2018 Congressional Generic Ballot/House Polls Megathread 2 - no debates please

[ltomlinson31]

Do we have evidence that Siena has a pro-incumbent bias and not a pro-Democrat bias?

Like I said, their pro-incumbent bias is only true for summer polls. Later in the campaign, their polls are significantly better.

Their October polls of the two races I mentioned earlier had Bishop down 50-45 (he lost 53-45) and Maffei down 52-42 (he lost 58-39).

They also had Hayworth up on Maloney by 13% and 7% in 2012 in NY-18 (Maloney ended up winning by 3%). In NY-24 the same year, both of their polls had Maffei and Buerkle tied (Maffei ended up winning by 5%). Not really sure if they're necessarily pro-incumbent, but there is that info from 2012 (also worth noting that was when the districts were redrawn).

[Strong Candidate]

NY-22:Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

The GCB is R+13. Like I’ve been saying, a generic Republican would be winning easily here, but Claudia Tenney is not a generic Republican. You can see that from the fact that Brindisi is getting twenty-four percent of the Republican vote and is only underwater 31-37 among Republicans in terms of favorability that this is a perfect match-up for Democrats to pick off disaffected Rockefeller Republicans.

Are there really a ton of moderate GOPers in this district?

Yes, these upstate New York congressional districts are full of Republicans of a moderate sort. Don't forget that this district's previous Representative was the only Republican congressmember to endorse Hillary, and that when he retired, his chosen successor only lost the primary by 10 points despite being viewed as an overwhelming underdog. During the 2006/08 waves, huge swathes of upstate with massive Republican registration advantages elected Democratic Representatives. While these voters will support generic Republicans over generic Democrats most of the time, they will support Democrats in the right match-ups.

[Co-Chair Bagel23]

NY-22: Siena, Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

Phew... not changing my house rating then, good result.

[Littlefinger]

NY-22:Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

The GCB is R+13. Like I’ve been saying, a generic Republican would be winning easily here, but Claudia Tenney is not a generic Republican. You can see that from the fact that Brindisi is getting twenty-four percent of the Republican vote and is only underwater 31-37 among Republicans in terms of favorability that this is a perfect match-up for Democrats to pick off disaffected Rockefeller Republicans.

Are there really a ton of moderate GOPers in this district?

Yes, these upstate New York congressional districts are full of Republicans of a moderate sort. Don't forget that this district's previous Representative was the only Republican congressmember to endorse Hillary, and that when he retired, his chosen successor only lost the primary by 10 points despite being viewed as an overwhelming underdog. During the 2006/08 waves, huge swathes of upstate with massive Republican registration advantages elected Democratic Representatives. While these voters will support generic Republicans over generic Democrats most of the time, they will support Democrats in the right match-ups.

Sample was also heavily elderly skewed with only 27% under the age of 55. Great news for Brindisi to be up in a sample like that. Tilt D.

[Fmr. Gov. NickG]

NY-22: Siena, Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

It is Sienna, so it does have a pro-incumbency bias, but these are still great numbers, and the undecideds in the poll look like they will lean D.

How does a polling firm have a pro-incumbent bias?  Like, what sort of sampling or weighting procedure would produce this?

[Zaybay]

NY-22: Siena, Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

It is Sienna, so it does have a pro-incumbency bias, but these are still great numbers, and the undecideds in the poll look like they will lean D.

How does a polling firm have a pro-incumbent bias?  Like, what sort of sampling or weighting procedure would produce this?

they tend to give incumbents a better sample. For instance, both this poll and the other NY poll have had a GOP bias of around 3-4 in registration, but when they do D districts, they have a 3-4 D bias.

[Kodak]

NY-22:Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

The GCB is R+13. Like I’ve been saying, a generic Republican would be winning easily here, but Claudia Tenney is not a generic Republican. You can see that from the fact that Brindisi is getting twenty-four percent of the Republican vote and is only underwater 31-37 among Republicans in terms of favorability that this is a perfect match-up for Democrats to pick off disaffected Rockefeller Republicans.

Are there really a ton of moderate GOPers in this district?

Upstate NY Republicans are a mix of New England-style moderates, ancestrally Democratic blue-collar workers who voted for Trump, and rural farmers and gun nuts who wish New York City would sink into the ocean. Most of them will happily ticket-split to vote for Cuomo and Schumer. Many of them will do the same here.

[Zaybay]

NY-22:Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

The GCB is R+13. Like I’ve been saying, a generic Republican would be winning easily here, but Claudia Tenney is not a generic Republican. You can see that from the fact that Brindisi is getting twenty-four percent of the Republican vote and is only underwater 31-37 among Republicans in terms of favorability that this is a perfect match-up for Democrats to pick off disaffected Rockefeller Republicans.

Are there really a ton of moderate GOPers in this district?

Upstate NY Republicans are a mix of New England-style moderates, ancestrally Democratic blue-collar workers who voted for Trump, and rural farmers and gun nuts who wish New York City would sink into the ocean. Most of them will happily ticket-split to vote for Cuomo and Schumer. Many of them will do the same here.

Its like the rustbelt combined with moderate GOP suburbs. Its very swingy, as we are probably going to see this election.

[mencken]

Do we have evidence that Siena has a pro-incumbent bias and not a pro-Democrat bias?

Like I said, their pro-incumbent bias is only true for summer polls. Later in the campaign, their polls are significantly better.

Their October polls of the two races I mentioned earlier had Bishop down 50-45 (he lost 53-45) and Maffei down 52-42 (he lost 58-39).

Yes, you did cite that they showed two incumbent Democrats ahead that ended up losing. That did not answer my question at all.

Do we have evidence that Siena has a pro-incumbent bias and not a pro-Democrat bias?

Like I said, their pro-incumbent bias is only true for summer polls. Later in the campaign, their polls are significantly better.

Their October polls of the two races I mentioned earlier had Bishop down 50-45 (he lost 53-45) and Maffei down 52-42 (he lost 58-39).

They also had Hayworth up on Maloney by 13% and 7% in 2012 in NY-18 (Maloney ended up winning by 3%). In NY-24 the same year, both of their polls had Maffei and Buerkle tied (Maffei ended up winning by 5%). Not really sure if they're necessarily pro-incumbent, but there is that info from 2012 (also worth noting that was when the districts were redrawn).

Thanks for at least answering my question. Seems like a Type I error here: most polls overestimated Republicans in 2012 and Democrats in 2014, but you know, anything that supports the "blue wave" narrative against any smidgen of epistemic uncertainty works I suppose.

[mencken]

^exactly no one on this site will be able to tell you with any credibility which way the polls will be off this year. Republicans ended up with a ton of egg on their faces in 2012 while Democrats got that in 2014 and 2016. 2010 was surprsingly a mixed bag in polling errors given Democrats exceeded polling numbers in a lot of Senate/gubernatorial races that year. And in 2006 Republicans did a little better than expected given the final GCB numbers

I never made such a claim. I merely asserted that it was specious to claim that a particular pollster had a pro-incumbent bias because they showed Democratic incumbents doing better in 2014 (and Republican incumbents doing better in 2012). I never heard such a claim before this poll came showing Tenney doing better than expected.

 Republicans might lose only a few hostile open seats and controversial incumbents or they might lose virtually every seat where Trump won by <10%. Most of the hacks here will jump over themselves to ignore the first possibility.

[Florida Man for Crime]

I never made such a claim. I merely asserted that it was specious to claim that a particular pollster had a pro-incumbent bias because they showed Democratic incumbents doing better in 2014 (and Republican incumbents doing better in 2012). I never heard such a claim before this poll came showing Tenney doing better than expected.

 Republicans might lose only a few hostile open seats and controversial incumbents or they might lose virtually every seat where Trump won by <10%. Most of the hacks here will jump over themselves to ignore the first possibility.

You are correct. Polling error is not predictable in advance (in part because pollsters modify their methodology in response to past errors), and anyone who thinks it is has definitely not observed many elections, or else has not learned anything from observing them.

[Fmr. Gov. NickG]

NY-22: Siena, Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

It is Sienna, so it does have a pro-incumbency bias, but these are still great numbers, and the undecideds in the poll look like they will lean D.

How does a polling firm have a pro-incumbent bias?  Like, what sort of sampling or weighting procedure would produce this?

they tend to give incumbents a better sample. For instance, both this poll and the other NY poll have had a GOP bias of around 3-4 in registration, but when they do D districts, they have a 3-4 D bias.

But why would they do this?  Like, why would you use a different sampling or weighting procedure for races with a Democrat versus Republican incumbent?

[Zaybay]

NY-22: Siena, Aug. 20-26, 499 likely voters

Anthony Brindisi (D): 46
Claudia Tenney (R, inc): 44

It is Sienna, so it does have a pro-incumbency bias, but these are still great numbers, and the undecideds in the poll look like they will lean D.

How does a polling firm have a pro-incumbent bias?  Like, what sort of sampling or weighting procedure would produce this?

they tend to give incumbents a better sample. For instance, both this poll and the other NY poll have had a GOP bias of around 3-4 in registration, but when they do D districts, they have a 3-4 D bias.

But why would they do this?  Like, why would you use a different sampling or weighting procedure for races with a Democrat versus Republican incumbent?

From what I gather, they overestimate the incumbent effect. For incumbents, they weirdly think that more Rs/Ds would come out for their incumbent when that doesnt really happen. And its especially bad in the waves of 2014 and 2018.

[Dr Oz Lost Party!]

In this NBC poll of millennials, the the GCB has the GOP garnering paltry support from Blacks, Asians, and Latinx with 8%, 9%, and also 9% respectively for the 2018 midterms.

That is honestly impressive...

https://t.co/jrMBgwf8Ff

[ajc0918]

In this NBC poll of millennials, the the GCB has the GOP garnering paltry support from Blacks, Asians, and Latinx with 8%, 9%, and also 9% respectively for the 2018 midterms.

https://t.co/jrMBgwf8Ff

Devastating.

[GeorgiaModerate]

Siena will drop a NY-19 poll tonight.

[Politician]

In this NBC poll of millennials, the the GCB has the GOP garnering paltry support from Blacks, Asians, and Latinx with 8%, 9%, and also 9% respectively for the 2018 midterms.

That is honestly impressive...

https://t.co/jrMBgwf8Ff

BUT BUT MUH KANYE ENDORSEMENT

MUH #WALKAWAY

[Zaybay]

Siena will drop a NY-19 poll tonight.

Is Siena doing the entire state, or something?

[Beet]

It's okay... Millennials are age 22-37, about a 15 year span in a country with a life expectancy of 78. As long as we hold other age groups, we'll be fine. Don't forget, Frederick III was supposed to be the youthful liberal "new generation" of the Second Reich, as well.

[Devout Centrist]

It's okay... Millennials are age 22-37, about a 15 year span in a country with a life expectancy of 78. As long as we hold other age groups, we'll be fine. Don't forget, Frederick III was supposed to be the youthful liberal "new generation" of the Second Reich, as well.

You don’t get it, do you?

[wesmoorenerd]

Siena will drop a NY-19 poll tonight.

Prediction: Faso +1

Other prediction: Delgado wins in November anyway

[Zaybay]

Siena will drop a NY-19 poll tonight.

Prediction: Faso +1

Other prediction: Delgado wins in November anyway

yeah, based on their other polls, I would say Faso will lead, but Delgado will win anyway.

[ajc0918]

FL-16 - Anzalone Liszt Research Group (Shapiro Internal)

Buchanan (R) - 48%
Shapiro (D) - 44%

https://voteshapiro2018.com/wp-content/uploads/sites/5/2018/08/POLL-SUMMARY-FL-CD-16-AUG-2018-2.pdf

FL-16 is Trump+11.

[Florida Man for Crime]

Siena will drop a NY-19 poll tonight.

Is Siena doing the entire state, or something?

Probably all the competitive races. They're on a roll.

[wesmoorenerd]

FL-16 - Anzalone Liszt Research Group (Shapiro Internal)

Buchanan (R) - 48%
Shapiro (D) - 44%

https://voteshapiro2018.com/wp-content/uploads/sites/5/2018/08/POLL-SUMMARY-FL-CD-16-AUG-2018-2.pdf

FL-16 is Trump+11.

Interesting result. A few weeks ago a poll came out with Shapiro down eleven. I'd say he's down high single digits realistically. Likely R, closer to Lean than Safe.