2022 Generic Ballot / Recruitment / Fundraising / Ratings Megathread

[wbrocks67]

July: R 230 - D 205 (R+25)
August: R 226 - D 209 (R+17)
September R 223 - D 212 (R+11)

GCB is R+1 here, was R+2 in August, and R+4 in July.

[Chancellor Tanterterg]

I love how everyone just decides that the most recent poll that they’ve seen is the most accurate, even if it’s only a day or two after another poll

Like just look at the reaction to Wisconsin polling

Barnes+2 = Democrats may pull this off, it’s a tossup!

Johnson+1 = Titanium Ron Jon can’t be beaten… Democrats are fools for investing in this race.

That’s a longstanding problem with Atlas political analysis in general tbh.

[Lief 🗽]

I love how everyone just decides that the most recent poll that they’ve seen is the most accurate, even if it’s only a day or two after another poll

Like just look at the reaction to Wisconsin polling

Barnes+2 = Democrats may pull this off, it’s a tossup!

Johnson+1 = Titanium Ron Jon can’t be beaten… Democrats are fools for investing in this race.

ABC/WaPo is a very high quality national poll. Of course people are going to pay more attention to it than the daily slop from Rasmussen or Morning Consult.

[GeorgiaModerate]

I love how everyone just decides that the most recent poll that they’ve seen is the most accurate, even if it’s only a day or two after another poll

Like just look at the reaction to Wisconsin polling

Barnes+2 = Democrats may pull this off, it’s a tossup!

Johnson+1 = Titanium Ron Jon can’t be beaten… Democrats are fools for investing in this race.

ABC/WaPo is a very high quality national poll. Of course people are going to pay more attention to it than the daily slop from Rasmussen or Morning Consult.

They are a very high quality poll, but it's still not wise to put too much faith into any single poll.  Even the best pollsters will have occasional outliers (and ISTR that this poll has had some doozies in the past).

[Frodo]

Voters divided amid intense fight for control of Congress, poll finds

[GeorgiaModerate]

538 has now added the ABC/WaPo and CBS/YouGov polls, and unsurprisingly they've shifted the model slightly toward Republicans.  The Deluxe model now has 69% D Senate, 70% R House, while Classic has 78% D Senate and 71% R House.

[MT Treasurer]

ABC/WaPo is 51-46 R for LVs. Which would be a big change from what the specials have shown, but that’s not impossible.

I can’t emphasize this enough, but I really don’t think it would be so much a "big change" from the specials as an entirely different electorate. Very important distinction, and also why I really caution people against reading this much into special elections.

Fair point, but looking at the last several elections, what has been a better indicator of Election Day: special elections or the generic ballot polling? We can be cautious about special elections, but we need to be even more cautious about polling. ABC’s LV screen isn’t the gold standard.

This LV screen increasing R performance seems at odds with your point that special elections would be more D than the general election. Who is a more likely voter than someone who shows up for a special?

If I had to choose between the three (with the caveat that you shouldn’t have blind faith in any one indicator), I’d trust national polls over special elections and (most) state polls any day, and yes, I’d have said the same thing before that ABC/WaPo poll. Of course it’s possible that national polls are 'off' as well, but the national average is almost certainly going to be off by much less (no more than 2-3 points, so we’re also talking MoE here) than state polls, as has generally been the case in previous cycles as well.

As for your second question, it’s a fascinating dynamic and I’m not yet entirely sure whether there’s a substantial share of Republican voters who only vote in November (or high-profile primary) elections, but I’ve suspected as much even before this cycle. However, if the special elections really are reflecting the composition of the November electorate, then I would agree that the national polls are probably underestimating Democrats and that we’re looking at a Democratic-controlled Congress in 2023.

[Devils30]

July: R 230 - D 205 (R+25)
August: R 226 - D 209 (R+17)
September R 223 - D 212 (R+11)

GCB is R+1 here, was R+2 in August, and R+4 in July.

The sample is also like 43% identifying as conservative while Dems win moderates by 26 and independents. I am kind of glad fewer pollsters are herding but there is so much uncertainty this year. 72% are 45 and over as well...that number is absurd even for a midterm. Split should be closer to 65-35.

[Senator Incitatus]

Huh. I hadn't actually looked at GCB polls recently and was under the impression from the rest of the forum that Democrats were consistently leading. (edit: Maybe that was the case, but I saw that tweet that showed Republicans up small or big in four polls and was shocked.)

[Amenhotep Bakari-Sellers]

Early voting is underway just because Rs are tied in GCB doesn't mean they are gonna win I have said this, numerous times in this pt in time they were R 5 on the GCB in 2010/14 they're not they are tied the Senate is definitely going D even bigger than 50 seats

Why was the R plus 5 on GCB because in 2010/14 voters wanted to repeal Obamacare that's why, it's now a Pandemic

I expect a big Senate majority and Crist Demings, Ryan and Beasley and Barnes to get elected , our Govs and Senate candidates are always stronger than H members


Watch when the polls close DeSantis and Rubio will be underperform and start losing to Demings and Crist, why because of VBM
The Rs need a big lead to withstand Early voting or VBM they don't have one it's tied that's why Molinaro lost

Watch Crist and Demings and Joe Cunningham start leading Rs when the polls close in SC and FL why because of VBM not Same day vote , they're the first contested states up we will get an idea in IN and KY but if D's win those states like MO, IA they will have to come from behind not ahead like Crist and Demings will be ahead in VBM

[wbrocks67]

Huh. I hadn't actually looked at GCB polls recently and was under the impression from the rest of the forum that Democrats were consistently leading. (edit: Maybe that was the case, but I saw that tweet that showed Republicans up small or big in four polls and was shocked.)

Dems have been leading in probably 75-80% of the GCB polls, if not more. The only recent ones that have had an R lead sans Traf/Rass have been the new ABC/Wapo, CBS/YouGov (they've had one all summer) and Data for Progress (they've had one in the summer too). McLaughlin too, but it's McLaughlin

[wbrocks67]

July: R 230 - D 205 (R+25)
August: R 226 - D 209 (R+17)
September R 223 - D 212 (R+11)

GCB is R+1 here, was R+2 in August, and R+4 in July.

The sample is also like 43% identifying as conservative while Dems win moderates by 26 and independents. I am kind of glad fewer pollsters are herding but there is so much uncertainty this year. 72% are 45 and over as well...that number is absurd even for a midterm. Split should be closer to 65-35.

Yeah, a lot of wonky results all over the place. Many pollsters seem to be expecting that Rs are still going to have a turnout edge in November. That is still where the CBS/YouGov and ABC/Wapo are (the former has been edging towards Ds though).

ABC/Wapo, like I said though, is quite the outlier, at least in its LV model. But it also has Trump winning a 2024 rematch, which says to me it's likely a bit too GOP leaning.

[Kahane's Grave Is A Gender-Neutral Bathroom]

R+1, with a 9 point D shift: ITS RED WAVE, R+30 INCOMING

[Amenhotep Bakari-Sellers]

R+1, with a 9 point D shift: ITS RED WAVE, R+30 INCOMING

Yea and Greg LANDSMAN and Cindy Axne are tied or leading and Golden is ahead 51/46 those seats are supposed to go R and they're not, pundits just assume ME 2 is R leaning and it's not the only reason why Trump won it was Susan Collins in 2020 Golden still survived like he will in 22 and Angus King will pull him thru in 24 King is up for reelection and Golden will challenge Collins and win 26

[Person Man]

R+1, with a 9 point D shift: ITS RED WAVE, R+30 INCOMING

Sarcasm is fun!

[Amenhotep Bakari-Sellers]

D's have been leading in GCB this is no difference than the NBC poll that showed a tied race on GCB a few weeks ago 46/46


Guess what Rs were leading in 2010 in repeal of OBAMACARE 48/42 +6

[philly09]

https://twitter.com/CBSNewsPoll/status/1574042616936767494

[Amenhotep Bakari-Sellers]

There is no less enthusiasm among young people voting because Rs were plus 6 on GCB in 2010/2014 this pt in 2010/ 14 and it's tied and they need to be plus 5 on GCB in order to win the Midterms, the fact it's been tied is good news when Gas prices were high Rs were plus 7 on GCB
.
Real clear politics average was 48/42 R why , because Voters wanted to repeal Obamacare that's why Ds lost 60 H seats

Why are users putting so much infersis in Washington Post Rs are underpolling their 2010)14 numbers and it's good news to be tied in a Midterm not bad the outparty is supposed to sweep all of the special election and they went 2/3 only winning VA and TX 34 and losing AK, NY and Cali recall, young voters came out in Cali recall

[philly09]

https://twitter.com/inadorable1/status/1574057937890430984

[wbrocks67]

I mean, what are the turnout rates for 18-29 year olds? 47% may be meh compared to other age groups, but that still feels massively large for that age group comparatively with usual youth turnout?

[GeorgiaModerate]

I mean, what are the turnout rates for 18-29 year olds? 47% may be meh compared to other age groups, but that still feels massively large for that age group comparatively with usual youth turnout?

Comparing midterms, it was 36% in 2018 and only 20% in 2014, according to the Census Bureau.

[David Hume]

I’ve always been very clear about my position in 2018. The polls were correct even though there was a gigantic turnout gap, with many aspects of it (Nevada for example) being unrecognized by pollsters. The polls should have underestimated democrats more given how far they exceeded expectations.

That’s also one case. The average GCB miss is 1.1 in the direction of democrats according to Nate Silver’s graphic, and there’s a logical (though not conclusive) case to be made that the low-propensity vs high-propensity coalitions are becoming wider than ever.

This is a very important point which makes the "but the polls were (largely) accurate in 2018!" argument appear in an altogether different light. It’s possible that the polls were (largely) accurate only because pollsters didn’t account for the turnout gap, i.e. the drop-off in turnout among Trump supporters and non-college-educated voters offset what otherwise would have been a Democratic bias.

The point here is — and we’ve seen this in special elections — that polls are most likely to miss in high-turnout elections.

Would you consider the GA runoff and VA NJ 2021 elections high turnout? The polls were quite accurate for those three.

[Tintrlvr]

I mean, what are the turnout rates for 18-29 year olds? 47% may be meh compared to other age groups, but that still feels massively large for that age group comparatively with usual youth turnout?

Comparing midterms, it was 36% in 2018 and only 20% in 2014, according to the Census Bureau.

That's eligible voters, not registered voters. The 47% presented above is a percentage of registered voters.

[Liminal Trans Girl]

I don't believe that theres any Real R shift in the polls from the most recent bunch

ABC/Wapo R+1 This was a shift towards the Dems IIRC

ABC/Wapo R+5 Since 538 doesn't provide how many people were polled, I'm not inclined to believe this poll

McLaughin R+4 Not only have they not held no polls with 21 days of an election since 2018, which is a big red flag, they only held one poll within 21 days of the election and It was off by 8 points and going back many of their polls held within 21 days of an election have been wildly off. I wouldn't take this one at face value

Rasmussen R+2 Its Rasmussen.

YouGov/CBS is the only one I really trust and it has 9% Undecided

[wbrocks67]

The ABC/Wapo LV model is weird. It wasn't in any of the crosstabs and Wapo didn't even report it, ABC did. And the way they worded it was odd - it was something like "in one of the likely voter models", as if there was multiple