Nate Silver's Blog, page 80

2018 College Football Predictions

How this works: Our model uses the College Football Playoff (CFP) selection committee’s past behavior and an Elo rating-based system to anticipate how the committee will rank teams and ultimately choose playoff contestants, accounting for factors that include record, strength of schedule, conference championships won and head-to-head results. It also uses ESPN’s Football Power Index (FPI) and the committee’s rankings to forecast teams’ chances of winning. (Before Oct. 30, when the CFP will release its first rankings, the AP Top 25 poll is used instead.) The teams included above either have at least a 0.5 percent chance of making the playoff or are in the top 25 in at least one of the three rankings we use in our model: FPI, the Elo-based rating or CFP/AP. Every forecast update is based on 20,000 simulations of the remaining season. Full methodology »

Design and development by Jay Boice and Rachael Dottle. Statistical model by Nate Silver.

How Our College Football Playoff Predictions Work

References

Associated Press Top 25 poll / College Football Playoff selection committee’s rankings / Elo rating / ESPN’s Football Power Index

The Details

The goal of any statistical model is to represent events in a formal, mathematical way — ideally, with a few relatively simple mathematical functions. Simpler is usually better when it comes to model-building. That doesn’t really work, however, in the case of the College Football Playoff’s selection committee, the group tasked with picking the nation’s four best teams at the end of each season. As you might imagine from a bunch of former coaches and college-administration types, they can sometimes resist the clean logic that an algorithm would love to impose. So while we’ve found that our model can do a reasonably good job of anticipating their decisions, it has to account for the group behaving in somewhat complicated ways.

That’s one of the challenges our College Football Playoff forecast faces, but one of the fun parts, too. Unlike our other prediction models, which only really try to predict the outcomes of games, it also tries to predict the behavior of the humans on the selection committee instead. Here’s a rundown of how we go about doing that.

The key characteristics of the model are that it’s iterative and probabilistic. It’s iterative in that it simulates the rest of the college season one game (and one week) at a time, instead of jumping directly from the current playoff committee standings to national championship chances. And it’s probabilistic in that it aims to account for the considerable uncertainty in the playoff picture, both in terms of how the games will turn out and in how the humans on the selection committee might react to them.

Games are simulated mostly using ESPN’s Football Power Index. We say “mostly” because we’ve also found that giving a little weight to the playoff committee’s weekly rankings of the top 25 teams helps add to the predictions’ accuracy. (We use the Associated Press Top 25 poll as a proxy for the committee’s rankings until the first set of rankings is released in the second half of the season.) Specifically, the model’s game-by-game forecasts are based on a combination of FPI ratings and committee (or AP) rankings — 75 percent on FPI and 25 percent on the rankings.1

In many ways, that’s the simple part. While predicting games isn’t always the easiest endeavor, there’s a science to it that we’ve applied across our many sports interactives over the years. But the next part, the process of predicting the human committee, is unique to our college football model.

After each set of simulated games, our system begins to guess how the committee will handle those results. These predictions account for the potential margin of victory in each game and for the fact that some wins and losses matter more than others. To assist with this part of the process, alongside a separate formula based simply on wins and losses, we use a version of our old friend the Elo rating. In other sports, we use Elo to help predict the games, but in this case, we mainly rely on it to model how college football’s powers that be tend to react to which teams won and how they did it. This special version of Elo is designed to try to mimic the committee’s behavior.

We’ve calculated these Elo ratings back to the 1869 college football season. Between each season, ratings are reverted partly to the mean, to account for roster turnover and so forth. We revert teams to the mean of all teams in their conference, rather than to the mean of all Football Bowl Subdivision teams. Thus, teams from the Power Five conferences2 — especially the SEC — start out with a higher default rating.3 As a consequence of this, our system also gives teams from power conferences more advantages, because that’s how human voters tend to see them.

This conference-centric approach both yields more accurate predictions of game results and better mimics how committee and AP voters rank the teams. For better or worse, teams from non-power conferences (except Notre Dame, that special snowflake among independents) rarely got the benefit of the doubt under the old BCS system, and that’s been the case under the selection committee as well.

Some of the model’s complexity comes in trying to model when the selection committee might choose to break its own seemingly established rules. For example, we discovered in 2014 — when the committee excluded TCU from the playoff even though the Horned Frogs held the No. 3 spot in the committee’s penultimate rankings and won their final game by 52 points — that the committee isn’t always consistent from week to week. Instead, it can re-evaluate the evidence as it goes. For example, if the committee has an 8-0 team ranked behind a 7-1 team, there’s a reasonable chance that the 8-0 team will leapfrog the other in the next set of rankings even if both teams win their next game in equally impressive fashion. That’s because the committee defaults toward looking mostly at wins and losses among power conference teams while putting some emphasis on strength of schedule and less on margin of victory or “game control.”

We’ve had to add other wrinkles to the system over the years. Before the 2015 season, for example, we added a bonus for teams that win their conference championships, since the committee explicitly says that it accounts for conference championships in its rankings (although exactly how much it weights them is difficult to say).4 And late in 2016, we added an adjustment for head-to-head results, another factor that the committee explicitly says it considers. If two teams have roughly equal résumés but one of them won a head-to-head matchup earlier in the season, it’s a reasonably safe bet that the winner will end up ranked higher.

Still, there are no guarantees. Not only do we account for the uncertainty in the results of the games themselves, but we also account for the error in how accurately we can predict the committee’s ratings. Because the potential for error is greater the further you are from the playoff, uncertainty is higher the earlier you are in the regular season. In early October, for example, as many as 15 or 20 teams will still belong in the playoff “conversation.” That number will gradually be whittled down — probably to around five to seven teams before the committee releases its final rankings.

Editor’s note: This article is adapted from previous articles about how our College Football Playoff predictions work.

Model Creator

Nate Silver FiveThirtyEight’s founder and editor in chief. | @NateSilver538

Version History

1.5 Forecast updated for 2018 season.Oct. 4, 2018

1.4 Forecast published for 2017 season; game-by-game forecasts incorporate team rankings, power conferences given a boost, AP poll used before committee releases rankings.Oct. 5, 2017

1.3 Head-to-head results incorporated into model.Dec. 2, 2016

1.2 Forecast published for 2016 season.Nov. 1, 2016

1.1 Forecast published for 2015 season; conference champion bonus added, uncertainty increased.Nov. 3, 2015

1.0 College Football Playoff model first published for the 2014 season.Nov. 21, 2014

Related Articles
Just Win, Baby (And You’ll Probably Make The College Football Playoff)
Should Alabama’s Résumé Have Trumped Ohio State’s Conference Crown?

Is Kavanaugh Helping Republicans’ Midterm Chances?

In midterm elections, with different candidates on the ballot in every state and district, it’s rare to see the sort of sharp, turn-on-a-dime shifts in the polls that we frequently saw during the 2016 presidential election, for example. Instead, races are more localized. But the past few weeks — during Republicans’ attempts to confirm Brett Kavanaugh to the Supreme Court — have felt more like a presidential election, where the news has largely been nationalized.

True to form, there have been some of the same sorts of arguments about the polls that I’m used to in presidential years, with competing narratives that may or may not square with the data. One plausible narrative is that the Kavanaugh hearings are helping to excite Republican voters and reduce the “enthusiasm gap” with Democrats. As The Upshot’s Nate Cohn points out, you can cobble together a credible case that polls since last Thursday’s Senate hearings have been comparatively good for Republicans. You could cite, for example, cite two new North Dakota polls showing Democratic incumbent Sen. Heidi Heitkamp down by double digits, or the several polls showing a close-ish Senate race in New Jersey, or a Quinnipiac University poll showing Democrats’ generic ballot lead down to 7 points from 14 points previously, or Upshot/Siena College polls showing GOP incumbents holding up well in districts in southwestern Ohio and coastal Virginia.

By the same measure, if you were trying to cite a series of strong Democratic polls since the hearings, you wouldn’t have much problem. You could highlight recent polls showing good numbers for West Virginia Democratic Sen. Joe Manchin, or several recent surveys that found Florida Sen. Bill Nelson having taken a small lead, or an Ipsos poll showing Democrats expanding their lead on the generic ballot since the hearings, or double-digit leads for Democrats in Upshot/Siena polls of congressional districts in Arizona and Minnesota. And there are some plausible stories behind this hypothesis, too. Kavanaugh was not a popular pick to begin with, and he has become more unpopular still in some (although not all) polls. He’s also particularly unpopular with groups such as college-educated women who typically turn out at high rates at the midterms.

The whole purpose of the various polling averages and forecasts we construct at FiveThirtyEight is to avoid this sort of cherry-picking of polls; instead, we want to evaluate all of the data in a comprehensive way. So for the rest of this article, I’m going to take a quick look at what our various metrics have said at several key points in Kavanaugh’s confirmation process, including when he was nominated in July, before and after the initial Senate hearings last month, when a Washington Post article revealed the name of Christine Blasey Ford (who accused Kavanaugh of sexually assaulting her when they were in high school) on Sept. 16, and before and after last week’s hearings with Ford and Kavanaugh.

First, here are the numbers from our generic congressional ballot polling average — and from two versions of our Trump approval rating average, one using all polls (our default version) and another using polls of registered and likely voters only.

How key metrics have changed during Kavanaugh’s confirmation process





Trump net approval


Date
Event
GOP gen. ballot margin
polls of RV/LV*
All polls




July 8
Day before Kavanaugh nominated
-7.4
-9.2
-10.8


Sept. 3
Day before initial confirmation hearings
-8.8
-11.9
-14.0


Sept. 15
Day before Ford’s name disclosed
-9.1
-12.6
-13.4


Sept. 26
Day before hearing on Ford allegations
-8.6
-8.6
-11.4


Oct. 4
Today
-8.0
-9.2
-10.7

*Registered voters and likely voters

From a 35,000-foot view, the story in the generic ballot numbers is largely one of stability.5 If you want to be more precise, however, the trend in the generic ballot now depends on what point in time you’re comparing against. The GOP’s current deficit on the generic ballot, 8.0 percentage points, is a bit worse than it was before Kavanaugh was nominated, when it was 7.4 percentage points. It’s slightly better than it was when Ford’s name was disclosed, however, when it was 9.1 percentage points, or since just before last week’s hearings, when it was 8.6 percentage points.

Trump’s approval ratings have largely followed the same trajectory as the generic ballot, having slumped in early-to-mid September and since rebounded slightly. It’s not clear how much of that is Kavanaugh-related, however, as the president was dealing with a lot of other news in August and early September, such as the guilty pleas of Paul Manafort and Michael Cohen. Merely staying out of the headlines while Kavanaugh was the lead story may have helped Trump’s numbers revert to the mean. It’s also not clear if Trump’s numbers have improved since last week’s Senate hearings; in the all-polls version of our average, they’ve gotten a bit better, but in the registered-voter version, they’ve gotten slightly worse. (There hasn’t yet been time for the polls to reflect any impact of Trump having mocked Ford at a rally on Tuesday night.)

We can also look at the topline numbers from the various versions of our House and Senate forecasts, which incorporate state-by-state and district-by-district polling as well as national indicators. Unfortunately, we can’t go back all the way to the date of Kavanaugh’s nomination in July because we weren’t publishing our numbers then. But we can cover the period since Kavanaugh’s initial hearings in early September. First, here’s the House:

How Republican chances in the House have changed since initial Kavanaugh hearings

According to FiveThirtyEight’s 2018 House forecast

Chance of GOP House Control


Date
Event
Lite
Classic
Deluxe




Sept. 3
Day before initial confirmation hearings
34%
22%
27%


Sept. 15
Day before Ford’s name released
26
17
22


Sept. 26
Day before new confirmation hearings
22
20
23


Oct. 4
Today
29
25
28

In the Classic and Deluxe versions of our House forecast, Republicans’ numbers have reverted back to where they were in early September, with around a 25 percent chance (1 in 4) of keeping the House. However, they’re somewhat better than than they were in mid-September, when their chances had slumped to as low as 17 percent (about 1 in 6) in the Classic version of our model. They’re also a bit better than before last week’s hearings, when they were around 20 percent (1 in 5).

The Lite version of our forecast, which heavily emphasizes district-by-district polls, tells a somewhat different story. In the Lite forecast, Republicans’ House odds are a bit better than they were last week. However, they’re worse than they were a month ago, having fallen to 29 percent from 34 percent. What that means is that district-level polls have generally been getting worse for Republicans, even if national indicators have stabilized or improved slightly.

But there’s some pretty darn good news for Republicans in our Senate forecast:

How Republican chances in the Senate have changed since initial Kavanaugh hearings

According to FiveThirtyEight’s 2018 Senate forecast

Chance of GOP Senate Control


Date
Event
Lite
Classic
Deluxe




Sept. 3
Day before initial confirmation hearings
72%
68%
68%


Sept. 15
Day before Ford’s name released
72
68
69


Sept. 26
Day before new confirmation hearings
69
68
69


Oct. 4
Today
76
77
77

Republicans have been favored to keep the Senate all along. But their position has improved quite a bit over the last week in all three versions of our model. In our Classic Senate forecast, for example, Republicans are now 77 percent favorites to hold the chamber, up from 68 percent before last week’s hearings.

A lot of this comes down to Heitkamp and North Dakota, where Republican Kevin Cramer is now a 2-to-1 favorite despite the traditionally strong performance of opposition-party incumbents in potential wave elections. Heitkamp’s problems might well be Kavanaugh related — she hasn’t yet been clear about how she’ll vote, but polls show a clear majority of North Dakotans favoring Kavanaugh’s confirmation. At the same time, Manchin’s numbers have held up well in West Virginia despite his having taken a similarly ambiguous stance on Kavanaugh, and some Democrats who have said they’d vote against Kavanaugh, such as Missouri’s Claire McCaskill and Indiana’s Joe Donnelly, have had decent polling numbers lately. I’m inclined toward the obvious-seeming conclusion that Kavanaugh has hurt Heitkamp, but it isn’t totally cut-and-dried.

One other thing the forecasts won’t tell you is whether changes in the topline numbers reflect shifts in voter preferences or instead changes in voter propensity to turn out. If the GOP position on the generic ballot has improved by half a point over the last week, for instance, that could be because their voters are closing the enthusiasm gap or it could be for some other reason.

But one hint comes from polls that publish both registered- and likely-voter results; the difference between these numbers is a good measure of the enthusiasm gap or turnout gap. Currently, we’re showing that likely voter polls are only about 0.4 percentage better for Republicans than registered-voter polls. That’s much smaller than the typical gap between likely- and registered-voter polls, which usually favors Republicans by anywhere from 1 to 6 percentage points in midterm years, reflecting that Democrats tend to rely on minority and young voters who don’t always turn out at the midterms. It is, however, slightly improved for Republicans from the numbers we were seeing earlier this year, when there wasn’t any gap at all on average between registered- and likely-voter polls. To complicate matters, Republicans are generally doing worse in district-level polls than you’d expect them to do in generic ballot polls, even though district polls are almost always conducted among likely voters. One possibility is that Kavanaugh is helping with Republican base turnout, but also hurting the GOP among swing voters with a high propensity to turn out, such as suburban women.

Overall, I’m inclined to conclude there’s actually something there for Republicans — that their position has genuinely improved from where it was a week ago (although, not necessarily as compared to where it was a month ago). But I’m also wary of the idea that this is necessarily a turning point, since it wouldn’t take much — a couple of good generic ballot polls for Democrats, plus a handful of good state-level results in places like North Dakota — to reverse the GOP gains in our forecast. There is truth in the idea that Republicans have had a decent week of polling, but it can also be exaggerated by cherry-picking data that’s consistent with a particular narrative.

Finally, it should go without saying that this is still a dynamic situation, and it doesn’t necessarily follow that the party that “wins” the battle over Kavanaugh will benefit electorally. The opposite could prove true. A CBS News/YouGov poll conducted earlier this week found that more voters would be angry than enthusiastic if Kavanaugh was confirmed — but also, more voters would be angry than enthusiastic if Kavanaugh was not confirmed. Whichever party doesn’t get its way on Kavanaugh will have more reason to feel aggrieved — and perhaps more motivation to turn out to vote.

Election Update: Here’s How We Decide Who’s Involved In A Scandal

Welcome to our Election Update for Wednesday, Oct. 3!

As of 11:30 a.m. on Wednesday, the Classic version of our forecast model gave Democrats a 7 in 9 chance (77 percent) of winning the House and Republicans a 5 in 7 chance (72 percent) of holding the Senate. Those are some of the best numbers Republicans have gotten in weeks. The forecast has been a bit more volatile lately, but don’t read too much into it — there was a drought of generic-ballot polls last week, so the model was extra sensitive to the new generic-ballot polls added this week.

One race whose outlook changed significantly, though, didn’t change because of generic-ballot polling, but because of scandal: the California 39th Congressional District. Democrat Gil Cisneros pulled into a virtual dead heat with Republican Young Kim after Melissa Fazli recanted her accusation of sexual harassment against Cisneros, which prompted us to remove the scandal penalty our model had been applying to Cisneros’s chances. The scandal variable is new in our forecasts this year, and we’ve fielded a handful of reader questions about how it’s applied, so we figured we’d bring some transparency to the matter in an Election Update.

It can be surprisingly difficult to agree on what is and is not a scandal. But we wanted to take as much personal judgment as possible out of the equation, so we settled on the following definition: A scandal is a credible accusation of wrongdoing according to some legal or ethical standard. In other words, we’re talking about breaking the law or committing a widely agreed-upon moral transgression. (We know that’s a bit ambiguous, but not every scandalous misdeed is against the law — adultery, for example.) What’s more, under our scandal definition, a candidate doesn’t need to be charged with or convicted of a crime; an allegation is enough to affect public opinion. In fairness, it’s possible for a candidate to be exonerated and shed the scandal tag completely — but the bar for that is high. If the allegations can’t be proved because there’s insufficient evidence, that’s not enough to clear a candidate of the scandal tag; the allegations must be demonstrated to be false or be rescinded. (Think of the difference between a prosecutor dropping charges versus a defendant being found not guilty.)1 Of course, even if a scandal has been dismissed, the lingering fallout can continue to drag a candidate down in ways that show up in other areas of our forecast, such as polling or fundraising numbers.

In races where a candidate is involved in a scandal, we apply a scandal penalty as part of our “fundamentals” calculation. For new scandals — by which we mean scandals that became public knowledge since the seat was last contested — the penalty typically averages around 8 percentage points, based on our research into the effects of scandals on past congressional elections.2 Our research has also found, however, that the effects of scandals fade quickly and that they have only a marginal effect on a candidate’s chances if the candidate has won an election in the time since news of the scandal broke. Thus, for scandals that became public knowledge prior to the current election cycle, the scandal penalty is discounted. Specifically, it’s discounted by a factor of 1/(n+1) where n is the number of years since the scandal became public knowledge. For example, a scandal that was disclosed in 2010, eight years ago, will receive only 1/9th of the penalty3 that a new scandal would get.

Note that not every scandal that has been in the public eye for some time is an “old” scandal — it can still be considered “new” if the seat has not been contested in that time. For instance, the corruption scandal surrounding New Jersey Sen. Bob Menendez, which first became public knowledge in 2013, is considered “new” because his Senate seat was last contested in 2012. By contrast, even though Montana Rep. Greg Gianforte body-slammed a reporter just last year, that is not considered a new scandal because a special election was held after the incident. (Gianforte won.) A few candidates have more than one scandal, in which case the scandal penalty is dated from whenever the most recent scandal broke. More rarely, as a scandal develops, it brings out news of a substantially more serious charge than the one that was made in the original allegation (to take a hypothetical example, a candidate who was originally accused of verbal sexual harassment might later be accused of sexual assault). But it takes a lot for the clock to be reset; additional reporting or the revelation of new facts about the scandal does not a new scandal make.

Also to be clear, the scandal tag applies to our “fundamentals” calculation only and does not apply a penalty to the candidate’s polling numbers. In races where there are a lot of polls, our models eventually default to being almost entirely polling-driven. If the scandal has considerably more or considerably less impact than the typical scandal, the polls (and eventually, of course, the election results) are the best way to judge that.

Here’s a complete4 list of every candidate on the 2018 ballot to whom the scandal penalty applies.

The 35 candidates our forecasts penalize for a scandal

Candidates in 2018 races for the U.S. House, U.S. Senate and governorships who have been involved in scandals, by year the scandal developed and whether it happened in the current election cycle

Candidate
Race
Year
new?




Rod Blum
Iowa 1st
2018
✓


Brandon Brown
South Carolina 4th
2018
✓


Randy Bryce
Wisconsin 1st
2018
✓


Tony Cárdenas
California 29th
2018
✓


Steve Foster
Georgia 14th
2018
✓


Jim Jordan
Ohio 4th
2018
✓


Kris Kobach
Kansas (Gov.)
2018
✓


Adam Laxalt
Nevada (Gov.)
2018
✓


Marty Nothstein
Pennsylvania 7th
2018
✓


Archie Parnell
South Carolina 5th
2018
✓


Bill Schuette
Michigan (Gov.)
2018
✓


Scott Taylor
Virginia 2nd
2018
✓


Steve Von Loor
North Carolina 4th
2018
✓


Chris Collins
New York 27th
2017
✓


Greg Gianforte
Montana at large
2017*



Duncan Hunter
California 50th
2017
✓


Omar Navarro
California 43rd
2017
✓


Dana Rohrabacher
California 48th
2017
✓


Antonio Sabato
California 26th
2017
✓


David Schweikert
Arizona 6th
2017
✓


Bobby Scott
Virginia 3rd
2017
✓


James Comer
Kentucky 1st
2015



Henry McMaster
South Carolina (Gov.)
2015
✓


Mark Meadows
North Carolina 11th
2015



Allan Fung
Rhode Island (Gov.)
2014



Bob Menendez
New Jersey (Sen.)
2013
✓


Scott DesJarlais
Tennessee 4th
2012



Alcee Hastings
Florida 20th
2011



Gavin Newsom
California (Gov.)
2007



David Scott
Georgia 13th
2007



Don Young
Alaska at large
2007



Beto O’Rourke
Texas (Senate)
2005



Ken Calvert
California 42nd
1994



Sherrod Brown
Ohio (Senate)
1989



Tom Carper
Delaware (Senate)
1982

* Although Gianforte’s scandal developed in 2017, which is normally part of the 2018 election cycle, it happened before the last election for his seat — a special election held on May 25, 2017.

Source: News reports

It’s important to note that we limited the scandal ruling to instances where a candidate was accused of crossing a clear line, like breaking a law or breaking his or her marriage vows. That leaves out a number of things I would call “controversies,” like committing a gaffe or saying something offensive. That means it’s also not a scandal when a candidate has an extreme ideology, even if that ideology is widely reviled, like Nazism or white supremacism — ideology remains a matter of opinion, and we can’t be in the business of deciding where to draw the line between what’s extreme and what’s mainstream. Nor is it a scandal when a candidate deviates from social norms on things like bagels and Bigfoot erotica; a candidate’s personal tastes are just that — personal. And it’s not a scandal when a politician does something that’s legal but has bad optics. For example, Rep. Tom Marino lost out on the job of drug czar because he pushed a bill championed by pharmaceutical companies that made it harder for the Drug Enforcement Administration to stop shipments of drugs the agency believed were destined for illegal street sales — but even if the bill, as opponents charged, exacerbated the opioid crisis, it doesn’t meet our definition of scandal because he didn’t break any rules. By the same token, it’s also not a scandal when a candidate has been close to some sketchy activities but he or she hasn’t explicitly been implicated in any wrongdoing. For example, that’s why Tallahassee Mayor Andrew Gillum, the Democratic candidate for Florida governor, won’t carry the scandal variable in our gubernatorial model. The FBI is reportedly investigating actions that Tallahassee’s community redevelopment agency took during Gillum’s time in office, but there is no indication — at least so far — that Gillum himself did anything wrong or even knew about any wrongdoing.

So now that you’ve got the ground rules, do you know of any scandals that aren’t incorporated into our model? Get in touch.

Our Third-ish 2020 Democratic Primary Draft Got Weird

Welcome to FiveThirtyEight’s weekly politics chat. The transcript below has been lightly edited.

sarahf (Sarah Frostenson, politics editor): We’re back with our third snake draft of 2020 Democratic presidential contenders, god help us all. Previous drafts can be found here and here. And remember, we’re trying to pick who’d win the nomination, although our picks tend to get less selective and more inventive(?) as the rounds wear on.

The rules are as follows: Six rounds, so between the four of us, 24 potential 2020 Democratic nominees. Let’s determine the order. (And yes, we really do write our names on slips of paper and pick randomly!) We’re going to have Geoffrey Skelley, our new elections analyst, announce today’s order. Welcome, Geoff!!

geoffrey.skelley (Geoffrey Skelley, elections analyst): The office is currently doing the draw.

Clare has first pick.

Geoff is second.

Sarah is third.

Nate is fourth.

natesilver (Nate Silver, editor in chief): OMG what bullshit.

clare.malone: heh heh

geoffrey.skelley: There are many witnesses.

sarahf: OK, Clare, you’re up first! Take it away.

clare.malone (Clare Malone, senior political writer): Elizabeth Warren.

natesilver: Dammit.

geoffrey.skelley: Wow, that was my pick.

Stunning, I know.

clare.malone: My picks are gonna haunt your dreams, Nate.

sarahf: So she’s been popular in previous drafts, why is she your No. 1 now, Clare?

micah (Micah Cohen, managing editor): I’m really pissed I’m not participating in this draft.

clare.malone: Warren has captured the mood of the party for a long time as far as economic angst goes, she’s been a consistent and eager Trump antagonizer, and she’s gonna be raising mucho $$$$$$$$$$$

natesilver: She’s also seemed more candidate-y recently — like her weekend speech about Kavanaugh, which was nominally a speech for her Senate race, felt very much like something she could deliver in Iowa or New Hampshire.

clare.malone: Right, that speech is basically the reason we’re having this chat.

sarahf: For any readers that missed it, Warren said in a town hall this weekend in Holyoke, Massachusetts, that “after Nov. 6, I will take a hard look at running for president.” So yeah, definitely she seems like she’s considering running.

OK, Geoff, who’s your pick?

geoffrey.skelley: I’m going to go with Kamala Harris.

Tough call, was debating leading with a certain former vice president. But when I think about candidates who can put together winning coalitions, I think of candidates who could have a strong appeal to the Democratic Party’s African-American base.

sarahf: Betting markets seem to agree with you, Geoff.

geoffrey.skelley: Harris is also fresh and Democrats may be poised to go for a woman nominee again. Plus, Harris will have access to that California $$$$.

sarahf: And I’m going to continue the #2018yearofthewoman with my pick … Kirsten Gillibrand.

clare.malone: So, Sarah, a question for you on that one: Worried at all about the way that she has been screwed by some in the donor class?

sarahf: For sure. I also think her ties to Clinton are problematic for a 2020 run.

But I think she has a lot of experience going for her. She’s been in the Senate since 2009 and was in the House before that. Plus, she has some bipartisan appeal as well. Part of what we saw in 2016 I think had to do with the fact that both Trump and Clinton were deeply unpopular, which means I don’t think Clinton’s loss necessarily means that a woman like Warren, Harris or Gillibrand can’t win.

geoffrey.skelley: Gillibrand is probably the leading NY candidate, which ain’t nothing in a Democratic field.

sarahf: Nate, you’re up.

clare.malone: Nate’s gonna go with noted populist Democrat Jamie Dimon, I can feel it.

natesilver: OK, we’re going snake so I get two picks, right?

geoffrey.skelley: Yeah.

sarahf:

Politics Podcast: What The New Polling On Kavanaugh Means

FiveThirtyEight












 












More: Apple Podcasts |
ESPN App |
RSS
| Embed


Embed Code

The first batch of polls is out in the wake of Judge Brett Kavanaugh’s testimony last Thursday. The FiveThirtyEight Politics podcast crew discusses how the public is reacting. The crew also looks at how faith in the court has evolved in recent decades and debate whether a Kavanaugh confirmation would affect that evolution.

You can listen to the episode by clicking the “play” button in the audio player above or by downloading it in iTunes , the ESPN App or your favorite podcast platform. If you are new to podcasts, learn how to listen .

The FiveThirtyEight Politics podcast publishes Monday evenings, with occasional special episodes throughout the week. Help new listeners discover the show by leaving us a rating and review on iTunes . Have a comment, question or suggestion for “good polling vs. bad polling”? Get in touch by email, on Twitter or in the comments.

Emergency Politics Podcast: Flake News!

FiveThirtyEight












 












More: Apple Podcasts |
ESPN App |
RSS
| Embed


Embed Code

In an emergency edition of the FiveThirtyEight Politics podcast, the crew reacts to Sen. Jeff Flake’s call for an FBI investigation into the sexual assault allegations against Supreme Court nominee Brett Kavanaugh. Flake indicated that he is reluctant to vote to confirm Kavanaugh without such an investigation.

You can listen to the episode by clicking the “play” button in the audio player above or by downloading it in iTunes , the ESPN App or your favorite podcast platform. If you are new to podcasts, learn how to listen .

The FiveThirtyEight Politics podcast publishes Monday evenings, with occasional special episodes throughout the week. Help new listeners discover the show by leaving us a rating and review on iTunes . Have a comment, question or suggestion for “good polling vs. bad polling”? Get in touch by email, on Twitter or in the comments.

Election Update: Why Our Model Thinks Beto O’Rourke Really Has A Chance In Texas

When building a statistical model, you ideally want to find yourself surprised by the data some of the time — just not too often. If you never come up with a result that surprises you, it generally means that you didn’t spend a lot of time actually looking at the data; instead, you just imparted your assumptions onto your analysis and engaged in a fancy form of confirmation bias. If you’re constantly surprised, on the other hand, more often than not that means your model is buggy or you don’t know the field well enough; a lot of the “surprises” are really just mistakes.

So when I build election forecasts for FiveThirtyEight, I’m usually not surprised by the outcomes they spit out — unless they’re so surprising (a Republican winning Washington, D.C.?) that they reflect a coding error I need to fix. But there are exceptions, and one of them came in the U.S. Senate race in Texas between Republican incumbent Ted Cruz and Democratic U.S. Rep. Beto O’Rourke. I was pretty sure that once we introduced non-polling factors into the model — what we call the “fundamentals” — they’d shift our forecast toward Cruz, just as they did for Marsha Blackburn, the Republican candidate in Tennessee. That’s not what happened, however. Instead, although Cruz is narrowly ahead in the polls right now, the fundamentals slightly helped O’Rourke. Our model thinks that Texas “should” be a competitive race and believes the close polling there is no fluke.

We’ll return to Texas in a moment, but first, here’s a table comparing the polls and fundamentals in the five Senate races where elected Republican incumbents are defending their seats. (We covered races with elected Democratic incumbents in Part 1 of this series and open-seat races in Part 2). As you can see, there isn’t really a lot of disagreement between the polls and fundamentals in these races:

Republican incumbents are polling about as well as expected

Forecasted margin of victory for Republican senators who are running for re-election, according to FiveThirtyEight’s fundamentals and adjusted polls as of Sept. 26

Forecasted margin of victory


Race
Incumbent
Fundamentals
Adjusted Polls




Nevada
Heller
-0.5
-0.9


Texas
Cruz
-0.3
+3.8


Nebraska
Fischer
+15.1
+13.0


Mississippi
Wicker
+19.2
+16.6


Wyoming
Barrasso
+40.8
—

The Mississippi special election is not listed because Sen. Cindy Hyde-Smith is an appointed rather than an elected incumbent and our model treats races with appointed incumbents as open-seat races. There has been no polling of the Senate races in Wyoming.

I’m not going to discuss Nebraska, Wyoming or the regular election in Mississippi much further.1 You wouldn’t expect them to be competitive based on the fundamentals, and they haven’t looked competitive when polls have been taken there — although I wouldn’t mind seeing a poll of Nebraska, which hasn’t had a nonpartisan survey all election cycle or any polling at all since January.

I would note, however, that our fundamentals calculation doesn’t expect all Republican incumbents to be in competitive races just as a default — it has Nebraska Sen. Deb Fischer winning by 14 percentage points and Wyoming Sen. John Barrasso winning by 41, for example. Rather, it’s factors particular to Texas that make the model think Cruz is a weaker incumbent and O’Rourke is a stronger challenger than usual. So let’s talk about Texas in more detail, and then we’ll loop back around to the other close race, Nevada.

Before this year, we treated incumbency as just another variable in our fundamentals model. That was a mistake, because there are all sorts of complicated interactions between incumbency and the other variables that go into the fundamentals. To take an obvious example, the margin of victory in a state or district’s previous election is a lot more meaningful when there’s an incumbent running than when there are two new candidates on the ballot.

So this year, we built separate fundamentals models2 for races with elected incumbents, compared with open-seat races. One of the most important differences is that a state or district’s overall partisanship, as measured by voting in elections for president or state legislature,3 is less important in races with incumbents. Something like presidential voting is a very useful indicator when you don’t have a lot of other data to go by. But in races with incumbents, we have a lot of information pertinent to the specific incumbent and his or her strengths. For instance, our fundamentals calculation for Florida’s 26th Congressional District knows that Republican Carlos Curbelo won the district by 12 points in 2016 even as the area voted overwhelmingly for Hillary Clinton for president. That doesn’t necessarily mean that Curbelo will survive the “blue wave” this year. (He’s only a slight favorite for re-election.) But it does mean the presidential vote doesn’t tell you everything you need to know about the district

Cruz, on the other hand, shows signs of being a weak incumbent — and O’Rourke shows signs of being a tough challenger. Here’s a detailed calculation of exactly what goes into the fundamentals model in Texas.

Some factors hurting Cruz have nothing to do with Cruz himself, but rather with the state of Texas. Historically, the incumbency advantage is larger in small, idiosyncratic states and smaller in larger, more diverse ones. This is why Democratic Sen. Mazie Hirono’s incumbency advantage in Hawaii is much larger than Cruz’s in Texas or Sen. Bill Nelson’s in Florida, for example. In addition, Congress’s overall approval rating is low, which hurts incumbents in all states and all parties.

Still, Texas is a red state — redder in statewide elections than in presidential ones, in fact — and Cruz won by a fairly healthy (although by no means overwhelming) margin in 2012. That ought to be enough to offset a blue national environment as measured by the generic congressional ballot. If you add up the first four indicators in the table — incumbency, state partisanship, Cruz’s previous margin of victory and the generic ballot — they’d project him to win by about 9 percentage points.4

It’s the other factors that push the race toward toss-up status, however. When a challenger has previously held an elected office, they tend to perform better with each level higher that office is. To run for Senate, O’Rourke is giving up his seat in the U.S. House, which is a higher office than had been held by Cruz’s 2012 opponent, Paul Sadler, a former state representative. Strong incumbents tend to deter strong challengers from entering the race, but Cruz wasn’t able to do so this time. Cruz also has a very conservative voting record, one that is perhaps “too conservative” even for Texas. The model actually penalizes O’Rourke slightly for his DUI scandal, but because the scandal has been public knowledge for a long time, the model discounts its importance.

Fundraising is another influential factor hurting Cruz. Ordinarily, you’d expect an incumbent to have a pretty healthy fundraising advantage. Instead, O’Rourke had more than doubled Cruz in dollars raised from individual contributors as of the end of the last filing period on June 30 — an advantage that will probably only increase once the campaigns file their next fundraising reports, which will cover up through Sept. 30. (Our model considers money raised from individual contributors only — not PACs, parties or self-funding.) If fundraising were even, Cruz would still lead in our fundamentals calculation by 4 percentage points, but O’Rourke’s money advantage is enough to bring the overall fundamentals forecast to a dead heat.

One could get into some pretty good arguments about exactly how fundraising should be included in the model. Should out-of-state or out-of-district contributions get less weight, for example? Are Republican donors contributing less in the post-Citizens United era because they expect super PACs to fill in the gaps for them? Still, individual fundraising totals have one really nice quality, which is that they represent hard evidence — tangible action undertaken by individual voters. If you thought you could never trust the polls, fundraising might be one of the first things you’d look at instead. And the fundraising numbers have generally been really good for Democrats, in Texas and in other races for Congress, perhaps reflecting their enthusiasm advantage.

Now that you’ve read all those words explaining why the fundamentals look the way they do in Texas, I should probably tell you that they don’t actually have that much influence on our top-line forecast there. That’s because a lot of polling has been done in Texas, and our model doesn’t weigh the fundamentals heavily when it has a lot of polling. Nonetheless, the fundamentals help explain why it isn’t necessarily a surprise that the polling shows a close race in Texas or that O’Rourke has gradually been gaining ground. At the moment, Cruz leads in our adjusted polling average by 3.8 percentage points; adding in the fundamentals pushes the forecasted margin to 3.3 points, a close race.

Let me also show you the detailed fundamentals calculation for Nevada, which is a more straightforward race:

Republican Sen. Dean Heller is a fairly typical incumbent who should have a decent-sized incumbency advantage, and Nevada is a fairly average swing state. He’s drawn an experienced opponent in U.S. Rep. Jacky Rosen, however, who has raised slightly more money than Heller has — and the overall political environment is blue. All of that adds up to a race that “should” be a toss-up, which is exactly what the polls in Nevada show too. Nevada may not be as high-profile a race as Texas, but it’s just as important in determining control of the Senate.

Politics Podcast: What Happens After The Kavanaugh Hearing?

FiveThirtyEight












 












More: Apple Podcasts |
ESPN App |
RSS
| Embed


Embed Code

The FiveThirtyEight Politics podcast team reacts to the Senate Judiciary Committee testimony on Thursday of professor Christine Blasey Ford and Judge Brett Kavanaugh. One of the lessons of the day was how partisanship shapes what is supposed to be one of the least partisan branches of American government.

You can listen to the episode by clicking the “play” button in the audio player above or by downloading it in iTunes , the ESPN App or your favorite podcast platform. If you are new to podcasts, learn how to listen .

The FiveThirtyEight Politics podcast publishes Monday evenings, with occasional special episodes throughout the week. Help new listeners discover the show by leaving us a rating and review on iTunes . Have a comment, question or suggestion for “good polling vs. bad polling”? Get in touch by email, on Twitter or in the comments.

Will The GOP Base Really Stay Home If Kavanaugh Isn’t Confirmed?

Welcome to FiveThirtyEight’s weekly politics chat. The transcript below has been lightly edited.

micah (Micah Cohen, managing editor): As far as we know, Republicans are pushing ahead with Brett Kavanaugh’s nomination to the Supreme Court — for the moment, anyway — amid accusations of sexual assault and misconduct from at least two women. Nate made a semi-compelling argument that that’s a bad move politically. The counter-argument was succinctly expressed by Washington Free Beacon editor-in-chief Matthew Continetti in an interview with Politico:

“A defeated Kavanaugh nomination would not only demoralize the conservative base, it could seriously jeopardize Trump’s relationship with the conservative legal movement, and that could be crippling for conservative influence in the Trump era.”

So, that’s our question for today: If President Trump withdrew Kavanaugh’s nomination, would that depress conservative turnout in the midterms? And somewhat relatedly, would it have a long-term effect on “Trump’s relationship with the conservative legal movement”?

Give me your topline views, and then we can get into it.

Also, welcome, politics editor Sarah Frostenson, to your first politics chat!!!

sarahf (Sarah Frostenson, politics editor): Hello, hello!

natesilver (Nate Silver, editor in chief): My topline view is that:

It might be partially true, but probably also exaggerated by commentators who are extrapolating their own personal investment in seeing Kavanaugh confirmed to feelings among “the voters” or “the base.”
Even if it’s partly true, the alternatives for the GOP could be a lot worse.
The base might be despondent initially, but there’d be an opportunity to rectify that if Trump successfully rolled out another conservative candidate.

micah: Having a topline view with three parts is very on-brand.

perry (Perry Bacon Jr., senior writer): Trump has appointed a huge number of conservative judges to the federal courts, most notably Neil Gorsuch to the Supreme Court. The conservative legal movement will be fine with him as long as he keeps doing that, so replacing Kavanaugh with another conservative would do the trick. And I think there is very limited evidence that conservative voters, as opposed to conservative activists, are motivated by Supreme Court nominations.

sarahf: Damn, I think Perry took some of my topline views.

I would add, though, that given the partisan divide on issues of sexual misconduct, Kavanaugh withdrawing or President Trump asking for him to withdraw (which I think is unlikely) could backfire among the base.

perry: So you think Republican voters will be turned off because they think Trump would be bowing to liberal views on sexual harassment/misconduct/assault.

micah: Yeah, and maybe that plays more of a role in how the GOP base reacts than their views on the Supreme Court.

sarahf: Yes, I think that could happen. They might see him as weak, giving into liberals, etc.

micah: OK, so let’s take each of those one by one …

First, the idea that Republican/conservative voters really care about the Supreme Court — more so than Democratic voters — does have some evidence to back it up, right?

perry: So the exit polls from 2016 would seem to say that. Among voters who said Supreme Court appointments were the most important factor in their vote — about one-fifth of all voters — Trump beat Hillary Clinton 56 percent to 41 percent. Among the 14 percent who said such appointments were “not a factor at all,” Clinton won 55-37.

Still, I think this is a bit of a misnomer. How much did Clinton talk about the Supreme Court? How much did liberal activists prioritize the court during the campaign? I don’t have a study on this, but my guess is less than Republicans. Remember, Trump felt compelled to put out a list of his likely Supreme Court nominees before he was elected. The court is a core part of the GOP’s campaign message.

Also, it’s a lot easier to say, “I’m deciding who to vote for because of the Supreme Court or abortion,” rather than, “because I have negative attitudes about blacks and Latinos,” which we know motivates a lot of conservative voting.

natesilver: The other issue with that exit poll result is that when you ask voters to volunteer a list of which issues are most important to them, the Supreme Court barely registers.

So, sure, when you remind people about the Supreme Court, many of them say it’s important. And Republicans are a bit more inclined to say so than Democrats, which isn’t meaningless. But it isn’t generally a top-of-mind issue.

perry: It is true, though, that if you wanted to accurately describe Trump’s base or the Republican base, white evangelicals would be a big part of the description. But I don’t know if it follows that therefore because white evangelicals are the base, abortion/LGBT issues motivate them, as opposed to taxes, immigration, health care, etc.

natesilver: Or just overall tribal behavior. They’ve clearly forgiven a lot of Trump’s sins because he’s on their “team.”

micah: Let’s posit for the moment that the prioritization of the Supreme Court among rank-and-file GOP voters is overblown — isn’t it still a priority among GOP activists, Sarah? And doesn’t Trump have to worry about that?

Like, I’m not sure if there’s data to back this up, but I feel pretty confident in saying that Republican activists are more engaged on the courts than Democratic activists.

sarahf: He does, but I would counter that he’s already done quite a bit to remake the federal judiciary. The Supreme Court gets all the attention, but if we look at the lower federal courts, we can see that 26 of Trump’s federal appeals court picks have been confirmed (more than any other recent president at this point in their first term).

micah: Yeah, so it’s unlikely they would abandon him over this?

sarahf: Plus, there’s 142 vacancies to fill across the federal judiciary. More than 100 in U.S. District Courts alone!

perry: Micah, it’s a common view that conservative activists care more about the courts than liberal activists, in part because the Federalist Society has an outsized role on the right, and I don’t think the American Constitution Society (the liberal equivalent) has that same role on the left. But the NAACP Legal Defense Fund, for example, is pretty focused on judges and the law, and it is longstanding and influential.

I would say, though, that Trump has been more invested in appointing judges in his first two years in office than Barack Obama was.

I think we can say that confidently.

sarahf: And to Perry’s point, I think this registers with his evangelical base.

natesilver: We’re also having this fight over Kavanaugh against a backdrop where we’re in the midst of what looks like it’s going to be a blue midterm, maybe even a wave year for Democrats. And that tends to color voters’ impressions of everything.

Because the public is skeptical of Trump and the GOP overall, everything they do is going to be taken in a more skeptical light.

So some fights that might be winnable in, say, 2014 might have a lot more downside for Republicans this year.

micah: OK, but Nate, how do you square the exit poll results showing people who cared about the Supreme Court most overwhelmingly backed Trump with the open-ended Gallup question finding hardly anyone thinks of it off the top of their head?

natesilver: If you ask someone about something, of course they’re gonna be more cognizant of it.

It’s kind of like if I ask you, “How important is personal fitness to you?” Most people are going to say it’s “somewhat” or “very” important. But a lot of those “somewhat important” people probably aren’t going to the gym, aren’t really watching their diet, etc.

micah: It’s very important to me!

I play basketball every day!

Oh, you meant in general.

natesilver: I’m not making an accusation about you personally, Micah, but nice #humblebrag.

micah: lol

sarahf: That’s fair criticism, but we are talking about a conservative majority on the highest court in the land. It could be that it really does matter to Republican voters more this time around.

natesilver: But, Sarah, if they don’t get Kavanaugh, they probably get … someone equally conservative. Or maybe more conservative.

Unless they screw it up by spending several weeks fighting this possibly very stupid fight, to the point where they jeopardize the timing of his replacement.

sarahf: Right, and the bench they have to pull from isn’t weak; this is from July:

Yeah, it’s 41 days until the midterms? Democracy moves slow. I think the reluctance of the GOP to move on from Kavanaugh is tied to that.

perry: There is a core of Republican voters who care deeply about, say, limiting abortion and were not happy with the Supreme Court’s gay marriage ruling. I just think people who volunteer for campaigns and knock on doors are pretty politically active and will not make the weird decision of not voting Republican simply because Trump, who has executed lots of GOP policies, isn’t able to get Kavanaugh through. They might be annoyed. But the idea that they will not vote at all seems far-fetched. The people I would worry about if I were Trump or Senate Majority Leader Mitch McConnell are voters who backed Trump in 2016 but often stay home at midterms and might do so again. I doubt that group is following every development in this story.

natesilver: Plus, there’s the chance that Trump doesn’t abide by the script and nominates Thomas Hardiman, who’s more moderate than some of the other names on the short list, or something. But, again, this isn’t about finding a good outcome for the GOP; it’s about cutting their losses.

micah: OK, so let’s talk about Sarah’s point that this could be viewed by GOP voters as Trump caving on a cultural issue more than about the court in particular.

There’s definitely evidence that cultural issues have a lot of sway with voters, no?

natesilver: I’m not sure what you mean by “cultural issues” — but, sure, in some sense cultural issues (especially if we recognize the very, very strong overlap between “culture” and race) is the whole reason Trump was elected.

perry: I’m not sure. But if I were Sens. Ted Cruz or Dean Heller, I would not equivocate on this issue and suggest that I believe Christine Blasey Ford (the woman who says Kavanaugh sexually assaulted her when they were both in high school). Just in terms of politics, I do think the GOP base expects Republican politicians to view sexual misconduct claims skeptically, particularly those made against Republican men.

sarahf: Perry wrote this piece earlier in the week: Republicans largely don’t believe Ford’s accusations are credible.

natesilver: At the same time, #MeToo has also been challenging to a lot of male elites. They clearly care about it a lot. And, again, there’s probably a lot of transference wherein they assign their own feelings to those of the amorphous base.

micah: How does Trump figure into all this?

natesilver: I think the best thing Trump could do for the GOP is to stay the hell out of it.

He’s absolutely the last person you want talking about the credibility of women accusers.

But, yeah, if he invests himself too much in the story, it becomes more painful for Republicans to fail to confirm Kavanaugh.

perry: He was literally criticizing one of the accusers while this chat was happening.

natesilver: I know. And if McConnell is sort of play-acting and ultimately would be just as happy to replace Kavanaugh with someone else, Trump is a big wild card there.

micah: Nate, you and Trump and McConnell are not on the same page.

natesilver: I’m not sure that “Cocaine Mitch” and I aren’t on the same page.

micah: We argued about this on the FiveThirtyEight Politics podcast, but I don’t see how you square Mitch’s actions so far with a strategy of ditching Kavanaugh?

natesilver: There’s a good chance he’s trying to save face.

He may also be trying to expedite the process so that Kavanaugh gets voted up or down (or has to be withdrawn) within a week instead of after several weeks.

I put a lot of emphasis on the fact that McConnell reportedly didn’t want Kavanaugh to be chosen originally.

I give him some credit for just being a good overall strategist who sees a couple of moves ahead.

perry: I understand Nate’s argument, but I found McConnell’s remarks on Friday (in a speech with conservative activists, he pledged that Republicans would “plow right through” the opposition to Kavanaugh and get him confirmed) to be somewhat in tension with that. But maybe the remarks were part of the strategy.

He does think ahead. I agree.

natesilver: Yeah, I will say — I was more convinced of the McConnell-is-bluffing theory before the story got so much more involved with the new accusers and Trump tweeting and everything else.

Maybe now McConnell has to commit to the bit.

sarahf: I dunno. I think that at this point, McConnell is busy on doing what he has to do to get Kavanaugh nominated. He might have had reservations about him initially, but I think all signs point to him pushing this through. This was written before news of other allegations broke, but I haven’t seen any reporting that makes me think this isn’t still the case.

micah: +1

OK, to wrap up: If the consensus here is that the danger to Republicans in the midterms of pulling Kavanaugh is overblown, is the same true of the danger of pushing him through?

In other words, would voters who are currently GOP-leaning or at least open to voting for a Republican candidate react negatively to Kavanaugh getting confirmed? (Obviously Democrats would.)

natesilver: I’m not sure you necessarily encounter a point of diminishing returns. If, I don’t know, 60 percent of women are slated to vote Democratic for Congress, there’s still 40 percent of women who aren’t yet planning to do so.

It’s also always probably true that pundits (and data journalists?) overrate the long-term electoral importance of whatever the story of the day is. Although we’re close enough to Nov. 6 that this may be fairly fresh in voters’ minds.

perry: It is hard for me to see a bad political outcome here for Democrats. Their base is angry about this issue and will likely be more so if Kavanaugh is confirmed. Democratic senators who are up for re-election this year in red states — Joe Donnelly, Joe Manchin and Heidi Heitkamp, for example — know their states better than I do, but it’s difficult for me to see a “no” vote on Kavanaugh hurting them. Cruz has to vote “yes,” as does Heller, but I don’t think they will be eager to defend that vote.

sarahf: The question of undecided voters is an important one. And on the subject of Kavanuagh, more than a quarter of voters say they don’t know enough to have an opinion, according to a Wall Street Journal/NBC News conducted last week. But the latest polls definitely aren’t good news for Kavanaugh (and this was before the New Yorker story). That WSJ/NBC poll found that 38 percent of registered voters oppose the Kavanaugh nomination, up from 29 percent in August. Below is a table from an article we did on Kavanaugh’s declining popularity (including the shift in the WSJ/NBC poll):

Early signs that Kavanaugh’s popularity has dipped

Polls of Brett Kavanaugh’s approval conducted before and after Sept. 16, when The Washington Post published Christine Blasey Ford’s allegations of sexual assault

Net Support



Pollster
Before
AFter
Change




NBC/WSJ
+4
-4
-8


Morning Consult
+5
+1
-4


HuffPost/YouGov
-6
-4
+2


Ipsos
-3
-9
-6

micah: Yeah, the Journal article on the poll said “college-educated women are particularly sour on Mr. Kavanaugh: 49% of them oppose his nomination, while 28% support it.”

natesilver: There’s a decent argument that even though Kavanaugh might be unpopular, he isn’t any more unpopular than Trump or, say, the GOP’s health care bill.

At the same time, there’s this:

You've watched us produce a string of pretty good results for the Democrats over the last week. It's only two polls being released today, but it may be that other polls in the field over the same period might be finding a similar lurch to the left

— Nate Cohn (@Nate_Cohn) September 25, 2018

Lot of particularly bad polls for the GOP this week, especially in upscale suburban-type places.

sarahf: But Gallup found that the Republican Party’s favorability rating among Americans is the highest it’s been in more than seven years?