2022 Generic Ballot / Recruitment / Fundraising / Ratings Megathread

[GeorgiaModerate]

Marist finds D+7 after Dobbs.

https://www.npr.org/2022/06/27/1107733632/poll-majorities-oppose-supreme-courts-abortion-ruling-and-worry-about-other-righ

It seems like the argument that “everybody already knows how they feel about abortion, nothing will change, red tsunami incoming” doesn’t work well when the legal status of abortion changes dramatically and becomes illegal in half of the country.

To recap:

In April (pre-Dobbs decision leak) they had Republicans leading 47-44.  In May (post leak) they had Democrats up 47-42.  This month it's 48-41 D.

Prior to April, they had not done a poll since November so it's hard to compare trends before that (i.e. it's possible April was an outlier).

Survey data

[GeorgiaModerate]

Although the horrendous effects of this decision will be lasting, the public outrage will not be. The ruling itself will likely be forgotten in a few weeks, and then they'll start talking about gas prices again.

This could very likely be like the ACA repeal where the outrage is sustained.

Yes, and the ruling won't be forgotten because there will continue to be women unable to get abortions in red states, along with developments such as additional states either banning abortions or expliclity legalizing them.  I have no doubt that there will be plenty of stories to keep this all fresh in the news.

[GeorgiaModerate]

So you're saying we have a chance?

Yes.  Republicans are still favored to take House control, but it would no longer be a huge upset if the Democrats held on to a tiny majority.  It would have been one prior to Roe being overturned.

[GeorgiaModerate]

https://twitter.com/NateSilver538/status/1542228065123811329

Any guesses?

[GeorgiaModerate]

The GCB question in the Economist/YouGov weekly tracker had an error and has been removed from their data for surveys prior to this week, with the following disclaimer:

The results for the question, "If an election for U.S. Congress were being held today, who would you vote for in the district where you live?" that were previously included in this document are no longer being reported after YouGov discovered a programming error that resulted in some registered voters not being asked the question. Since the people asked were not a representative sample of all registered voters, the data does not meet YouGov’s standards and has been removed.

The problem has been corrected for this week's survey, which shows D 43, R 40.

[GeorgiaModerate]

The GCB question in the Economist/YouGov weekly tracker had an error and has been removed from their data for surveys prior to this week, with the following disclaimer:

The results for the question, "If an election for U.S. Congress were being held today, who would you vote for in the district where you live?" that were previously included in this document are no longer being reported after YouGov discovered a programming error that resulted in some registered voters not being asked the question. Since the people asked were not a representative sample of all registered voters, the data does not meet YouGov’s standards and has been removed.

The problem has been corrected for this week's survey, which shows D 43, R 40.

Would make sense why it randomly shifted to R+5 the past few weeks.

Though 43-40 is still useless though with so many undecideds, ugh.

Yes, it would.  I've dropped a note to 538 suggesting they remove the broken ones from their database.

[GeorgiaModerate]

The GCB question in the Economist/YouGov weekly tracker had an error and has been removed from their data for surveys prior to this week, with the following disclaimer:

The results for the question, "If an election for U.S. Congress were being held today, who would you vote for in the district where you live?" that were previously included in this document are no longer being reported after YouGov discovered a programming error that resulted in some registered voters not being asked the question. Since the people asked were not a representative sample of all registered voters, the data does not meet YouGov’s standards and has been removed.

The problem has been corrected for this week's survey, which shows D 43, R 40.

Would make sense why it randomly shifted to R+5 the past few weeks.

Though 43-40 is still useless though with so many undecideds, ugh.

Yes, it would.  I've dropped a note to 538 suggesting they remove the broken ones from their database.

Would also make sense as we noted last week - there was a huge disparity between the "preferred outcome" questions (D+2) and the GCB question (R+5)

Hopefully 538 drops them.

G. Elliott Morris has a Twitter thread with more details on the methodology error at https://twitter.com/gelliottmorris/status/1544674762332557313.  He notes that the error was present in all surveys from April 30 through last week's.

[GeorgiaModerate]

The GCB question in the Economist/YouGov weekly tracker had an error and has been removed from their data for surveys prior to this week, with the following disclaimer:

The results for the question, "If an election for U.S. Congress were being held today, who would you vote for in the district where you live?" that were previously included in this document are no longer being reported after YouGov discovered a programming error that resulted in some registered voters not being asked the question. Since the people asked were not a representative sample of all registered voters, the data does not meet YouGov’s standards and has been removed.

The problem has been corrected for this week's survey, which shows D 43, R 40.

Would make sense why it randomly shifted to R+5 the past few weeks.

Though 43-40 is still useless though with so many undecideds, ugh.

Yes, it would.  I've dropped a note to 538 suggesting they remove the broken ones from their database.

I got a reply that said they were going to keep the problematic polls in the database for now.

[GeorgiaModerate]

https://twitter.com/stphnfwlr/status/1545096431605891075

[GeorgiaModerate]

From a Democratic strategist, but still interesting:

https://twitter.com/SimonWDC/status/1545389554743627777

[GeorgiaModerate]

This is a typical midterm, Dems will lose big time, the only question is when do the polls move in the GOP favor like they did for 2014.

Before this Spring I would have agreed with this, but things have changed.  A typical midterm features high motivation by the party out of power and lower motivation by the party in power.  The overturning of Roe v Wade, by all indications, is amping up Democratic energy to a competitive level.  This is not to say that the normal midterm advantage for the out party has been wiped out; that is unlikely.  But it's fair to say that it looks like the advantage will be considerably reduced from a typical midterm.

[GeorgiaModerate]

CNN/SSRS has GCB at 46-46 (same poll has Biden 38/62 approval)

https://www.cnn.com/2022/07/19/politics/cnn-poll-midterms/index.html

This is pertinent too:

The overwhelming majority (88%) of voters who approve of Biden plan to vote Democratic in this year’s midterms; among voters who disapprove, 70% plan to vote Republican and 19% to vote Democratic.

This is yet another poll that supports the theory that a chunk of Biden's disapproval is coming from Democrats who will nevertheless support the party in the midterms.

[GeorgiaModerate]

Echelon Insights, July 15-18: D+4 among LV, D+7 among RV.

https://twitter.com/EchelonInsights/status/1549857224461369344

More details in other tweets in the thread.

[GeorgiaModerate]

...
I suppose that's a good poll for Democrats.

My heart can't take many shocks like this.

[GeorgiaModerate]

538 still refusing to remove those wrong YouGov GCB polls

https://twitter.com/gelliottmorris/status/1550469729726365697

Is this tweet suggesting those polls are making the average friendlier for democrats?

No, he's referring to tweets that have noted the 538 average even WITH the bad YouGov surveys included has moved from R+2.3 to R+0.9 recently.  If the bad polls were removed (which they should be) the average would move somewhat further toward the D's (how much, I have no idea).

[GeorgiaModerate]

Watch the polls FAIL like in 2020!

The generic ballot showed a lead for the Democrats just as Youngkin and other fellow GOP members were romping their way through otherwise blue-ish states. I trust them about as much as Alex Jones atm.

Youngkin ran in 2020?

[GeorgiaModerate]

The Economist/YouGov weekly tracker moves to D+6 this week (44/38) after being at D+3 for the last couple weeks.

https://docs.cdn.yougov.com/ldb835m1bk/econTabReport.pdf

[GeorgiaModerate]

https://twitter.com/DecisionDeskHQ/status/1552361352978542592

[GeorgiaModerate]

I’m excited for when the GCB is like D +2 and most people are truly expecting the house to hold. It’ll make for a great meltdown, even better than it would have been before when we were heading for R+7

FWIW, 538 is currently predicting almost no PVI advantage for either party.  See https://projects.fivethirtyeight.com/2022-election-forecast/house/ and scroll down to the section "How the popular vote for the House translates into seats".

[GeorgiaModerate]

The topline number is still good for the GOP, problem is the momentum was continuous for them in the late months of 2010-14 and it does appear Dems are showing signs of late life. GOP is a big big favorite in the House but still probably have like a 20% chance of losing everything.

Other than the Dobbs report what else caused the shift in the momentum. Also, what do you think would shift it back in the GOPs favor?

Dobbs is the main driver, but there's also Trump being unfavorably back in the news due to the J6 hearings, and possibly gas prices coming down quite consistently for the last month-plus.

[GeorgiaModerate]

Given that Rasmussen is terrible, they should still be giving results of like R+5 in a neutral environment. Not only is their GCB close but it keeps narrowing.

There was a point where I think reality was truly R +6 or 7 and November couldn’t come fast enough, but at this point we need time for a change of course. 52-54 dems would not be out of the question which is enough for extremist legislation

It's not uncommon for a party to peak too soon in a cycle.  If the 2016 election had been held a few months earlier, Hillary would have won easily and the D's would likely have held at least one house of Congress.  Similarly, the D's probably peaked too soon in '18 and '20.  Right now it looks like the same thing may be happening to the R's this year -- but there's still plenty of time for that to change.

[GeorgiaModerate]

Given that Rasmussen is terrible, they should still be giving results of like R+5 in a neutral environment. Not only is their GCB close but it keeps narrowing.

There was a point where I think reality was truly R +6 or 7 and November couldn’t come fast enough, but at this point we need time for a change of course. 52-54 dems would not be out of the question which is enough for extremist legislation

It's not uncommon for a party to peak too soon in a cycle.  If the 2016 election had been held a few months earlier, Hillary would have won easily and the D's would likely have held at least one house of Congress.  Similarly, the D's probably peaked too soon in '18 and '20.  Right now it looks like the same thing may be happening to the R's this year -- but there's still plenty of time for that to change.

Did they? The Democrats held a consistently large polling lead over the Republicans leading up to the 2018 midterms. I don't remember any peaking happening.

In very late 2017 and early 2018, the Democrats led the GCB average by 10 points or more, probably due to fallout from the ACA repeal effort.  There was talk of a 60-70 seat gain in the House.  It started declining in the Spring and settled down in the mid-to-upper single digits thereafter.  Not a huge difference, and a flip in control of the House was never in danger, but it's an illustration of the same principle; the Republicans would have liked an election in early 2022, while the Democrats would have liked it in early 2018.

[GeorgiaModerate]

A tied GCB would still yield a GOP House majority (although not an overwhelming one).

538’s model actually believe the Dems are favored to win the majority with a tied popular vote.  I believe their model tends to overrate incumbency across the board, and Dems would probably lose a tied PV, but it’s pretty close.  I think they’d be favored with a D+1.0 PV.

For another viewpoint, G. Elliott Morris said on Twitter not long ago that he thinks the Democrats need a D+2.5 PV for the majority.

[GeorgiaModerate]

I think a small shift toward the Democrats is plausible, but it’s too soon to be concluding that Democrats are now in good shape. I think left-leaning voters becoming somewhat more engaged was inevitable, thus why I never bought WA-SEN being competitive, and I do think there’s an upper limit to how well Republicans can do and that they’ll surely leave at least a few competitive races on the table, but the fundamentals are still really bad for Democrats. The Republican Party hasn’t been popular in its own right in quite some time, but that hasn’t stopped them from being quite successful, since their strategy has always been to paint the Democrats and the left as “dangerous.”

An argument could be made the fundamentals are bad for both parties.

It's finally the Libertarians' opportunity!

[GeorgiaModerate]

Conservatives believe that voters don't care about Dobbs. And what they do care about is Hunter Biden, the Durham probe, and more 2020 election audits and decertification.

By all means, they should focus on those items in their campaigns.