NYT LIVE POLL THREAD 2

[psychprofessor]

Stauber +30, Radinovich under 30...that's quite the Kavanaugh bump in MN-08. Not too many angry women outside of Duluth!

[morgieb]

Stauber +30, Radinovich under 30...that's quite the Kavanaugh bump in MN-08. Not too many angry women outside of Duluth!

The Trump/Republican approval numbers seem wayyyyyyy too high for a seat Obama won twice.

[BudgieForce]

Stauber +30, Radinovich under 30...that's quite the Kavanaugh bump in MN-08. Not too many angry women outside of Duluth!

The Trump/Republican approval numbers seem wayyyyyyy too high for a seat Obama won twice.

They seem way too high even when you factor in Trump won this seat by 16 points.

It could be what Ebsy and xingkerui said and that the call center is reaching out to rural areas first. Because if not, it would be a very very strange result for a district that elected a democrat in 2016.

[Senator Incitatus]

The PA-1 (so far) and NJ-3 polls are real outliers in both races. (Obviously doesn't preclude them being accurate.) Weird given their proximity.

[Ebsy]

Looking closer at the calls made, doesn't actually seem that unbalanced between the Iron Range/the rest of the distrct. More likely that the Upshot can't poll and that this whole project should be abandoned.

[Comrade Funk]

The PA-1 (so far) and NJ-3 polls are real outliers in both races. (Obviously doesn't preclude them being accurate.) Weird given their proximity.

Leads me to believe it's a trend. GOP will be smashed in Northeast suburbs.

[BudgieForce]

Looking closer at the calls made, doesn't actually seem that unbalanced between the Iron Range/the rest of the distrct. More likely that the Upshot can't poll and that this whole project should be abandoned.

I wonder if its a different call center than last time.

[Dr. Arch]

Looking closer at the calls made, doesn't actually seem that unbalanced between the Iron Range/the rest of the distrct. More likely that the Upshot can't poll and that this whole project should be abandoned.

There should be a TON more dots in Duluth alone. This sample is incredibly unbalanced in terms of regions in this district.

[BudgieForce]

Arizona-6 has some of the lowest response rates I've seen so far.

[RussFeingoldWasRobbed]

RIP Radinovich…..

[Bidenworth2020]

RIP Radinovich…..

lol

[GeorgiaModerate]

TX-SEN is finished with a final of 51-43 Cruz.  But among "almost certain to vote", it's O'Rourke+3, which makes this race tied for the second highest enthusiasm gap so far.

[GeorgiaModerate]

The current "design effect" for MN-08 is 1.36.  Anything over 1.2 is bad and makes the results suspect, according to Nate Cohn.

[Former Dean Phillips Supporters for Haley (I guess???!?) 👁️]

There is no possible way that Trump's approval rating has actually shifted 32 points in MN-08 since the last time they polled it (Sept 6-9). Of course, Trump's approval may have shifted somewhat, and there is also sampling error, but it will not have shifted anything remotely like that much in such a short period of time.

So, either the sample will eventually balance out and Trump's approval rating will go towards -1 (rather than +32, which is where it is now), or else they are somehow polling entirely different samples/getting entirely different respondents.

If they are somehow polling entirely different samples/getting entirely different respondents, my hypothesis for how that could be is that MN does not have party registration, which means that they can't weight by party registration (which they do in other states).

Compare the self-IDed party crosstab in the earlier MN-08 poll (34% Dem 32% Rep) to what it is currently in the ongoing MN-08 re-poll (23% Dem, 42% Rep), and it is clear you have an issue there. Not that self-IDed party can't change, but it is not going to change that much in the course of just a month.

Again, with luck the sample will even back out, but if not...

If that is the case, then they would seem to be having a serious response bias problem. That would also mess up all the other polls done in Minnesota (MN-02 and MN-03), and NYT/Siena probably would also have the same underlying issue messing up the polls in other states that do not have party registration - including, for example Texas. So if that is the case, then for example the TX-SEN poll could be off. If there is a serious response bias issue, Quinnipiac and other pollsters would also probably be having that issue, I would think.

[DataGuy]

Radinovich is winning Duluth voters 49% to 34% so far. That will not be anywhere near enough to win the district. Duluth is only 13% of the district's total population, and Stauber is winning the remaining 87% by a large double-digit margin.

[Senator Incitatus]

The PA-1 (so far) and NJ-3 polls are real outliers in both races. (Obviously doesn't preclude them being accurate.) Weird given their proximity.

Leads me to believe it's a trend. GOP will be smashed in Northeast suburbs.

I'd buy that for sure, but these are far from the only two polls in the northeastern burbs. Lance +1, Zeldin +8, and Wild +8 don't really fit with Kim and Wallace up double digits against incumbents.

Obviously this poll is early, but I'm just curious to watch it start out like the NJ-03 poll, which Cohn devoted a dozen tweets to explaining and has promised to re-poll. I'll also be interested in the re-poll of NJ-07 to see if this might be a cresting wave or if that's still a true toss-up. (Would like to see a poll of NJ-11, which you think a New York-based organization would prioritize.)

[psychprofessor]

Radinovich is winning Duluth voters 49% to 34% so far. That will not be anywhere near enough to win the district. Duluth is only 13% of the district's total population, and Stauber is winning the remaining 87% by a large double-digit margin.

Stauber’s best demo is 18-29 year olds! Winning!

[Xing]

There are bound to be a few outliers, and while NJ-03 and PA-01 could be too Democratic-friendly, some results like NV-SEN, ME-02, and NE-02 seem too Republican-friendly.

[Senator Incitatus]

There are bound to be a few outliers, and while NJ-03 and PA-01 could be too Democratic-friendly, some results like NV-SEN, ME-02, and NE-02 seem too Republican-friendly.

Agreed, just noting that the Democratic-friendly outliers (if this continues) so far are bordering districts. Seems noteworthy.

[Sestak]

welp. There goes Bredesen.

old thread.

[Sestak]

new thread.

[Holy Unifying Centrist]

I still think there is a slight chance dems can take the senate. Would require Beto to somehow overperform the polling average + probably a Ted Cruz gaffe. I suppose Heitkamp still has a chance of winning, but I think it is slimmer than Beto's at this moment.

[Xing]

Democrats still have a clear path to 50 seats, although they need to run the table to pull it off. 51 seats is just exponentially harder, and basically either requires TX polls to be massively wrong, or for Heitkamp to make an enormous last minute comeback.

[Fmr. Gov. NickG]

In the MN-08, how is it possible that the total only sums to 95%?

[Dr. Arch]

He wanted Kavanaugh? He can have Kavanaugh.