We are now just over three weeks away from the election, which means a lot of things. First, it means you should vote-most states have absentee ballots & early voting in full swing (or will by Friday), and quite frankly if you're voting absentee, you should have your ballot in the mail at least two weeks prior to the election, so this is (practically-speaking) your last days to vote. Get it done. Especially with Covid, you don't want to be within ten days from the election & suddenly find yourself fighting the disease...and can't safely go to the polls without infecting other people.
But you knew you're supposed to vote if you're reading this blog or literally have been on any social media in the past month. What it means for me, personally, is that there's a few articles that I wanted to start getting out related to facts, figures, & hypotheticals about this election. Therefore, my goal this week is going to be to have a week entirely focused on politics, starting today through next Tuesday, most of which will relate back to this year's election (we'll still keep doing our articles devoted to October's Monster Movies & our Star of the Month, Nancy Kwan). Then, we'll be back to some politics (if something comes up, as well as resuming our regular 2019 Oscar articles) next week, and I'll be finalizing my predictions in every major contest, which I'll be publishing every day the week before the election. I don't believe in skipping races or going into Election Night with "Tossups," so today I wanted to talk through some of the biggest questions I have as I start to prepare for my giant biennial elections preview, either in terms of interpreting data or disregarding my own personal prejudices about the data, as we approach that final week of predictions. I suspect some of the below questions if you're trying to read the tea leaves of this election are on your mind as well. Let's discuss.
1. How Wrong are the Polls?
Will Biden avoid Clinton's fate?
Four years ago, despite common discussions now, the polls weren't that wrong. Most national polls got it right, that Hillary Clinton would win the popular vote, but not by much. Where they got things wrong were in states like Wisconsin & Michigan, under-polled states in a volatile election (Clinton maintained her lead throughout, but its margin ebbed & flowed dramatically), where they didn't control for education, much to the advantage of Donald Trump.
In 2020, a few of these issues are solved for. Most reputable polling firms control for education now in a way they didn't before, so that should (theoretically) be more accurate in 2020. Biden's lead has also been maintained all year, but by a solid margin-Biden has, in fact, grown his lead in the past few weeks from 6-8 points to 9-12, though that could mellow out the further we get from Trump's disastrous debate performance. And state-level polling has also been much more consistent (with only a couple of exceptions like Florida & Arizona), enough so that the polling is all saying the same thing-a Biden win.
But polls are always a little bit off, and inevitably they could be more than a little off with so many strange aspects to this election (the pandemic, the delayed postal service, and a uniquely unpopular president all make turnout predictions more difficult). The question is-in what direction? You'll hear a lot of people say that if the polls were wrong in the same way as they were in 2016, Biden would still win in 2020, albeit by a narrower margin than what we're seeing right now. But the polls aren't going to be wrong in the same way. They might be less wrong in either direction (resulting in a Biden win), they might be more favorable to Trump than four years ago (that's really the only way he wins at this point), or because pollsters tend to over-compensate in these scenarios, they might be under-estimating Biden by a larger margin. We'll get to this below, but there are races that have historical trends (Arizona tends to be red, Ohio/Iowa are trending that way) going up against polling (Arizona is clearly leaning blue for Biden, the other two states are tossups), that if, say, Biden is being under-polled could deliver unexpected victories. I'm going to be watching the polls over the next two weeks as I finish up my predictions, but guessing how big a win might be for Biden (or to gage Trump's chances), you have to assume the polls are slightly wrong...but how & in what direction is up for debate still, and something we'll hopefully get a better handle on in the next two weeks as we see shifts in the remaining undecideds in the electorate.
I'm not going to try & predict, either here or in my elections article, legal issues that will happen as a result of the election. It's clear at this point that predicting Donald Trump's behavior is a lost cause (we'll see what happens in that regard after the election, though I personally wish they'd stop asking if he'll concede, as it makes it seem like that matters or if he has a choice in stepping down-he doesn't), and I can't predict what states might end up in court. It's also clear, though, that there are efforts by Republicans to stifle turnout, or to invalidate as many mail-in ballots as possible. The question for predicting is-how egregious will that be? And will their efforts actually backfire as Democrats seem to be more likely to return their ballots early this year?
While it's a fool who looks at early voting returns and assumes Democrats are leading (with the pandemic & the push to vote early by one party and not the other, it's impossible to tell whether these are voters who would've voted for Biden anyway or actual signs of an expanding electorate...voting early is brilliant for campaigns because it helps bank voters & has you focus on a smaller pool to turnout, but pundits interpret at their peril), voting early does have its drawbacks, as we've seen with "naked ballots" in Pennsylvania or the lack of drop boxes in major metropolitan areas in Ohio & Texas. The question isn't whether it will decrease the Democrats margin-it's whether or not it will enough to decide any contests, and I just don't know right now. The huge push by Democrats to vote early & increase access also matters here-if people who wouldn't have normally voted do vote early, that will help Democrats, possibly enough to outweigh these anti-voting measures. But it's arguably the hardest thing to predict, even as "naked ballot" numbers start to come in from key states. Of all of these questions, it's the one that I think about the most, and have the least answers to interpret. But it will be a factor on Election Day (if potentially not a deciding one).
3. Is There Any October Surprise That Could Still Happen (That Will Matter)?
President Donald Trump (R-FL)
Millions of people have already voted, particularly in major metropolitan areas. Hennepin County (Minneapolis) has already gotten 24% of its registered voters to the polls, while Dane County (Madison) continues to astound with 30% of its registered voters already casting ballots. These are Democratic strongholds posting gargantuan numbers in swing states. It's possible that by the time the next debate happens (if it happens), more than half the country could have already voted. While these people certainly made up their minds early, that doesn't mean they weren't universally winnable in the back-stretch of the election, and thus if the current environment favors Biden, its begs the question...is there an October surprise left, and will Trump have enough voters left over for it to make a difference?
The potential for a surprise is always there. Obviously the biggest question remains the candidates' health. Donald Trump already has Covid, and while it seems (publicly) like he's on the mend after a close scare, Covid is a strange disease. Any scare in his health (or the health of Mike Pence, Joe Biden, or Kamala Harris) would matter in the waning days of the campaign. This is also true for any key Senate or House campaigns. And of course Trump's tax returns remind us that there's always a late-breaking story that could pop up. But October surprises won't have the same impact they did four years ago (with the James Comey letter) because of the emphasis on voting early, and thus any remaining twists in the campaign could be of increasingly limited impact.
4. Will Any of the "Lean R" Senate Races Break?
Sen. Lindsey Graham (R-SC)
We now shift gears to the Senate (I'll be predicting not only the presidential race, but also all of the congressional & gubernatorial races the last week before the campaign). There are a few things that seem likely in the battle for the Senate. Unless something strange changes, Alabama will flip red, while Maine, Colorado, & Arizona go blue. We'll get to North Carolina in a second, but all of the other blue seats seem safe (even if Michigan is a bit tenser than it should be). The big question for me is the "Lean R" races, and this goes back to-what if the polls are underestimating Biden?
Right now, if the polls are right, the only other seat that is probably at risk for a flip is Iowa, which is the truest of tossups. But if Biden is being underestimated (particularly if there's a groundswell of straight-ticket voting or if there's a lot of new Democrats in the closing days of the campaign, which is possible), we could see some of the other Republican-held seats with competitive Democrats switch seats. This has happened before (1980 & 2008 being the most obvious examples), where the victor won by enough that he transformed the Senate. Montana, South Carolina, Georgia (both races), Kansas, Alaska, & even Texas have the potential to flip in the right scenario, and these candidates have shown competitiveness in the polls, but the fundamentals of the races favor the Republicans. But most of these candidates (save Georgia & Texas) are outrunning Biden, and GA/TX Biden could win. If Biden's leading by more than the polls, or even the high-end of the polls, it's probable that at least one of these seats flip, and potentially a number of them if he does well enough.
One of the stranger aspects of this election has been, with shockingly consistent polling in the presidential race, why the Republicans haven't been able (so far-they're clearly trying to do this covertly) to run a checks-and-balances campaign against Biden. Essentially this is when you say "we know we're losing the White House/House, so either vote for us or the Democrats will have no one to keep them in check." Republicans essentially did this in 1996 when Bob Dole was obviously going to win the White House, and it worked-the GOP picked up two seats & won Senate races in 5 states that Clinton won in the presidential election.
This isn't working in 2020 for two reasons, though. One-Republicans running this kind of campaign in a public enough way would run the risk of Trump deriding them, thus costing them potential support from their base (something Dole, a longtime party man & former congressional leader, wouldn't have done), and two, despite robust polling showing Biden as the heavy favorite, many people still assume Trump will win. Because he beat the (much more marginal) polls in 2016, the perception is that Trump can do it again (he might be able to), and thus it's harder to convince people voting for Biden that they should have a Republican check in the Senate because they are convinced Biden won't win.
5. Will Cal Cunningham's Scandal Matter?
State Sen. Cal Cunningham (D-NC)
There is very little good news for Republicans in this article, but this is the one glimmer for them. State Sen. Cal Cunningham has been involved with a scandal that's broken in the last two weeks, where it was revealed this the married Cunningham had an affair with at least one woman, and some of their intimate correspondence was made public. While it might sound odd to assume that Sen. Thom Tillis is able to make hay out of this (relatively tame, so far) scandal while also running on the same ticket as Donald Trump (who publicly cheated on his current-wife with an adult film star, and then one of his attorneys paid the actress to keep quiet about the affair), Democrats seem to be held to a higher standard when it comes to these types of indiscretions. Most (including me) assumed Cunningham was likely to win, and would be the fiftieth seat for the Senate Democrats, and he still could be-initial polling shows limited movement in this race, where Tillis remains quite unpopular. However, Cunningham doesn't have a lot of ground to lose before this becomes Tillis's race, and so it's possible that his affair could cost the Democrats the Senate if they don't win here (or find another spot on the map in redder territory to make up for Cunningham's loss).
6. Straight Ticket Voting in the House Races?
Rep. Kendra Horn (D-OK)
There are currently 30 seats held by Democrats where Donald Trump won in 2016, and six such seats held by Republicans that were won by Hillary Clinton in 2016. One of the big questions for November 3rd is-will these seats stay for Trump (or Biden holding the Clinton seats), and if they do, how common will ticket-splitting be? In 2016, no Senate race went for a different party for President & Senate, and while this wasn't the case for all House seats, in the Trump Era ticket-splitting is an endangered species, and thus these incumbents are clearly vulnerable.
It's worth noting, of course, that it's probable that Trump doesn't win all of these seats again in 2016 (Biden is currently favored to win all six of the Hillary ones, making these incumbents range from surefire losers like NC-2 and TX-23 to potential pickups in New York & California). IL-17 and NV-3, for example, barely went to Trump, could well go for Biden, and no one assumes that their incumbents are vulnerable. But there are obvious seats like New Mexico-2, Oklahoma-5, & Minnesota-7 where Trump is likely to win-how much can the Democrats overrun Biden even in the best of circumstances, or are people like Kendra Horn & Collin Peterson goners?
There's also the question of if Biden over-performs enough in the suburbs, how many additional districts that he wins that went to Trump four years ago. There are several districts, even in Trump states, that went for statewide Democrats in 2018 like GA-7 (for Stacey Abrams), IN-5 (for Joe Donnelly), MO-2 (for Claire McCaskill), and TX-10/TX-24 (both went for Beto O'Rourke). This could've been just a sign of a good midterm, but if not most of these seats are open, and thus could be pickups for the Democrats.
7. The Toughest Races to Call
State Sen. Wendy Davis (D-TX), candidate in the 21st
There are a few races that are genuinely too hard to call right now. If I had to corner it down to ten, it'd be NC-President, OH-President, IA-Senate, SC-Senate, NC-Senate, MN-1, NY-11, OK-5, TX-21, & VA-5. These races are all ones that I wouldn't be comfortable calling a winner. Some it's because the fundamentals of the race have shifted enough that it's hard to tell how many crossover votes might happen (the Cunningham scandal complicating North Carolina, Bob Good's indiscretions doing the same in VA-5), whereas others it's a shifting electorate that's hard to guess (New York's 11th district, for example, is pro-Trump, but the Democrats are running a robust campaign, and Staten Island runs from just a little Republican to very Republican depending on the cycle). And of course, there's races like Iowa Senate and Ohio President that just look super close in polling. I don't believe in predicting Tossups in my guide (because regardless of who wins, "Tossup" is not who will be the victor), but I will say I have no idea whom I'll predict in any of these contests at this juncture, and will be looking pretty heartily at all of them in the next week to decide which way the wind is blowing.
No comments:
Post a Comment