Welcome to FiveThirtyEight’s weekly politics chat. The transcript below has been lightly edited.
sarahf (Sarah Frostenson, politics editor): It certainly has been quite the news cycle, what with five impeachment hearings last week and a Democratic debate, so in light of the Thanksgiving holiday, this week we’re taking a step back to play a good old-fashioned game of buy/sell/hold with PredictIt prop bets (plus some I made up).
We’ll talk where impeachment is headed and our best guesses for what’s happening in the Democratic primary. PredictIt’s prices (given in cents) as of noon Eastern on Tuesday were translated into probabilities. (We know that’s not exactly right, but it’s close enough.)
And in case you forgot how to play buy/sell/hold
- Buy means “I think the chances of this happening are higher than indicated.”
- Sell means “I think they’re lower.”
- Hold means “I’m a coward and am unwilling to take a stand.”
OK, to kick us off. Buy, sell or hold: President Trump will be impeached in his first term. (73 percent).
nrakich (Nathaniel Rakich, elections analyst): The chances of that happening are much higher than 73 percent. Probably like 95 percent. So I guess I’m buying? (This game always confuses me.)
perry (Perry Bacon Jr., senior writer): I would hold. I still think there is some chance Democrats back down — since the polling suggests support for impeachment remains divided along party lines. (So if you’re a Democrat in a district Trump won, being pro-impeachment is probably a controversial view among your constituents.) The House Judiciary Committee is having a hearing exploring the constitutional grounds for impeachment on Dec. 4, though, so it does suggest that Democrats haven’t gone too wobbly on the impeachment plan.
nrakich: Interesting, Perry!
Wouldn’t it look terrible for Democrats to back down at this stage? The polls haven’t changed much since the Ukraine scandal broke — but they’re much better for impeachment than they were before it broke. And impeachment is still more popular than it is unpopular.
geoffrey.skelley (Geoffrey Skelley, elections analyst): I think I’ll buy at that price, given that Democrats need 216 of their 233 members to vote yes (considering vacancies), but as Perry says, there are some Democrats who’ve sounded somewhat less certain recently.
nrakich: Hmm. I grant you that there are some Democratic members who might want to vote “no” on impeachment out of self-interest. But the self-interest for the Democratic Party as a whole is obviously not to say, “Our bad, the president we’ve been railing against as a criminal for three years actually isn’t a criminal after all.”
sarahf: We still need some more polls before we understand the effect of the hearings, but as FiveThirtyEight’s editor-in-chief Nate Silver pointed out, support overall is slowly ticking back upward. Now, that might not last, or it might just be noise, but as he says, it pushes back against this idea that what Democrats have done is wildly unpopular.
So I’d buy, too.
But OK. Buy, sell or hold: The Senate will convict Trump on impeachment in his first term. (13 percent)
perry: Sell. There is just no chance of a removal vote. None. Maybe Trump resigns if somehow impeachment support gets really high, but I don’t see Republicans voting for his removal. I don’t think there is any real chance Trump resigns either.
nrakich: I think the chances that the Senate removes Trump are not literally zero percent, but they aren’t as high as 13 percent. (I’d maybe say 5 percent?) So yeah, sell.
geoffrey.skelley: I would sell at 13 percent. It’s just such a high bar to get 20 Republicans to vote to remove Trump, which would be needed for removal, assuming all Democrats vote in favor. But as FiveThirtyEight contributor Lee Drutman wrote, if the GOP starts opposing Trump, it might be hard to predict in advance. It could happen fast and all at once.
sarahf: Right, and there’s not really any signs of congressional Republicans breaking with Trump at this point, right?
Or in other words, is there anything that would make you not sell at this point?
geoffrey.skelley: I mean, GOP Rep. Will Hurd was one of just 13 Republicans in the House to vote to oppose Trump’s national emergency declaration over the U.S.-Mexico border, but he’s on the Intelligence Committee and looks like a “no” on impeachment. If he’s not breaking, I’m not sure who will.
But if someone like Hurd was more open to impeachment, especially because he’s not running for reelection, I’d take that as a sign that a break with Trump might be possible.
nrakich: But even so, Hurd isn’t your average congressional Republican, and I don’t think there’s any sign of those average Republicans breaking with Trump, which is what would need to happen for removal.
geoffrey.skelley: I don’t have a probability, but buy, sell, hold: Two GOP senators vote to remove Trump from office.
perry: Sell. There could be one senator (Trump skeptic Mitt Romney of Utah) who votes for removal. But I think the more likely number is zero Republicans back removal.
nrakich: I agree. It’s more likely that a Democrat (i.e., Sen. Joe Manchin of West Virginia) votes not to remove than that a Republican votes to remove.
geoffrey.skelley: Oh, I like that take.
nrakich: Removing a president is just such a drastic step. Even if Romney or moderate Sen. Susan Collins disagree with the president, that doesn’t mean they would vote to oust him from office.
perry: I could definitely see Manchin and Sen. Doug Jones of Alabama voting against removal.
nrakich: Jones would be in a tricky place — he is actually not that moderate. But you’d have to think a vote to remove would be political suicide in Alabama.
geoffrey.skelley: Given he’s trying to win reelection in a state Trump will probably carry by 20+ points, voting to remove might make his defeat a done deal.
perry: On the other hand, Jones may know he’s very likely to lose his reelection bid in 2020 anyway, so maybe he votes for removal because that’s his real view. He might assume he has a better chance of being selected as attorney general for a Democratic president in 2021 than winning another term in the Senate from Alabama. (Jones was the U.S. attorney for the Northern District of Alabama from 1997-2001.)
geoffrey.skelley: Maybe a yes vote by Jones is a sign he knows he’s a goner. A no vote signals he’s still in it to win it.
sarahf: But 56 percent of Americans said in our poll with Ipsos that Trump had committed an impeachable offense, and there were some Republicans, too, open to the idea of impeachment. You don’t think if public opinion continues to tick upward, that wouldn’t convince any Republican senator to break?
perry: But in our impeachment tracker, support for impeachment is at 12 percent among Republicans. Why would any Republican in Congress vote for impeachment?
geoffrey.skelley: To Sarah’s point, based on our poll with Ipsos, it looks like about 28 percent of Republicans might be open to impeachment if they come to believe Trump withheld aid or tried to cover it up, compared to the 11 to 12 percent of Republicans supporting impeachment in our tracker. But that still would mean 70 to 75 percent oppose impeachment, and a bunch of GOP senators don’t want to risk Republican voters’ wrath in primaries in 2020 (or in the future).
sarahf: OK, let’s leave impeachment alone for now — looks like there will be plenty more to speculate on soon enough with the first House Judiciary Committee hearing scheduled for Dec. 4 — and move on to Democratic primary wagers.
South Bend, Indiana, Mayor Pete Buttigieg has been on a bit of an upswing lately. Strong performance in the last debate, some good polls in Iowa and New Hampshire, and the betting markets currently have him winning the Iowa caucuses (37 percent). Buy, sell or hold?
nrakich: Ooh, that’s a tough one. I guess I’d hold — I just don’t know what will happen, so I don’t have strong evidence to argue one way or another.
perry: Sell. I think Buttigieg is doing very well in Iowa. Perhaps he is the favorite there. But I would say he has a 30 percent chance to win the state, Warren a 25 percent chance, Biden and Sanders both a 20 percent shot and maybe a 5 percent chance for someone who is not one of those four.
geoffrey.skelley: I’m going to sell only because it seems like there are four candidates truly in the mix in Iowa, and I’m not sure anyone’s chances are above 30 percent out of the maybe 95 percent of outcomes that involve one of them winning.
Perry and I are very much on the same page!
sarahf: That’s interesting, Perry, regarding the order you’ve put the candidates in — PredictIt has it has Buttigieg, Sanders, Warren and then Biden in terms of their odds winning Iowa. So a bit more bullish for Sanders than I think the conventional wisdom holds, but right about where you’d expect Biden given his lackluster poll numbers in the state.
So tell me, since you and Geoff seem to be of the same mind, why do you think Sanders stands a chance of winning Iowa?
geoffrey.skelley: I think it mainly comes down to the fact that the vote could be sufficiently fragmented, and that means Sanders could win with 22 to 23 percent or something.
perry: I’m not really bullish on Sanders. I just think Buttigieg’s chances of winning Iowa are less than 37 percent.
geoffrey.skelley: The Democratic caucuses are also difficult to project because if a voter’s candidate isn’t clearing a certain threshold at their caucus site — usually 15 percent — that means a voter has to support another candidate who has cleared it. This is called the viability threshold, and it means that voters’ second-choice picks will be really important in Iowa, which is one reason Sanders could do so well. He is the backup pick for 13 percent of likely caucus goers, according to a CNN/Des Moines Register/Mediacom Iowa poll earlier this month.
sarahf: On that note, I’m going to go off book and give you all a wager from my ~mind~ (dangerous, I know). Let’s talk about how many candidates win 15 percent or more of the statewide vote in Iowa, which is the mechanism by which candidates win delegates (this is different than the viability threshold Geoff mentioned above, which looks at each caucus site individually — confusing, I know). Since 1992 when the 15 percent threshold was first introduced, there’s never been a primary or caucus (in Iowa or elsewhere) where four or more candidates have earned more than 15 percent of the statewide vote, but there have been many instances where three candidates have cleared it. So, first up — buy, sell, hold: Three candidates win 15 percent of the vote in Iowa this year (45 percent).
nrakich: Yes! I would buy that, Sarah. I think three candidates will win 15 percent or more. In fact, I think there’s a pretty decent chance that four will, which, as you note, would be unprecedented.
perry: I would buy three candidates getting to 15 percent in Iowa, and I think I would buy four candidates getting to 15 percent as well.
sarahf: What would you put those odds at, Nathaniel, for four candidates clearing it?
nrakich: Oh man, totally out of thin air … I’d say there’s like a 75 percent chance that at least three candidates exceed 15 percent, and a 30 percent chance that four candidates do.
sarahf: I don’t know. It’s hard for me to know whether to buy. Every four years as political journalists we go through this whole rigamarole where we salivate over the idea of a brokered convention.
So what makes you all think four candidates is a real possibility this year compared to say, 2016 on the GOP side, when things were also crowded?
nrakich: I mean, who knows if it would lead to a brokered convention. For one thing, a step like this is probably necessary but not sufficient for convention chaos.
But I think 2020 is different from 2016 because the polling shows that four candidates are already at or above 15 percent in Iowa right now — and that’s before the reallocation of all the lower-tier candidates’ supporters.
geoffrey.skelley: I’d argue, though, that we might see more movement before the caucuses that could lead one of the four leading candidates to slump and fall short of 15 percent, leaving us with only three candidates. Right now, that looks like Biden, but maybe it’s actually Warren or one of the others.
sarahf: OK, in the vein of candidates clearing thresholds statewide to win delegates … currently, Michael Bloomberg sits at 12 percent for winning the Democratic nomination, but I’m more interested in wagers on whether he wins a state (any state) on Super Tuesday. Again from my brain — bidders beware. Buy, sell, hold (5 percent).
perry: Buy. I think Bloomberg has the chance to do well. So. Much. Money.
nrakich: But, Perry, trying to buy political office doesn’t always work if voters simply aren’t interested in what you’re selling. That said, I think 5 percent is pretty low, so I’ll buy too.
I just think things are so unpredictable after the first four states.
Maybe Biden wins all four and the primary is effectively over by Super Tuesday, in which case I’d expect Bloomberg to get no traction. But maybe Buttigieg, Warren, Biden and Sanders each wins one early state and it’s a free-for-all!
geoffrey.skelley: I remain very skeptical of Bloomberg’s chances, though I see the case people are making. It’s possible that things are such a mess after the first four states that Bloomberg could swoop in and present himself as the solution to the chaos. But I think it’s far more likely that Bloomberg’s inability to appeal to Democrats keeps him from taking advantage even if that does happen. So I’m going to sell at 5 percent that he wins any Super Tuesday state.
sarahf: It’s just such an odd strategy to me that he’s decided to skip out on the first two early states to make up ground in the other states. You’ve looked at the historical numbers, Geoff — has this ever worked?
geoffrey.skelley: To be fair, there haven’t been that many times when candidates have openly minimized campaigning in the early states — or even skipped Iowa or New Hampshire outright. But it didn’t work for Henry “Scoop” Jackson in 1976, Al Gore in 1988 or Rudy Giuliani in 2008.
sarahf: I rest my case.
nrakich: I also would be more convinced of this if Bloomberg had said he’s going to go all out to win, say, Virginia.
But spreading his efforts across multiple Super Tuesday states might make it more likely that he comes in first place in none of them.
Now, you need to win more than just one Super Tuesday state to win the nomination, so I get why he’s not doing that. But come on, Bloomberg, why won’t you change your campaign strategy to go along with our prop bet???
sarahf: Haha OK, let’s end on where things stand overall in the 2020 primary. Elizabeth Warren has long been a favorite of the betting markets, but Joe Biden is actually in the lead at the moment with a 24 percent shot. So buy, sell or hold: Biden wins the 2020 Democratic nomination?
nrakich: At 24 percent? Buy.
Biden, to me, remains the single most likely Democratic nominee — especially so now that some of the wind has been taken out of Warren’s sails.
It’s still competitive, but to me that’s good enough for Biden to have a 30-40 percent chance.
perry Hold. I’m just confused. He leads nationally but is fairly weak in Iowa and New Hampshire. And I wonder if losing in those states will affect his numbers in subsequent primaries. I just don’t know.
nrakich: My pushback to that, Perry, is that he’s only weak in the first two early states, and there “weak” just means “not in first place.” It would be entirely unsurprising to me if he gains a few points and wins both Iowa and New Hampshire in the end.
geoffrey.skelley: Based on my analysis of early primary polls earlier this year, a candidate polling at about 25 percent nationally in the second half of the year before the primary has around a 25 percent chance of winning the nomination.
I haven’t calculated things yet for the second half of this year — the year isn’t over, after all — but Biden is probably polling somewhere around 25 percent. So I think I would hold at 24 percent based on that historical record.
sarahf: Whoever defined the ground rule of holding as “I’m a coward and am unwilling to take a stand” was shortsighted. Hold confidently, Geoff!