkowski (R-AK) could be in some trouble.



Q: Looking at your map, it appears that enthusiasm has dropped for this President in every state that's been polled since the last election. Yet there appears to be a rock solid 40% support in every poll that has anything to do with his performance. Why the discrepancy? S.B., New Castle, DE

Q: We've all heard about "shy Trump" voters. What about the millions of non-voters who are part of Trump's base but did not vote in 2016 because they didn't actually believe he would be such a racist, misogynist and xenophobe but were pleasantly surprised by the last 4 years? I'm worried this "new" bloc of voters is being undercounted. S.S., Raleigh, NC

A: The problem pointed out by S.B. is a semi-mystery that will presumably be resolved, one way or another, next week. One possibility, which we've pointed out before, is that there are a sizable number of "hold your nose, vote for Trump, and hope for the best" voters from 2016 who have lost faith and will not be repeating that behavior in 2020. If that is the case, then the 40% was his base all along, and he was put over the top by another 6-7% who have since jumped ship. A second possibility is raised by S.S.: Trump has lost support among some groups (suburban women, seniors), but those folks have been replaced by voters who were previously apathetic/uninvolved but have now been "activated" by the President. A third possibility is that the polls have missed something, either overstating his loss of support among certain groups, or overstating the strength of his support among those who are still with him.



Q: NBC News has a tool on the Texas page of their website that breaks down early voting and that shows voter makeup so far as 53% Republican, 36% Democratic and 11% Other/Unknown. That would be a Republican wipeout and contrary to all the polling. And if the polls are this far off in Texas, they could be off in many other states as well. Is it time to panic? Or is this a bunch of nonsense that we can safely ignore? I lean toward "ignore," but I would like to hear it from somebody else. J.K., New York, NY

A: We just don't know what to make of those numbers, which come from TargetSmart. That firm typically works for Democratic clients, so it's unlikely they are cooking the books in favor of the Republicans. That said, they have no track record when it comes to this kind of guesstimating because nobody has a track record. There's never been an election remotely like this, and there has certainly never been early turnout like this. So, we have no real basis for evaluating them.

That said, we will point out four things for your consideration:

  1. As reader M.S. pointed out in the mailbag, TargetSmart's numbers are consistently more bearish for the Democrats than in the states where party affiliation is known, and where TargetSmart's estimates are therefore unnecessary.

  2. As a known "Democratic" house, they are probably best served by erring a little bit in the direction of the Republicans.

  3. If you believe that some Republicans, and most independents, are voting for Joe Biden (two suppositions well supported by polls), then TargetSmart's projections aren't actually far removed from polling of Texas. If you give Biden 34 out of 36 Democrats, 6 of 53 Republicans, and 8 of 11 independents, then you have Biden 48% and Trump 52%, which is pretty close to where the polls have it.

  4. Since Texas has made absentee voting very tough for most people, it could be one of the states that sees majority Democratic turnout on Election Day.


Q: I've noticed that the betting markets all give Biden a shockingly low 65% or so chance of winning. What's going on there? Is it just partisan betting? Are the people who are buying Trump to win really just betting on hope? Or is there something to the idea that the betting markets are smarter than the polls? In any case, I've put a few dollars on Biden and hope to pick up an easy 50% on my money... D.S., San Diego, CA

Q: You used to occasionally cite the election odds as reflected on betting sites, where people are putting actual money down on their predictions. What are those sites saying about some of the more high-profile races? R.S., San Mateo, CA

A: We've thought about this question a fair bit. British bettors, for example, have already wagered $284 million with the book Betfair, a sum that leaves every other election (and, in fact, every sporting event) in the dust. That means there surely isn't something fluky that is skewing the results, like a millionaire Trump supporter dropping $5 million on The Donald.

So that's what's not happening. What is happening? We would suggest there are three primary dynamics going on:

  1. There are certain sorts of bets that aren't as "fun" to bet. One of those is the under in sporting events (since that, in effect, puts the bettor in the position of rooting for a boring game). And another of those is where you bet X to get some fairly small fraction of X back in return. In other words, for people who are just looking for entertainment, "a $10 bet will get you $40 in winnings" is much more fun than "a $40 bet will get you $10 in winnings." And so, this tends to push the odds closer together, for most bets.

  2. If trying to choose between the two bets (1-to-2 for Biden, 2-to-1 for Trump), it is pretty easy for someone to convince themselves that Biden's chances are overrated. After all, the pollsters have made mistakes before, right? And who can forget 2016? And what about Russian shenanigans? On the other hand, we rarely, if ever, hear of anyone who thinks Trump's chances are overrated.

  3. Once the money in the pot really starts to pile up, it's pretty hard for the odds to move a bunch. Certainly much harder than it is for polling numbers to move. At the moment, Betfair has taken roughly $187 million in bets on Biden and roughly $94 million on Trump. Let us imagine that every dollar bet from here on out is bet on Biden until the odds match exactly the odds that FiveThirtyEight is projecting right now (Biden 90% to win). Do you know how much money that would take? It would take $686 million in bets on Biden. Obviously, that's not happening.

If you look at the odds for other political events, you can clearly tell that they are out of alignment with reality. For example, PredictIt offers a very broad range of political bets. And if you peruse them, you see all kinds of crazy things. Like, for example, Rep. Steve King (R-IA) is given a 3% chance to win reelection...despite the fact that he's already lost his primary. Or the Republicans are given a nearly 15% chance to take over the House, which is crazy.



Q: I was doomscrolling today and came across Helmut Norpoth's website, as well as other sites talking about increased Republican voter registration in swing states. Would you please talk me down? S.G., Arlington, VA

Q: I found this item on the Internet. It's in German, but in short it talks about Swiss pollster Christoph Glauser, who is bragging that he was the only one who predicted Donald Trump's win in 2016. He says that in 2020, Trump's lead is even bigger than in 2016, and that he will win easily. As basis for his opinion, he is using data from sites like Google, where it seems that people are searching for Donald Trump way more often than for Joe Biden, and from that he concludes broader support for the President. What do you think of this approach? F.L., Innsbruck, Austria

Q: In my usual browsing around Twitter in scattered anxiety, I found someone linking to joeisdone.github.io, a site that claims to be purely data-driven but is obviously Republican-leaning due to its nauseating smugness. Their methodology doesn't seem very sound to me, talking purely about vote-by-mail efforts and ignoring the already massive early voting numbers—but I wanted to know if the experts had anything to say? D.N., Boston, MA

Q: Thoughts on item about Trafalgar pollster Robert Cahaly, headlined "'Shy' Trump voters will power his win, says pollster who called 2016 race"? L.L., Trumbull, CT

A: We got many questions of this sort this week. And before we answer, we would like to remind everyone about Dean Chambers, who ran a site in 2012 called unskewedpolls.com. He was the darling of right-wing talk and "news" programs, because he claimed he had the secret sauce, and that he knew for certain Mitt Romney was going to win. He could not explain what method he was using to "unskew" the polls, especially since he was working with other people's data, and not collecting his own. When Chambers proved to be way wrong, he insisted that his methods were sound, and eventually created a site called BarackOFraudo.com, which purported to "prove" that the Democrats stole the election through chicanery in four states (Ohio, Florida, Pennsylvania, and Virginia). That got Chambers another 15 minutes of fame, until pretty much everyone in the right-wing media figured out that he was making it all up, and it was embarrassing to give him any credence.

We tell this story to illustrate two things. The first is that there is definitely a market for "what people want to hear," especially among those on the right, even if what they want to hear is not rooted in fact. The second is that the risk-reward analysis for being way wrong does not always argue in favor of being right. When you make a wild prediction and you're right, people remember for a long time and you get oodles of credibility out of it. After all, everyone remembers that Joe Namath brashly and correctly predicted an upset in Super Bowl III. On the other hand, who remembers all the players who brashly and incorrectly predicted a Super Bowl upset? Furthermore, when someone is way wrong about political predictions like these, they seem always to come up with an explanation for what happened and how it wasn't their fault. My computer inadvertently flipped the numerator and the denominator, or I neglected to account for the fact that the moon was in the seventh house and Jupiter was aligned with Mars, or a bird flapped its wings in Brazil and that caused a chain reaction that wiped out a million votes in Florida. It's like those doomsday preachers who say the world is going to end on June 7, 2014, and then, when it doesn't, ask "Did I say 2014? I meant 2041! See you in 27 years!"

Anyhow, when we look at the websites of these various folks, we see lots of red flags. For example, Norpoth's site (which we've talked about before) starts by talking about what might happen if Bernie Sanders gets the Democratic nomination. That does not exactly speak to fresh, up-to-date analysis. Trafalgar's site does not list their clients, which is not a good sign with a pollster. And the github site? We looked at it three different times, and we still don't know what it's trying to tell us.

There are even bigger red flags when considering methodology. Like Chambers before him, Trafalgar's Robert Cahaly refuses to explain exactly how he accounts for all those shy Trump voters, he just does. Norpoth's method, which is based entirely on primary elections, has no room for things like...oh, a worldwide pandemic that has killed 225,000 Americans. And Glauser, who is basing his numbers on Google searches? Please. If it was possible to poll in that way, the pros would be doing it, and would not be spending gobs of time and money on telephone polls. Writing for this site, we search for Trump and Trump-related things at least a dozen times a day, and the same goes for Biden and Biden-related things. So, what does that say about our voting behavior? That we're voting for Biden and Trump?

What it really boils down to is this. Real prognosticators show their work. When the votes are counted next week (or the week after that, or maybe the week after that), maybe our prediction will be right and maybe it will be wrong. But at very least, we are transparent about both our methods and our data, so that anyone who wants to double-check our work can do so. Here is our data. Given the choice, we favor a projection like that, as opposed to one based on vague assertions about bias, and a bunch of barely explained black magick.



Q: I would appreciate your comments on investigative reporter Ron Suskind's report indicating that President Trump overturning an unfavorable election result is a very real threat. While I have generally agreed with your analysis that it would require too many people in too many jurisdictions to pull this off, I'm beginning to have my doubts. S.C., Geneva, Switzerland

A: Note that Suskind's piece covers all sorts of nasty things that Trump might do, and only one section involves stealing the election through the use of force. Reading that section, we find it to be extremely vague, and supported only by a couple of fairly unspecific quotes by someone who was unwilling to go on the record with his (or her) actual name. So, we don't put all that much credence in it. Further, if Trump—who, lest we forget, has consistently shown himself to be a coward at heart, as all bullies are—was to try to get violent, we strongly believe that the number of people willing to support that would be vastly smaller than the number of people who would stand in opposition. In particular, we must stress how seriously soldiers take their oaths to uphold the Constitution (and not the current occupant of the Oval Office). And that is before we talk about the fact that Trump's support among the troops is fairly middling.



Q: If the Supreme Court were to stop the vote count in Pennsylvania and Governor Tom Wolf (D-PA) was to say "John Roberts has made his decision, now let him enforce it" and carry on counting, surely Donald Trump or Bill Barr would send US Marshals or the FBI to forcibly enforce the court's decision, right? N.H.R., London, UK

A: They could try, but the counting of votes is decentralized, and would presumably be moved to secure or secret facilities that would not be easy to find or breach. Further, the amount of time it takes to count is surely smaller than the amount of time it takes for a lawsuit, a court ruling, and a deployment of federal law enforcement.



Q: How likely is it that Republicans in swing states will appoint so-called "loyal electors" and flip the election to Trump? V.F., New York, NY

A: If you mean sneaking "moles" into the ranks of Democratic electors, then the odds are zero. It varies state by state but, ultimately, electors are chosen by some aspect of the apparatus of the parties (central committees, or party chairs, or the presidential campaign). And they do their best to pick people whose loyalty they can count on, usually long-term party loyalists or prominent members of the party or both (which is why, for example, Hillary Clinton is an elector this year). Republicans have zero voice in choosing Democratic electors.



Q: Live blogging on election night? Yay and thank you! Where and how do we follow you? Or do we just keep hitting refresh on the web page? J.A., South Salem, NY

A: That's right: just keep reloading. We'll undoubtedly update many dozens of times that night.



Q: With what frequency will you update on election night? How late into the night do you plan to go, barring a landslide? Do you have any concern regarding the ability of your site to handle the peak in traffic you might expect this year? J.G.D., Bellevue, WA

A: We will certainly update multiple times per hour, and we will keep going until there's no possibility of new information. (Z) always stays up very late, and for (V) it will be morning/daytime, so there's no real problem in terms of tiring out. We have moved to a dedicated, quite powerful server, so we do not expect to have any issues with traffic.



Q: Is there a website you can recommend that will give comprehensive running totals of all the elections taking place across the U.S. on Tuesday? Obviously, the national results will get lots of coverage but I would be interested to know what is happening with state legislatures, considering the impact this could have on House districts in future elections. D.H., Leeds, UK

A: The New York Times' election updates page is very good. They have all the results, but they also do a good job of highlighting the interesting and/or important ones, usually with a few words of commentary as to why they are interesting/important.



Q: I really enjoyed your answer to L.J. in the Netherlands about what times to expect results to begin trickling in. Like most of your readers, I'm an election junkie and plan to spend most of the evening/night watching the results (with the prerequisite bottle of whiskey on hand, for either celebration or consolation, depending). My question, as it is every four years, is which channel to watch the incoming results on. The last thing I want is to tune into a network too eager to get the results in first, and that calls a state too early for one candidate or the other. This usually results in channel flipping, much to the annoyance of my wife. What network do you plan on watching election night coverage, or will you too be flipping back and forth between several? J.G., San Diego, CA

A: (V) will watch CNN, because that is what is available in Europe, and (Z) will primarily watch Fox News, so as to have access to a different set of information and commentary. Since the 2000 election, Fox has gotten much more serious about the integrity of their projections, and so we would not expect them or CNN to jump the gun. In other words, you can really trust either of them. Alternatively, you can wait until you hear from the Associated Press; they have the longest and best track record of projecting winners. Generally, on most newscasts, they will say "The AP has now called..."



Q: From your answer to the question from L.J. in the Netherlands, concerning the best way for someone in Europe to watch the election results come in, I infer that you consider CNN as providing the best election results coverage. For those of us in the U.S. who rely on broadcast TV and/or free, no-signup-necessary streaming services, which networks and web sites would you recommend? S.C., Mountain View, CA

A: CNN is good but, as noted above, (V) will be watching that because it's the only game in town (when your town is located in the Netherlands). It's not actually a judgment of quality. In terms of websites, we'd suggest us and the Times page linked above. FiveThirtyEight will have a live chat, and those are sometimes good. The Washington Post may have one, too. If so, theirs are better than FiveThirtyEight's, and they usually put them outside the paywall. As to TV, it is likely that CNN, MSNBC, and Fox News, among others, will be streaming their coverage on their YouTube channels. If you want to watch an over-the-air station on your TV, then the major networks are all fine and dandy (we like ABC the best), but we would actually go with PBS.



Q: If the Democrats could only capture one or the other, the Senate majority (let's say with a comfortable margin) or the presidency, which do you think would be better in the long run? R.K., Minneapolis, MN

A: The Senate. First, because they could pass a mountain of legislation, putting pressure on the White House and making a case for why their candidate should be elected in 2024. Second, because Donald Trump isn't really a Republican, and cares more about wins/his reputation. That being the case, a Democratic-controlled Congress might actually get him to sign some things, as long as they can manipulate him into thinking of those things as a "win." Alternatively, they would be in a position to conduct a real impeachment trial, and not that dog-and-pony show that Mitch McConnell put on.



Q: I know that it is unlikely, but if the presidential selection gets thrown into the House, the Republicans currently have the majority for 26 states, the Democrats have 23 and Pennsylvania is tied. I'm wondering if you could let us know which states are most likely to flip on Tuesday (or whenever the results are known). I've heard there's some chance to flip Montana, which has one Congressman, and Pennsylvania, which is currently tied. If both of those happen, then are we at a 25 to 25 tie. Which are the states most likely to flip? H.R., Jamaica Plain, MA

A: As you correctly point out, Pennsylvania is a tie, which means that if either party picks up a seat there, they gain a delegation. Beyond that, here are the seven places where the Democrats have a chance to improve their numbers, ranked (in our judgment) from most likely to least:

  1. In Florida, they would need to flip 1 seat (of 27) to gain a majority; 3 Republicans are retiring
  2. In Georgia, they would need to flip 2 seats (of 14) for a tie; 3 Republicans and 1 Democrat are retiring
  3. In Wisconsin, they would need to flip 1 seat (of 8) for a tie; 1 Republican is retiring
  4. In Alaska, they would need to flip 1 seat (of 1) to gain a majority
  5. In North Carolina, they would need to flip 4 seats (of 13) to gain a majority; 3 Republicans are retiring due to redistricting
  6. In Montana, they would need to flip 1 seat (of 1) to gain a majority
  7. In Kansas, they would need to flip 1 seat (of 4) for a tie; 2 Republicans are retiring

And here are the eight places where the Republicans have a chance to improve their numbers, ranked (in our judgment) from most likely to least:

  1. In Michigan, they would need to flip 1 seat (of 14) for a tie; 1 Republican and 1 Libertarian are retiring
  2. In Nevada, they would need to flip 1 seat (of 4) for a tie
  3. In Maine, they would need to flip 1 seat (of 2) for a tie
  4. In New Hampshire, they would need to flip 1 seat (of 2) for a tie
  5. In Arizona, they would need to flip 1 seat (of 9) to gain a majority
  6. In Minnesota, they would need to flip 1 seat (of 8) for a tie
  7. In Iowa, they would need to flip 1 seat (of 4) for a tie; 1 Republican and 1 Democrat are retiring
  8. In Colorado, they would need to flip 1 seat (of 7) to gain a majority

There are other places where the math works, but the politics don't. For example, Wyoming also has just one representative, but there's no way that seat is going to be flipped to the Democrats.



Q: If the presidential election ends up being decided by the House of Representatives, which House makes the decision? The current House, or the one elected on Nov. 3? H.B., State College, PA

A: The one elected Nov. 3. That job is one of the first orders of business for the new Congress, and will be taken care of on Jan. 6.



Q: I just made my second donation to the Biden campaign and it felt great. With all the talk of post-election lawsuits (and maybe the need for ads to influence public opinion), are campaigns allowed to set aside donations for those purposes? Please tell me Team Blue has a post-election strategy with this in mind? J.R., Boca Raton, FL

A: Yes, they are allowed. They are also allowed to raise funds specifically for that purpose. And Team Blue has a vast army of lawyers on retainer (as does Team Red) to deal with any post-election legal challenges. They will also have no issue finding volunteers, should they require them.



Q: If Joe Biden wins, couldn't he just reverse all of Donald Trump's executive orders in the first 90 days? J.K., Portland, OR

A: Many of them, yes. All of them, no. First of all, there are certain issues where patience is required. For example, if Biden intends to make changes as regards automobile fuel efficiency, he's got to give the auto makers some warning. Second, as Trump himself has taught us, it is necessary for a president to follow good practices when issuing XOs, and to have some clear and non-discriminatory basis for his choices. This is why, for example, the administration has struggled to get rid of DACA, despite Stephen Miller's repeated attempts.



Q: Is Donald Trump required to attend the inauguration if he loses? I can't imagine he will go to Joe Biden's inauguration voluntarily, as all cameras would be on him. He would be humiliated. Have other past one-term presidents refused to attend their successor's inauguration? S.H., Delaware, OH

A: It is not required, and surely he will skip out (if he hasn't already fled the country). The only previous presidents to skip their successor's inaugurations (beyond the nine presidents who died in office/resigned, of course) were John Adams, John Quincy Adams, and Andrew Johnson.



Q: How much of a "gap" exists between the last poll(s) of the election (exit polls notwithstanding) and the results? Put another way, in states that are close, how out of date will our information be on election night? S.M., Maple Plain, MN

A: Not much of a gap, particularly this year. As you can see, the polls are coming fast and furious, even though we're just a couple of days from the election. And the moment the regular polls stop is the moment the exit polls begin.



Q: How is early voting reflected in the most recent polls, in terms of methodology and accuracy? C.E., London, UK

A: In terms of methodology, it's actually solving the trickiest problem that pollsters face: figuring out who will actually vote, and who won't. Normally, pollsters begin election season by talking to registered voters, and then they shift to talking to "likely" voters. But it can be tough to figure out how likely a "likely" voter really is, and the various pollsters work hard to figure out which "likely" voters might be more flaky than they claim.

Anyhow, someone who has already voted is most certainly a "likely" voter, and so the pollster can have faith in their responses. Indeed, a few pollsters (like The Wall Street Journal) have limited themselves exclusively to people who have already voted. All of this will, on the whole, make the polls more accurate, since it removes some of the guesswork.



Q: In an election where turnout is higher than expected, isn't it likely that the polls will be inaccurate? I guess I'm assuming that the "extra" voters are not equally represented across all demographic buckets. D.M., New York, NY

A: Actually, as with the question above, the polls are likely to be more accurate. It will be easier for the pollsters to correctly estimate turnout, and also to know what the electorate is going to look like.



Q: With respect to shy Trump voters: maybe they are not shy, but rather refuse to engage with pollsters in greater numbers than Biden voters. If larger numbers of Trumpers hung up the phone or refused to click on the polling link in their email, wouldn't that skew the polls toward Biden? B.R., Chiang Mai, Thailand

A: No, not really. The pollsters would just keep calling until they got enough Trump voters to respond. And if they came up a little short (or even more than a little short), they can correct for it substantially with their weighting. If they believe 45% of the voters will be Republicans, but their sample is made up of only 30% Republicans, then they just weight each of those respondents as 1.5 people.



Q: If not the "shy Trump" voter, how about the Trump voter who revels in the idea of lying to pollsters? I have heard one or two interviews where Trump voters openly state they lie to poll questions because the press is evil. M.E., Syracuse, NY

A: We don't believe many of these people exist. First of all, in our experience, and like Trump himself, a lot of these folks talk the talk but they don't walk the walk. Second, it is unlikely that pollsters would consistently reach roughly the same number of liars. And yet, the polls (particularly the national ones) are quite consistent.



Q: As I look at the graphs for Minnesota, Wisconsin, Michigan and even Pennsylvania, the polls have consistently shown Biden with a decent lead over Trump, though some are within the margin of error. While you and others continue to point out these polls are a snapshot in time, the snapshots show a consistent lead for Biden. Doesn't that temper the "snapshot" argument? J.B., Williamsburg, VA

A: It does. When many polls, from many different pollsters, reach the same basic result, it significantly reduces the chance that result is an anomaly, or is caused by a wonky polling sample. That said, there still remains room for the possibility of an inaccurate model of the electorate and/or late movement in the polls due to some sort of October surprise.



Q: If the advantage in a state is within the margin of error of the surveys, why do you not list the states as "tied" instead of "leaning" one way or the other? M.B., Santo Domingo, Dominican Republic

A: Because the result produced by the poll is still the likeliest to be correct. And if we limited ourselves only to polls entirely outside the margin of error, we would be working with far fewer (and possibly wonkier) polls, and we also wouldn't really be able to project most of the swing states. If you want to ignore the close states, just add up the "strongly" and "likely" numbers in the legend to the right of the map. Or look at the Electoral-vote graphs page and focus on the second graph.



Q: Joe Biden leads by approximately 8 to 10 Points in national polling. Are there any major elections (presidential election, U.S. Senate election, gubernatorial election, election for the U.S. House of Representatives) in U.S. history in which a candidate won despite trailing by double digits in the last poll prior to the election? F.S., Cologne, Germany

A: None that we can find, because elections that are that big of a blowout usually don't get polled at all.



Q: You often hear about how a campaign's internal polls are markedly different from ones publicly released by polling firms. Are these internal polls "better" than the public ones? What information do campaigns have access to that polling firms don't have that would improve their modeling over public pollsters? Why would an internal poll come up with numbers that are much different from a public poll? Why do campaigns even bother running their own polls? J.P., Cleveland, OH

A: Pollsters, who are usually working on a somewhat tight budget, are trying to assess the state of a race. Parties, which usually have much more generous budgets, want to assess the state of the race, but are also interested in figuring out what they need to do next. So, they are more likely to have information on what voters say they want (would you favor a 10% increase in the use of solar power?) and they are much more likely to have very fine-grained information on how a neighborhood, or a congressional district, or a county is trending, so they know how to spend their money and where to direct their phone calls/text messages/door-knockers/mailers.



Q: A bit of clarification, please, on your inclusion of the Georgia special Senate election: Are the numbers you cite polls for the run-off in January, or for the jungle-election in November? R.E.M., Brooklyn, NY

A: The jungle election is Nov. 3. The top two will advance to a runoff on Jan. 5, 2021. A number of pollsters are already polling various runoff combinations, for example, Loeffler vs. Warnock, Collins vs. Warnock, etc. Our Georgia-special polls are the Loeffler vs. Warnock polls. Of course, if the runoff is Collins vs. Lieberman, our data is useless. But our best bet now is that it will be Loeffler vs. Warnock and the numbers we are using are for that race.



Q: Why are you leaving Maine out of our poll reports so often? Is it because you feel that Maine is unimportant? R.L.B., Bridgton, ME

A: We can only run the polls that, well, exist. As you can see here, we've run 16 Maine presidential polls this cycle. And as you can see here, we've run 19 Maine Senate polls. That hardly seems like neglect to us.



Q: The national Rasmussen poll you reported made me wonder again why you exclude them. They probably aren't skewing their results deliberately, since it will be hard for them to drum up business if their most prominent poll of the election cycle turns out to be off by 12 points. So they must honestly think they know something about the electorate that the other pollsters don't. The premise of your site is that crowd-sourcing gives the best picture available, so why exclude them if they don't fail your 538-ranking criterion? M.S., Jerusalem, Israel

A: As we note above, you should not assume that accuracy is a pollster's primary concern, or that inaccuracy will necessary destroy their reputation. In any event, we use FiveThirtyEight's pollster grades as a quick shortcut in many cases, but we also sometimes review pollsters on a case-by-case basis and reject those that do not pass the smell test. Rasmussen does not pass the smell test, and has not for years, given their heavily partisan client list and their questionable methodology.



Q: I remember you broke your "Rasmussen-Free Zone" policy not too long ago to make a point in a particular item you ran, but you have not included them in any of your poll tracking. However, I did notice you included polls from the presidential and senatorial races in Arizona from Pulse Option on Friday. Doesn't that polling house operate under the Rasmussen umbrella? If so, why would you run polls from them and not the parent operation? D.F., Norcross, GA

A: It used to be that Pulse was good enough, and separate enough from Rasmussen, to be just on the right side of the line, getting a C+ from FiveThirtyEight. Now, their orbits are back to overlapping so much that we aren't sure what to do. But given the very large number of pollsters, not using Rasmussen but using Pulse is probably a reasonable compromise.



Q: Just wondering how you choose which polls you use for reference? Is it the most recent one, or the most reliable? R.S., Chicago, IL

A: We eliminate pollsters with a grade lower than C on FiveThirtyEight, as well as those who raise red flags, as described above. Beyond that, we use all the polls we are aware of. The default rule is that if a pollster is new and unrated and there are no red flags, we use it. The map uses the most recent poll, no matter how old, and any other polls within a week of it. They are then averaged with equal weights. If you put your mouse on North Carolina on the map (but don't click), you will see the rating is based on eleven polls. FiveThirtyEight's pollster ratings are here.



Q: In 2016, The Los Angeles Times poll consistently had Donald Trump winning. I've yet to see them listed in any of your presidential polls this year. My question is, are they still polling this year? Also, could you direct me to the link that shows FiveThirtyEight's grading of pollsters? J.M., San Jose, CA

A: That is the USC/Dornsife poll. They are still polling, but they only do national polling, not state polling. So, we sometimes mention them in an item (as we did a couple of times this week), but they are not in our database because our database has only state polls.



Q: If the Democrats were to win the Congress and Senate and then expand the size of the Supreme Court, could the current SCOTUS just nullify this using judicial review? P.A.F., Los Angeles, CA

A: Not legally. The power to set the size of the Court is specifically granted to Congress in the Constitution, and is also supported by centuries of precedent. The Congress could add a further insurance policy by adding a passage specifically denying the Court appellate power over the bill. That's another privilege granted to Congress in the Constitution.



Q: Is there a liberal equivalent of the Federalist Society? R.L., Alameda, CA

A: Yes. It is called the American Constitution Society (ACS), and was founded as a direct response to Bush v. Gore. However, the ACS has nowhere near the funding or the membership of the Federalist Society. Further, while Republican presidents crave the Federalist Society's blessing for judicial picks, Democrats prefer to rely on the American Bar Association, prioritizing competence over ideological purity. So, the ACS has nowhere near the influence that the Federalist Society does.



Q: The fact that John Roberts sided with the liberals in the recent ballot question makes me wonder if he doesn't want the court to so obviously subvert democracy. I thought that the Chief Justice can choose to take a case or not. Is that true? M.L., Portland, ME

A: Not exactly. For the court to accept a case requires an affirmative vote from any four justices. And so, Roberts can be one of those votes, and he can certainly engage in a little politicking and/or arm-twisting, but he does not have sole discretion over which cases the Court takes.



Q: Can you explain then why Franklin Delano Roosevelt failed to pack the Supreme Court? M.S., Norwood, MA

A: Well, he proposed the plan, and sent it to the Senate, where it was not popular, and where it was ultimately defeated by an overwhelming 70-22 vote.

With that said, the mere threat was enough to cause two of the anti-New Deal justices to rethink their positions, and another to retire before that vote was held. Consequently, court packing became unnecessary, and FDR made no particular effort to win the Senate over to his point of view. So, there's a case to be made that it was all for show, and that he never seriously intended to follow through.



Q: Last week you responded to a question regarding running for president more than once. You cited two GOP candidates, Nixon and Reagan, and indicated the third time was the charm for them.

Noticeably absent from your list were any Democratic candidates who have run three times (or more) and then won. By my counting it hasn't happened since at least the start of the 20th century: Woodrow Wilson in 1912, FDR in 1932, Harry S. Truman in 1948, John F. Kennedy in 1960, Lyndon B. Johnson in 1964, Jimmy Carter in 1976, Bill Clinton in 1992 and Barack Obama in 2008. They all seemed to have won on their first campaign for President. This would seem to be a bad omen for Joe Biden, as this is his third attempt.
H.M., Tallahassee, FL

A: Nixon ran three times because he started his career so young, and Reagan did so because he didn't win until he was pretty old. The reason that there's no Democrat on the list is that a would-be president only has so many opportunities to run, in view of the fact that challenging an incumbent from one's own party is gauche and generally foolhardy. And because the Democrats have a propensity for electing men in their mid-to-late 40s or early 50s, they just haven't usually had enough possible cycles to build up three failed bids. That said, FDR was on a national ticket in 1920 as James Cox's running mate, and LBJ was most certainly a candidate in 1960, meaning he did have two runs, if not three.



Q: Bill Clinton managed to win states like Louisiana, Arkansas, Tennessee, and Kentucky twice. What has happened since 1996 so that we don't even remotely consider these states competitive anymore for the Democrats at the presidential level? S.K., Austin, TX

A: In part, Clinton's presidency coincided with the death throes of the old school "Southern Democrat," some of whom had yet to fully admit to themselves that they were now aligned politically with the party of Lincoln. And in part, Clinton was centrist-to-conservative, and the Democrats haven't had a taste for candidates like that in the last 15 years or so. If they nominated Sen. Joe Manchin (D-WV), they might well pick up a Louisiana or a Tennessee.



Q: You mentioned that "women (largely) could not legally vote in 1908." I thought that women couldn't vote at all until 1920. Were some women able to vote before 1920? J.H., Glendora, CA

A: Until the 19th Amendment made women's right to vote a federal matter, it was up to the states. And starting with Wyoming in 1869, a number of Western states extended suffrage to women, largely as an incentive to get them to relocate and to give some gender balance to populations that were 98% male (because most of the people who traveled west were men pursuing jobs as miners or cowboys or railroad builders). Colorado, Utah, and Idaho had followed suit by the end of the 19th century, and in the 1910s the progressives persuaded Washington, California, Oregon, Kansas, Arizona, Illinois, and Alaska to join the list.



Q: A few months back I wrote and suggested you discontinue the use of the term Midwestern states when referring collectively to Wisconsin, Michigan, and Pennsylvania, those that Trump unexpectedly won in 2016. I said it then, but my rationale bears repeating - Pennsylvania is not part of the Midwest. I suggested instead the term Rust Belt, since it's an accurate way to describe the same place (certainly as accurate as Sun Belt in describing the collection of North Carolina, Georgia, Florida, Arizona, and occasionally Texas) which I don't think should offend anyone in the region. It seemed like you thought it was a good enough idea since I noticed you use it pretty regularly after I submitted that suggestion. And thanks, by the way.

But now in the last month or so, I've seen you using the term "Midwest", in quotes like that, or occasionally just Midwest. What changed in the EV style guide to prefer this over Rust Belt, or even something else? Personally, I think "Midwest" is an even worse term than just Midwest without quotes, because it still implies to us Pennsylvanians that we're part of the Midwest, while running the risk of implying to actual Midwesterners that they are not.

If you don't like Rust Belt for whatever reason, here are some other suggestions:

  • Great Lakes states. Descriptive, technically accurate, kinda boring.
  • WMP, which could be pronounced as "wimp" and maybe even spelled as such if you need to loop in Iowa, or Indiana for some reason.
  • You could expand on that with various other acronyms including relevant nearby states. WOMP? WIMPO? MOP? Lots of options here.
  • Pierogi Alley. Self-explanatory, and truthful.
  • Rust Belt. I know, but seriously why not? This is what we do. We just sit here and rust.

I rest my case. C.M., Spring Brook, PA

A: We started using Rust Belt, but some denizens of that part of the country complained that term was insulting to that part of the country, since it suggests those states are decaying and past their prime. We couldn't think of a better option, so we settled with "Midwest." If you had suggested Pierogi Alley earlier, we might have gone with that.



Q: I joined the Democratic Party when I was old enough to vote, 20 years ago. However, around 2016 I started noticing my Republican friends calling it the "Democrat Party." I've seen this used more and more often since then, to the point where it's almost exclusively called the Democrat Party by them.

What is this all about? Why are they changing how people refer to the party? Is it some kind of slur?
J.A., Shepherdsville, KY

A: It is indeed an insult. Usually, when spoken, the last syllable is accented, so it comes out as DemocRAT. Further, following Donald Trump's lead, many Republicans have decided it's fun to deliberately mispronounce their adversaries' names. Witness all the folks who deliberately mispronounce Kamala Harris' first name.



Q: When Trump whines "Russia, Russia, Russia!" do you think that's an intentional homage to the classic "Brady Bunch" episode where jealous Jan laments, "Marcia, Marcia, Marcia!"? Or is it just political/pop-culture serendipity? J.F., Irvine, CA

A: Well, that's a show about a well-to-do professional whose second wife is an attractive blonde, who appears to interact exclusively with white people, whose kids are constantly getting into trouble, and who lets "the help" do much of the actual parenting. We are willing to believe that Trump's a fan.



Q: In last week's Q&A, you noted that The Washington Post has some questionable right-wing columnists that seem to be there only to "balance" the opinion section, namely Marc Thiessen and Hugh Hewitt. The New York Times does the same thing, most notably with Bret Stephens (best known for definitely not being a bedbug) and until recently Maureen Dowd.

So, in your opinion, what's the deal with this? Do these news outlets feel the need to do the obnoxious false balance thing because it helps keep subscriptions up? Do they believe that it's genuinely good to have people to challenge the assumptions of their presumably left-leaning reader base? Or something else?
A.L., Oakland, CA

A: Our best guess is that it's a defense against "liberal bias in the media" complaints. "How can you say we don't incorporate conservative voices when we have Marc Thiessen and Hugh Hewitt on staff?" is the general idea.

We doubt that it actually placates conservatives, and it surely doesn't do anything for liberals, or for anyone who is interested in a free and open exchange of thoughtful ideas. And by the way, we also dislike columnists who reflexively adopt the obvious left-wing position, like Sally Kohn or Jeff Yang. Either way, they're not telling us anything we didn't already know.



Q: I never thought that I'd make it into a 1% Club. I'm glad that it's here at this site. Anyhow, in reference to the questions posed by E.M. in Chicago, I wanted to expand a bit more on your readership and their feedback. Of those that are in the 1% Club, what percentage have responded to content from the site more than once, and what percentage has been of the negative variety? J.P., Kansas City, KS

A: This is just a guess, since we don't keep formal statistics. However, we would guess that once someone writes in once, they write in again at least 90% of the time. As to how many of the comments are negative, we assume you are asking about people who send in hate mail, and not those who send in critical but constructive criticism (or corrections). The hate mail isn't a huge percentage, maybe 2%-3%. It often arrives in clusters, however, clearly prompted by some sort of message posted to a conservative Facebook page or mail list or the like. For example, we got five e-mails in 10 minutes from five different people asking why we deliberately exclude Rasmussen polls, since they are the only "honest" pollster. We don't have a problem explaining our position on Rasmussen (as we did above), but the only honest pollster? Please.



Q: I've followed you daily since 2015. If Trump doesn't manage to steal the election, and assuming Biden's presidency is refreshingly free of scandal, corruption, and human rights violations, what will happen with your site for the foreseeable future? S.S.L., Norman, OK

A: We haven't thought that far ahead, but we are not likely to go dark for a year. We may possibly reduce the frequency of postings a bit, for want of enough news. Or maybe not, since we have a bunch of ideas we haven't been able to carry through, since the amount of news to be written up didn't leave room for them.

Today's Presidential Polls

We're three days from Election Day, and the polls of Pierogi Alley keep coming. Clearly, the pollsters don't want to get caught with their pants down again. Meanwhile, those who fear a constitutional crisis have to be pleased that North Carolina and Florida may well be decided early and cleanly. (Z)

State Biden Trump Start End Pollster
Connecticut 51% 26% Oct 08 Oct 21 Sacred Heart University
Florida 48% 46% Oct 16 Oct 26 YouGov
Florida 52% 45% Oct 28 Oct 29 PPP
Michigan 51% 44% Oct 27 Oct 29 RMG Research
Michigan 54% 41% Oct 21 Oct 28 Kiaer Research
Michigan 54% 44% Oct 29 Oct 30 PPP
North Carolina 47% 48% Oct 28 Oct 29 Pulse Opinion Research
North Carolina 49% 47% Oct 27 Oct 28 East Carolina U.
North Carolina 52% 46% Oct 25 Oct 28 Marist Coll.
Pennsylvania 52% 45% Oct 28 Oct 29 PPP
Wyoming 31% 59% Oct 08 Oct 28 U. of Wyoming

Click on a state name for a graph of its polling history.

Today's Senate Polls

For John James, Sen. Thom Tillis (R-NC), and MJ Hegar, the clock appears to be running out. For Merav Ben-David, it never started. She probably didn't really expect to win though, but cuddling with polar bears all day can get boring (see our Senate page). (Z)

State Democrat D % Republican R % Start End Pollster
Michigan Gary Peters* 51% John James 38% Oct 21 Oct 28 Kiaer Research
Michigan Gary Peters* 54% John James 44% Oct 29 Oct 30 PPP
North Carolina Cal Cunningham 47% Thom Tillis* 44% Oct 28 Oct 29 Pulse Opinion Research
North Carolina Cal Cunningham 47% Thom Tillis* 46% Oct 27 Oct 28 East Carolina U.
North Carolina Cal Cunningham 53% Thom Tillis* 43% Oct 25 Oct 28 Marist Coll.
Texas Mary Hegar 42% John Cornyn* 48% Oct 27 Oct 28 RMG Research
Wyoming Merav Ben-David 26% Cynthia Lummis 56% Oct 08 Oct 28 U. of Wyoming

* Denotes incumbent


If you wish to contact us, please use one of these addresses. For the first two, please include your initials and city.

To download a poster about the site to hang up, please click here.


Email a link to a friend or share:


---The Votemaster and Zenger
Oct30 Courts Get Involved Again, This Time in Minnesota
Oct30 Things for the Democrats to Worry About
Oct30 More on "Shy Trump" Voters
Oct30 Right-wing Media Try to Salvage Hunter Biden Story
Oct30 On Your Marks, Get Set, Go!
Oct30 The Delicate Art of Question Dodging
Oct30 Donald Trump, Flight Risk?
Oct30 Today's Presidential Polls
Oct30 Today's Senate Polls
Oct29 Biden Continues to Lead in the National Polls
Oct29 Early Voting Has Hit 51% of the 2016 Total Vote
Oct29 Anonymous Isn't Anymore
Oct29 Where Are the Candidates?
Oct29 Democrats Are Now with Trump
Oct29 A New Front in the Voting Wars: The Order of Counting Ballots
Oct29 Overseas Military Ballots Could Be Crucial in Florida
Oct29 Whose Fault Is It?
Oct29 Senate Rundown
Oct29 Schumer's Relationship with McConnell Is in Tatters
Oct29 Whither the Supreme Court?
Oct29 Today's Presidential Polls
Oct29 Today's Senate Polls
Oct28 Melania Trump Hits the Campaign Trail
Oct28 Jared Kushner Is Not Helping His Father-in-Law
Oct28 Biden Decides to Do a Little Swinging
Oct28 The Ballots Are Pouring In
Oct28 Abbott Wins the Ballot Box Battle, But Appears to be Losing the War
Oct28 Trump Campaign Backs Off in Florida
Oct28 One Last Funny Feeling
Oct28 Today's Presidential Polls
Oct28 Today's Senate Polls
Oct27 Barrett Is Confirmed...
Oct27 ..And May Soon Be Mucking Around in the Election
Oct27 Trump Thinks Media Should Not Cover COVID-19...
Oct27 ...Probably Because He's an Autocrat...
Oct27 ...Which Is Absolutely Killing the Republican Party
Oct27 Lou Dobbs Knows Who Is to Blame for the Trump Administration's Failures
Oct27 Six Reasons Not to Panic About the Election
Oct27 Today's Presidential Polls
Oct27 Today's Senate Polls
Oct26 Murkowski is Not Concerned and Will Vote to Confirm Amy Coney Barrett
Oct26 Nearly 60 Million Voters Have Already Cast Their Ballot
Oct26 Could COVID-19 Affect the Election?
Oct26 Could Trump Win the Midwest Again?
Oct26 Did Biden Slip on a Oil Slick?
Oct26 Pennsylvania Supreme Court Rejects Rejected Signatures
Oct26 Biden's Campaign Has Spent More on TV Ads Than Any Campaign in History
Oct26 Democratic Senate Candidates Outraise Incumbents Again
Oct26 How Trump's Digital Voter Suppression Operation Works
Oct26 A Voter's Guide to Worrying about the Election