In this week’s politics chat, we go looking for surprises. The transcript below has been lightly edited.

micah (Micah Cohen, politics editor): Welcome to a special, extra politics chat — our last before the election. (It’s been fun :crying_cat_face:.)

Today we’re talking X factors!

clare.malone (Clare Malone, senior political writer): Isn’t that a reality show?

natesilver (Nate Silver, editor in chief): I think that’s “Ex-factors,” Clare, which would be a good name for a dating show — like, two couples date one another’s exes or something like that.

clare.malone: In this chat, we’ll discuss all our former flames!

micah: That’s a terrible idea. But, what else should we be looking at, besides the polls, that could affect who wins on Tuesday?

harry (Harry Enten, senior political writer): Well, I think the early vote, obviously. Our model doesn’t take that into account. And while it’s tough to discern trends based on the early vote in most states, it looks good for Hillary Clinton in Nevada, for example, even though the polls in that state paint a far closer race.

micah: But beyond Nevada can we really say anything? African-American turnout for Clinton seems weak, maybe?

natesilver: We could really use a voter-list-based poll of Nevada, indeed. And one of Michigan. Those are the two states where I think we have the least idea what’s going on, and they’ve both had high polling errors in the past.

clare.malone: So those states in and of themselves are X factors?

harry: Right. The question I wonder about re: Nevada is if the polls are off there is it a sign they are too low on Clinton in other states? Or is it a one-off?

natesilver: My general view of the early voting numbers is that they basically tell us the same story as the polls. Clinton’s coalition has shifted more to Hispanics and upscale whites, but with probable underperformance relative to President Obama among African-Americans and young voters. You can see that in the state polls too.

micah: Have you gotten that sense in your travels, Clare?

clare.malone: I think the Latino community is probably the most interesting of these growing support groups — they haven’t really been very active as a political bloc, we don’t really think of them as having the same sort of institutional political organization as say, black communities, but I think this election could be the beginning of a more politicized Latino community. They’re being given an emotionally galvanizing election, and no matter the result this year, I think the Democrats are going to really start working on their outreach.

Because Latino communities aren’t just growing in the Southwest and California — they’re growing in big Eastern Seaboard areas, they’re working in agriculture in places like North Carolina.

natesilver: It’s also interesting that Democratic early voting numbers seem to be strong again over the past couple days in states like Nevada and Florida after slumping earlier this week — which is also sort of consistent with the enthusiasm swing in the polls.

harry: It’s also consistent with what happened in Nevada in 2012 — a strong finish for the Democrats.

micah: OK, so Latino turnout is an X factor. Early voting is an X factor.

What else?

natesilver: Well, we should probably mention that Donald Trump is very, very lucky that he didn’t pick Chris Christie as his running mate. What with his aides being convicted on all charges in the Bridgegate trial just now.

harry: His bridge to the presidency would be fully shut, if that were the case. (I’ll show myself out.)

natesilver: Harry, remember when you thought Christie was gonna make a comeback in the Republican primary? He was your Marco Rubio.

harry: My “Rubio” crushed your real-life Rubio.

natesilver: WTF? My candidate won Minnesota, DC and Puerto Rico.

clare.malone: Wait, what does that have to do with Election Day X factoring? Are we just bringing up a news event?

natesilver: Obviously, there’s the chance something unexpected could happen. You had a news alert this morning about potential al-Qaida activity. You see people worried about hacking and chaos of various kinds on Election Day. Obviously, we hope something like that doesn’t happen, but people are going to be on high alert.

clare.malone: Well, I hate to say this, but voter misinformation might be something to keep an eye on. The White House and other law enforcement agencies are getting prepared for the possibility of Russian cyber actions.

natesilver: It seems too late for an oppo dump or a major investigative story to break, on the other hand, but who knows?

clare.malone: There’s stuff like this, which ProPublica’s Electionland is keeping an eye on — misinformation directed mostly toward Democratic communities telling voters stuff like “you can vote via text!” Making people think the presidential election is like “American Idol.”

harry: I’d like to think that doesn’t have much of an effect. I’d like to think that ‘cause I’d like to think people aren’t that naive.

micah: There are also four more days for Trump to say something crazy (or, you know, for people to remember the thousands of crazy things he’s already said).

clare.malone: He hasn’t said anything crazy lately, you’re right. That’s pretty crazy.

natesilver: Yeah, I think an underrated story is that he’s managed to remain relatively composed since the final debate. That has helped the media to stay more focused on Clinton.

harry: I think it’s the story, no? He’s allowed Clinton’s own faults to be spotlighted without interruption.

micah: I find it pretty amazing that in the last week of this campaign, the dominant story may end up being about Clinton.

What about GOTV?

clare.malone: Is that a public access station? KIDDING!

harry: If get out the vote means anything, then obviously Clinton is much better prepared than Trump.

clare.malone: I think the Republican side is the more interesting one on this particular angle. If they can get low propensity voters to actually turn out in certain states, that would be interesting.

micah: Trump isn’t really doing much GOTV, but the Republican National Committee is.

clare.malone: Per that Bloomberg piece, the Trump team’s strategy is voter suppression of Democratic groups more than GOTV.

But from that Washington Post article Micah just mentioned: “The late field surge gives the party a larger footprint on the ground than it had at any point in 2012.”

natesilver: I’m skeptical that any of that is going to matter much. They’re relying on firms like Cambridge Analytica that, if we can be frank, may be more pseudo-science than data science.

harry: Didn’t Ted Cruz’s campaign pretty much think Cambridge Analytica did nothing for them?

clare.malone: I think the RNC has, to a certain extent, tried to learn its lessons from 2012 — they just were confounded by Trump’s attitude toward traditional campaigning.

Yes, the Cruz people spent a lot of money and were disappointed. But sort of quietly.

harry: Of course, Republican voters are older and vote more regularly (see the difference in turnout between midterm and presidential elections), so Republicans don’t need as big of a ground game.

natesilver: The thing is, there aren’t any magic tricks to GOTV. It just requires a lot of effort and a lot of people and a lot of money over a long period of time.

harry: This is a presidential general election. People usually don’t need motivation to vote. If GOTV moves things, it’ll just be a couple percentage points at most. That can mean something in a close race, but we shouldn’t expect something shocking.

micah: Other X factors? Clinton’s surrogate advantage?

clare.malone: Obama has been on fire. That guy is really campaigning out there!

harry: He has been. It’s quite the difference between having Obama out on the campaign trail versus Rudy Giuliani.

micah: Are you implying that Giuliani isn’t an effective surrogate?


harry: I’m implying Giuliani hasn’t held office in 15 years, and it’s amazing to me that he’s still relevant at all.

clare.malone: Trump’s surrogates are actually spectacularly inept, so it’s a testament to his candidacy’s power that it can withstand them. Ben Carson, in particular, is … weak.

micah: Are we missing any important X factors?

harry: I mean the polling stuff like “shy Trump” voters.

micah: Shy Trump! We should tackle that.

clare.malone: Well, if we are being literal, maybe there is a Marla Maples/Ivana Trump bombshell left in this campaign.

Shy Trump is more pertinent, though.

harry: In the primary, Trump mostly hit or underperformed his polling. And a new Morning Consult study shows there is perhaps the slightest bit of a shy Trump, but it’s really not significant at all.

Still, there’s always the possibility of a polling error.

clare.malone: But we should note that college educated men who sort of left him during the sexual harassment stuff are probably coming back on board with Trump. But they’re not shy about saying they’re coming back. The polls show it.

natesilver: How about “shy Clinton” voters?

harry: Also a possibility.

micah: Who would be a shy Clinton voter?

harry: Married women, I’d think. But could be anyone.

natesilver: Republican married women whose husbands are voting Trump, sure.

People who live in very pro-Trump areas where it’s socially unacceptable to express support for Clinton.

clare.malone: This is why it’s a secret ballot, folks! Interestingly, a lot of people I talked to in Ashtabula County, Ohio, did not want to tell me who they were voting for. That hasn’t happened a lot in this election, and I thought it was really interesting. Don’t know quite what to make of it — but they’re totally right. Secret ballot.

natesilver: I think we ought to have a strong prior belief that the polls are equally likely to be wrong in either direction. It’s very hard to predict the direction in which the polls might be biased.

If anything, it seems to go in the opposite direction of what the conventional wisdom expects. Although this year the conventional wisdom is a bit split between “Clinton ground game” on the one hand and “shy Trump” on the other.

micah: So polls could be off in Trump’s favor (he wins), they could be right (Clinton wins) or they could be off in Clinton’s favor (Clinton wins). She wins two of three — nicely matching the odds our models give her.

natesilver: Yeah, that’s not the worst way to think about it.

micah: Nate, rein in that effusive praise.

natesilver: My highest praise is “pretty good.” Pretty good is higher than very good.


natesilver: Clinton can afford a *minor* polling error in Trump’s favor, BTW, although as we’ve stressed repeatedly her situation gets iffy in the Electoral College if the race tightens much at all. But that’s why if she ticked back up to a 4-percentage-point lead instead of a 3-point lead at the end of the race here, it would be a pretty big deal. And it would be a huge deal if Trump narrowed it to a 2-point lead.

harry: I’m going to be watching the polls over the last few days because I got a feeling that some of the movement we saw to Trump was non-response bias. We already see our Now-cast, which is more aggressive about chasing down polling trends, moving somewhat back toward Clinton.

micah: Yeah, that seems like one of the big unanswered questions. Internal polls have reportedly been far more stable.

natesilver: I’m also curious whether we’ll see convergence/herding in the big national polls that come out this weekend.

harry: Always happens.

micah: Final thoughts?

natesilver: If Clinton’s up 3.5 points in the average, you’ll probably see some polls ranging from a tie to Clinton +7 or thereabouts. I wonder if a big national pollster would have the guts to publish a Clinton +11 or Trump +3 result at this stage.

harry: Some would.

clare.malone: Stories tend to break over the weekend in this election, so make of that what you will as we head into this final weekend.

Source: FiveThirtyEight

Leave a Reply