OH-Sen 2022: So you’re telling me there’s a chance

[Pollster]

Wonder if the seat being open gets Tim Ryan to finally pull the trigger on a statewide run.

[Pollster]

How much money will McConnell dump into this primary to stop Jim Jordan?

[Pollster]

He would need to run an absolutely pitch perfect, gaffe-free campaign and hope he draws an opponent who is nuclear-level toxic in the suburbs. Even then, the national environment would need to be favorable, and the Republican primary would need to be more bruising than the Democratic one. A lot needs to go right.

[Pollster]

https://twitter.com/RussWKYC/status/1382797896169447428?s=20

Moreno has notably vowed to not take PAC money.

[Pollster]

This man really doesn't know how to be a candidate (which is admittedly harder than it looks).

[Pollster]

Am increasingly hearing rumblings that Matt Dolan is increasing his polling numbers very aggressively. Nobody outright predicting a win yet, but many predicting last-minute attention and a "Never Trumper" media circus.

[Pollster]

A pollster friend of mine is currently running a rolling tracker (basically means a daily poll of ~200 respondents per day, with the most recent 3 days making up the traditional "full sample" done by pollsters to catch in-the-moment movement in fluid races) of the GOP primary for one of the main Democratic super PAC's and says that Vance and Mandel have plateaued at roughly 20-25% of the vote each, Dolan is increasing by about 1% per day, Gibbons crashing, Timken plateauing at about 10%, and undecided understandably dropping steadily as well. Appears that most undecided are opting for Dolan, though he claims "reason to believe" they will more likely break for Vance. Says the trajectory, if it holds, points to a Vance win with Dolan narrowly topping Mandel for second.

[Pollster]

Dolan's strategy here was incredibly unorthodox but clearly paid off - he waited for all the other candidates to saturate the electorate with their messaging to the point that their spending had diminishing returns (basically every primary voter already knew who they were/their branding and every dollar they spent started to have less impact on the race), and launched his major paid communications at that exact moment to be the main driver of the conversation at the point of highest voter engagement.

Best recent comparison I can think of is Andrew Gillum's 2018 primary win.

I personally would only advise a candidate to do this under very unique circumstances - it's risky for obvious reasons, especially if you don't have access to personal wealth and need to keep donors reassured that you're competitive. Fortunately for Dolan, he is personally wealthy, not reliant on donors, and is running in a race that largely meets those unique circumstances (multiple well-funded candidates splitting the Trump-aligned vote, pre-ordained establishment choice failing to launch, no conventionally "strong" or "rising star" candidate in the field, etc). Entirely possible though that he's laying out a blueprint for anti-Trump Republicans in primaries across the country.

[Pollster]

Dolan overtakes Vance by a single point in the final day of the rolling tracker, though the 3-day average still gives Vance a slight lead.

[Pollster]

Dolan overtakes Vance by a single point in the final day of the rolling tracker, though the 3-day average still gives Vance a slight lead.

what tracker

A pollster friend of mine is currently running a rolling tracker (basically means a daily poll of ~200 respondents per day, with the most recent 3 days making up the traditional "full sample" done by pollsters to catch in-the-moment movement in fluid races) of the GOP primary for one of the main Democratic super PAC's and says that Vance and Mandel have plateaued at roughly 20-25% of the vote each, Dolan is increasing by about 1% per day, Gibbons crashing, Timken plateauing at about 10%, and undecided understandably dropping steadily as well. Appears that most undecided are opting for Dolan, though he claims "reason to believe" they will more likely break for Vance. Says the trajectory, if it holds, points to a Vance win with Dolan narrowly topping Mandel for second.

[Pollster]

Dolan overtakes Vance by a single point in the final day of the rolling tracker, though the 3-day average still gives Vance a slight lead.

Do you guys happen to be tracking the PA race in the same way?

I'm sure somebody is, but I don't know them/they haven't told me if they are.

[Pollster]

Had dinner with a Republican pollster friend (yes, I have them) last night - he is not polling for Vance but relayed that their numbers for a different client show Ryan gaining significant ground and that there is growing concern that the Vance team doesn't understand that they have a real fight on their hands. Majewski in the Toledo seat is apparently in the same boat and has not made any public appearances recently outside of right-wing podcasts and streaming channels - folks in the party are dreading what his fundraising numbers will be later today.

Don't worry, I cheered him up with plenty of the terrible numbers I've seen prettymuch everywhere else in the country.

[Pollster]

Had dinner with a Republican pollster friend (yes, I have them) last night - he is not polling for Vance but relayed that their numbers for a different client show Ryan gaining significant ground and that there is growing concern that the Vance team doesn't understand that they have a real fight on their hands. Majewski in the Toledo seat is apparently in the same boat and has not made any public appearances recently outside of right-wing podcasts and streaming channels - folks in the party are dreading what his fundraising numbers will be later today.

Don't worry, I cheered him up with plenty of the terrible numbers I've seen prettymuch everywhere else in the country.

What else are you seeing? Can you give us any general insights?

One of the more interesting patterns we've started to pick up on is "frontline" Democrats are polling extraordinarily well right now (even in Trump seats) while many backbenchers in Biden +6-12 seats are polling the way you would be expecting frontliners to poll. The takeaway here is clearly that many backbenchers have lower name ID and have established less personal branding/separation from the national party than members in traditionally competitive seats. Most have plenty of money and face similarly undefined Republicans so would still rather be them than their opponents right now, but these are the types of races that can slip away fast.

[Pollster]

Had dinner with a Republican pollster friend (yes, I have them) last night - he is not polling for Vance but relayed that their numbers for a different client show Ryan gaining significant ground and that there is growing concern that the Vance team doesn't understand that they have a real fight on their hands. Majewski in the Toledo seat is apparently in the same boat and has not made any public appearances recently outside of right-wing podcasts and streaming channels - folks in the party are dreading what his fundraising numbers will be later today.

Don't worry, I cheered him up with plenty of the terrible numbers I've seen prettymuch everywhere else in the country.

What else are you seeing? Can you give us any general insights?

One of the more interesting patterns we've started to pick up on is "frontline" Democrats are polling extraordinarily well right now (even in Trump seats) while many backbenchers in Biden +6-12 seats are polling the way you would be expecting frontliners to poll. The takeaway here is clearly that many backbenchers have lower name ID and have established less personal branding/separation from the national party than members in traditionally competitive seats. Most have plenty of money and face similarly undefined Republicans so would still rather be them than their opponents right now, but these are the types of races that can slip away fast.

Interesting. So someone like Elissa Slotkin is likely polling well because they've been forced to defend their brand for a few cycles now, while someone in a traditionally Likely D slot (maybe someone like Jennifer Wexton) is used to not doing as much, and losing a bit of ground?

It would seem as though swing voters/competitive district "moderates" seem to be okay with separating their reps from the national party/Biden right now, while possible voter apathy is higher in bluer districts?

Wexton probably isn't the best example, since she was recently elected and defeated an incumbent in order to do so. I figure we're talking about people like Dina Titus, Sanford Bishop, and Joe Courtney -- longtime incumbents who haven't seen a competitive race in ages but who no longer have the luxury of resting on their laurels.

Yeah - both of you are on the right track. The task for these folks is to raise their favorables/their opponents' unfavorables before their opponent/GOP groups do the reverse. This is especially challenging in media markets that are home to multiple competitive races and where ad space is harder and harder to come by the closer you get to election day.

Vance and apparently Majewski seem to be making that task easier for Ryan and Kaptur.

[Pollster]

This race is quickly turning into Blunt vs. Kander 2.0, except the Republicans this time benefit from neither incumbency nor Presidential year turnout.

[Pollster]

This race is quickly turning into Blunt vs. Kander 2.0, except the Republicans this time benefit from neither incumbency nor Presidential year turnout.

I thought of that analogy today, actually — fun coincidence. In Blunt's case, I don’t think incumbency was a net advantage, however.

Incumbency advantage in the barest sense, basically that it was easier for him to get free media coverage and to use Congressional-sponsored travel around the state as de facto campaign events. These things count!

[Pollster]

What is the precedent for Tim Ryan winning? Would it be Heidi Heitkamp in 2012? Open seat, but demographics and Presidential votes are fatal to Dems? ND at least had a stronger track record of Democratic senators than Ohio, although Ohio has one which is better than zero.

2012 was clearly a better Dem year than 2022 can ever hope to be but ND then was redder than Ohio now.

Maybe Nevada 2006? Party in power wins a Senate race in a midterm in a state trending away from them? Even then, though, Ensign was the incumbent.

[Pollster]

https://twitter.com/SarahLongwell25/status/1565710650378276868

Unless "more than half" is just very opaque word choice here (and it very well may be), this is actually not a great sign. Ryan likely needs to be winning ~70% (if not more) of these voters to hit his win number statewide.

Caveat of course that we don't know the size of the focus group and that qualitative findings should never be extrapolated to the electorate at-large without confirming with quant.

[Pollster]

Unless "more than half" is just very opaque word choice here (and it very well may be), this is actually not a great sign. Ryan likely needs to be winning ~70% (if not more) of these voters to hit his win number statewide.

Is that the target for Brown-Trump voters or specifically women Brown-Trump voters?

Probably both, honestly.

[Pollster]

https://twitter.com/timryan/status/1568585673929621506?s=46&t=fbkdTuJDJKl_Vb...

[Pollster]

Big oppo drop on Marcy Kaptur's opponent in OH-9. Can't remember the last time a candidate for a major office was credibly accused of stolen valor.

https://twitter.com/BrianSlodysko/status/1572630032664625159?s=20&t=mKwdtwp4...

[Pollster]

Big oppo drop on Marcy Kaptur's opponent in OH-9. Can't remember the last time a candidate for a major office was credibly accused of stolen valor.

Really? It feels like it happens pretty much annually. Several of them are currently in office.

Also very hard to call what Majewski is guilty of here (calling himself an "Afghanistan War" veteran when he was deployed on a support mission to Qatar and calling himself a "combat veteran" because the U.S. government designated him one) "stolen valor"; it's the sort of misleading exaggeration that both veterans and politicians are well-known for. Something on this level happens pretty much every time a veteran runs for office.

This feels like something you'd see more often, but the more I think about it I can't point to many specific cases (doesn't mean they don't exist though). Regardless, whenever vet service is attacked the campaign usually refutes it forcefully, whereas Majewski's team practically concedes here. Also have heard that the candidate has made varying and differing statements about his service in different settings (which probably is what sparked the AP looking into) though I'm not too familiar with him or his team - first time candidate with bad handlers probably walked right into this one - so hard to draw a through-line for what exactly he's exaggerated.

[Pollster]

2020 NBC exit polls (this poll):

46% under age 50 (48%). Note this is even less realistic when considering it’s a midterm

39% R / 31%D / 30%I (37% R / 33%D / 26% I / 4.5% not sure). Why would the share of republicans drop in a midterm?

40% conservative / 21% liberal / 39% moderate (37% conservative / 21% liberal / 37% moderate / 5% not sure)

32% white evangelical Christians (26%)

34% under 50k income (36% under 40k income)

Not to mention that the issues don’t track with the recent student debt polls in the state or that NBC had 44% saying abortion should generally be illegal in 2020. Also, conducted by college students?

https://en.wikipedia.org/wiki/Central_limit_theorem

[Pollster]

2020 NBC exit polls (this poll):

46% under age 50 (48%). Note this is even less realistic when considering it’s a midterm

39% R / 31%D / 30%I (37% R / 33%D / 26% I / 4.5% not sure). Why would the share of republicans drop in a midterm?

40% conservative / 21% liberal / 39% moderate (37% conservative / 21% liberal / 37% moderate / 5% not sure)

32% white evangelical Christians (26%)

34% under 50k income (36% under 40k income)

Not to mention that the issues don’t track with the recent student debt polls in the state or that NBC had 44% saying abortion should generally be illegal in 2020. Also, conducted by college students?

https://en.wikipedia.org/wiki/Central_limit_theorem

I am aware of the CLT definition but can you explain it’s applicability here? I’m assuming you mean that certain traits of a sample can be biased while the sample is fine as a whole. But isn’t the whole point of weighting a sample to account for the fact that you think your original sample is inaccurate? I’m not sure that you can look at those numbers from the weighted sample and say they will resemble the Ohio electorate with a straight face.

Weighting is about representation, not measurement. The central limit theorem applies to both. To take party ID as an example, you could credibly argue self-ID Republicans are underrepresented here based on the 2020 exit poll. The central limit theorem would posit that, assuming the sample is not structurally biased, this underrepresentation would be corrected by, potentially, the self-ID independent subsample leaning more Republican than it possibly should. This is a very clean example, and doesn't take into account intersection with other demographics that could also have a measurement error that potentially corrects a representation error, or vice versa. And of course the margin of error complicates all of this as well. This is why "unskewing" polls is a fool's exercise and generally discouraged and why most internal pollsters use subsamples only directionally.

[Pollster]

Interesting, my two follow-up questions would be:

Would this still apply to features that are very strongly associated with voting behavior? For example, if the true proportion of self ID liberal and urban black voters this November in Georgia is 15% (arbitrary number) and a poll reports this group as 12%, is it possible for the sample to contain enough variation to still be accurate while underestimating a group by 3% where everybody in the group has basically a 100% chance of voting democrat? Or the converse, if a certain republican demographic is underrepresented, how much pro-R variation can you get from liberal and urban black voters?

Would you then say that we should be skeptical of polls with “perfect” cross tabs, such as ones with 8 different factors and the proportion looks right for each variable, because they have perhaps massaged their sample too much?

Short answers: yes and yes.

Long answer: for your first question, the margin of error and statistical noise alone prettymuch guarantee that the answer is yes. Do keep in mind that the margin of error on subsamples is greater than the sample overall (sometimes reaching double digits) - that is OK and to be expected - and makes situations like the one you're describing all the more likely. The answer to your second question is yes for mostly the same reasons. No poll is perfect - that is also OK and to be expected - and if one looks like it is, there was probably excessive weighting/data cutting done to make it so, to the point that the design effect is so large that the poll is useless.