Does Kavanaugh help Republican chances mid-way?

In the mid-term elections, with different candidates on each ballot in each state and district, it's rare to see the kind of sudden changes in the cast that we've frequently observed during the elections. 39, presidential election of 2016, for example. Instead, the breeds are more localized. But in recent weeks – during Republican attempts to confirm Brett Kavanaugh at the Supreme Court – have rather resembled a presidential election, where the news was largely nationalized.

As it should be, the polls to which I am accustomed during the presidential years were the subject of the same arguments, with conflicting accounts that agree or disagree with the data. A plausible narrative is that Kavanaugh audiences help to excite Republican voters and reduce the "gap of enthusiasm" with Democrats. Like Nate Cohn from The Upshot points out, you can concoct a credible case that polls have taken place since last Thursday's Senate hearings, comparatively good for Republicans. For example, there are two new polls in North Dakota showing incumbent Democrat Heidi Heitkamp, ​​a double-digit poll, or several polls showing a tight race in the Senate in New Jersey, or a poll of the Quinnipiac University showing the generic vote of Democrats. Upshot / Siena College polls show that GOP holders are holding up well in the southwestern Ohio and coastal Virginia districts.

By the same token, if you tried to quote a series of solid Democratic polls since the hearings, you would not have many problems. You can highlight recent polls showing good numbers for West Virginia Democrat Senator Joe Manchin, or several recent polls that revealed that Florida Senator Bill Nelson had taken a small lead, or an Ipsos poll showing that Democrats were broadening their lead over the generic ballot since the hearings, or double-digit in the lead for Democrats in the Upshot / Siena polls in the congressional districts of Arizona and Minnesota. And there are also plausible stories behind this hypothesis. Kavanaugh was not a popular choice to start, and he became even more unpopular in some polls (but not all). It is also particularly unpopular with groups such as university-educated women who typically present themselves at high rates at mid-term.

The goal of the different averages and surveys we build at FiveThirtyEight is to avoid this type of selective selection of surveys; instead, we want to evaluate all complete data. So, for the rest of this article, I'll quickly go through what our different indicators have said at several key moments in Kavanaugh's confirmation process, including when he was appointed in July, before and after the initial hearings of the Senate last month. , when a Washington Post article revealed the name of Christine Blasey Ford (who accused Kavanaugh of sexually assaulting her in high school) on Sept. 16 and before and after last week's hearings with Ford and Kavanaugh.

First, here are the figures from our Congressional Generic Poll Vote – and two versions of our average Trump approval rating, one using all polls (our default version) and one using only registered voters' polls. and probable.

At 35,000 feet above sea level, the generic ballot number is largely synonymous with stability. If you want to be more specific, however, the trend of the generic vote now depends on when you compare. The current GOP deficit on the generic ballot, 8.0 percentage points, is a little worse than before Kavanaugh's nomination, when it was 7.4 percentage points. That's slightly better than when Ford's name was leaked, but when it was at 9.1 percentage points, or since just before last week's hearings, when it was at 8.6 percentage points.

Trump's approval ratings largely followed the same trajectory as the generic ballot, having collapsed from early to mid-September and rebounded slightly since then. It is unclear to what extent this is related to Kavanaugh, as the President was informed of many other news in August and early September, such as the guilty pleas of Paul Manafort and Michael Cohen. The mere fact of not making headlines while Kavanaugh was the main story may have helped Trump's number to return to the average. It is also unclear whether Trump's figures have improved since last week's Senate hearings; In the version of our average, all the polls, they improved a little, but in the version with a registered vote, they slightly deteriorated. (The poll has not yet had time to reflect the impact of Trump's impact on Ford taunts at a rally on Tuesday night.)

We can also consult the figures in the different versions of the House and Senate forecasts, which include state and district polls, as well as national indicators. Unfortunately, we can not go back to the date of Kavanaugh's appointment in July because we did not publish our numbers at that time. But we can cover the period since Kavanaugh's first hearings in early September. First, here is the house:

In the Classic and Deluxe versions of our House forecasts, Republican figures have returned to their level in early September, with about 25% chance (1 in 4) of keeping the House. However, they are a little better than mid-September, when their chances were only 17% (about 1 in 6) in the classic version of our model. They are also a little better than before last week's session, when they were around 20% (1 of 5).

The Lite version of our forecast, which has a strong focus on district polls, tells a somewhat different story. According to Lite forecasts, the chances of the Republican home are a little better than they were last week. However, they are worse than they were a month ago: they went from 34% to 29%. This means that district-level surveys have generally deteriorated for Republicans, even though national indicators have stabilized or improved slightly.

But there is very good news for the Republicans in our Senate forecast:

Republicans have been favored to keep the Senate from the beginning. But their position has improved a little over the last week in all three versions of our model. In our estimates for the classic Senate, for example, Republicans are now 77% favorites, up from 68% before last week's hearings.

Heitkamp and North Dakota, where Republican Kevin Cramer is now a two-on-one favorite, despite the traditionally strong performance of representatives in opposition party square during potential wave elections. Heitkamp's problems may well be related to Kavanaugh – she has not yet specified how she will vote, but polls clearly indicate that the majority of North Dakota residents support Kavanaugh's confirmation. At the same time, Manchin's numbers held up well in West Virginia, even though he had taken an equally ambiguous stance against Kavanaugh, and some Democrats who said they would vote. against Kavanaugh, like Claire McCaskill of Missouri, and Joe Donnelly of Indiana, have achieved satisfactory results lately. I tend to think obviously that Kavanaugh hurt Heitkamp, ​​but it's not totally incisive.

The forecast will not tell you if the changes in the numbers in the hierarchy reflect changes in voter preferences or changes in the propensity to vote. If the position of the GOP on the generic ballot has improved by half a point over the last week, for example, this could be due to the fact that their constituents are filling the gap in the vote. Enthusiasm or for some other reason.

But one index comes from surveys that publish both the results of registered and probable voters; the difference between these numbers is a good measure of the gap of enthusiasm or participation. At present, we show that, according to polls, voter turnout on voters is only 0.4% higher than registered voters. This is far less than the typical gap between the polls of likely voters and those of registered voters, which generally favors Republicans by 1 to 6 percentage points at the mid-point, reflecting the fact that Democrats tend to rely on minority and young voters who do not always show up mid-term. This is, however, a slight improvement for Republicans compared to what we saw earlier this year, when there was no difference on average between the polls of registered and probable voters. To complicate matters, Republicans usually do worst in district polls you would expect it from what they do in general voting, even though district polls are almost always among likely voters. A possibility Kavanaugh helps with the Republican-based voter turnout, but also hurts the GOP among swing voters with a strong propensity to show up, such as suburban women.

Overall, I tend to conclude there is actually something for Republicans – that their position has actually improved compared to that of a week ago (although, not necessarily in relation to a month since). But I'm also wary of the idea that it's necessarily a turning point, because it would not take much – some good generic polls for Democrats, plus a handful of good state-level results in places like North Dakota – to reverse GOP gains in our forecasts. It is true that Republicans have had a decent voting week, but this can also be overstated by selection data that correspond to a particular story.

Finally, it goes without saying that this is still a dynamic situation and that the party that "wins" the battle for Kavanaugh will not necessarily benefit the electorate. The opposite could be true. A CBS News / YouGov poll conducted earlier this week revealed that more voters would be angry than enthusiasts if Kavanaugh was confirmed – but also that more voters would be angry than at home. enthusiastic if Kavanaugh was do not confirmed. Whatever party does not arrive in Kavanaugh, there will be more reason to feel aggrieved – and perhaps more motivated to vote.