PA GOV 2022: Stick a fork in it

[Amenhotep Bakari-Sellers]

This race has been over when an R internal had both Oz and Mastriano both down

[omar04]

If Mastriano doesn't book air time soon, there very well could be none left for him. TV ad space is not limitless and candidates in the past all over the country have learned this the hard way.

What do you think about the point that GOP operatives have reportedly seen internals showing Shapiro up by as much as a dozen points? Is the RGA thinking of triaging it?

Republican pollsters (and public pollsters) have been very slow to implement the same adjustments that Democratic pollsters have to counter non-response bias among non-college voters (of all races) and overrepresentation of the highly engaged among both party bases. I'm not surprised to hear that Republican internals are showing massive Democratic advantages in the rust belt (where the non-response bias issue is by far the most problematic) and if they cut Mastriano off because of it, music to my ears.

Why have they been slower to adjust? Issues with reaching enough non college voters to weigh properly?

[Pollster]

If Mastriano doesn't book air time soon, there very well could be none left for him. TV ad space is not limitless and candidates in the past all over the country have learned this the hard way.

What do you think about the point that GOP operatives have reportedly seen internals showing Shapiro up by as much as a dozen points? Is the RGA thinking of triaging it?

Republican pollsters (and public pollsters) have been very slow to implement the same adjustments that Democratic pollsters have to counter non-response bias among non-college voters (of all races) and overrepresentation of the highly engaged among both party bases. I'm not surprised to hear that Republican internals are showing massive Democratic advantages in the rust belt (where the non-response bias issue is by far the most problematic) and if they cut Mastriano off because of it, music to my ears.

Why have they been slower to adjust? Issues with reaching enough non college voters to weigh properly?

I really couldn't tell you. We know they are having the same problems and missed just as badly as we did in 2020 in mostly the same races/places. They just haven't prioritized fixing the problem, and I don't know why.

[GeorgiaModerate]

If Mastriano doesn't book air time soon, there very well could be none left for him. TV ad space is not limitless and candidates in the past all over the country have learned this the hard way.

What do you think about the point that GOP operatives have reportedly seen internals showing Shapiro up by as much as a dozen points? Is the RGA thinking of triaging it?

Republican pollsters (and public pollsters) have been very slow to implement the same adjustments that Democratic pollsters have to counter non-response bias among non-college voters (of all races) and overrepresentation of the highly engaged among both party bases. I'm not surprised to hear that Republican internals are showing massive Democratic advantages in the rust belt (where the non-response bias issue is by far the most problematic) and if they cut Mastriano off because of it, music to my ears.

Why have they been slower to adjust? Issues with reaching enough non college voters to weigh properly?

I really couldn't tell you. We know they are having the same problems and missed just as badly as we did in 2020 in mostly the same races/places. They just haven't prioritized fixing the problem, and I don't know why.

Thanks as always for your insights.  I've got a semi-related question, if you're up for it.   It's been reported by some people (e.g. Tom Bonier of TargetSmart, among others) that new registrations since the Dobbs decision are skewing younger, more female, and more Democratic, at least in some states.  Do you think this is true, and if so are you taking any steps to ensure these new voters are adequately represented in your surveys? 

[wbrocks67]

If Mastriano doesn't book air time soon, there very well could be none left for him. TV ad space is not limitless and candidates in the past all over the country have learned this the hard way.

What do you think about the point that GOP operatives have reportedly seen internals showing Shapiro up by as much as a dozen points? Is the RGA thinking of triaging it?

Republican pollsters (and public pollsters) have been very slow to implement the same adjustments that Democratic pollsters have to counter non-response bias among non-college voters (of all races) and overrepresentation of the highly engaged among both party bases. I'm not surprised to hear that Republican internals are showing massive Democratic advantages in the rust belt (where the non-response bias issue is by far the most problematic) and if they cut Mastriano off because of it, music to my ears.

I of course defer to you because you're an expert, but don't you think there is possibly just as much of an issue with pollsters likewise weighting to either or extreme? It certainly feels like on the other end of the spectrum, there are pollsters that are weighting this to a typically "red wave" year who may be missing the type of movement we've seen in all of the specials so far. Like, if you're modeling towards an R+3 environment because you think that's what it will be, versus say, a D+1 environment where it is now in the GCB.

[wbrocks67]

I just...

https://twitter.com/PoliticsPA/status/1568319716602658816

[Orwell]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

[GeorgiaModerate]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Tsk, tsk. Petty.

[S019]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

You call yourself a Democrat?

[TML]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Do you have a contingency plan if the scenario you just described fails to pan out?

[Pollster]

Thanks as always for your insights.  I've got a semi-related question, if you're up for it.   It's been reported by some people (e.g. Tom Bonier of TargetSmart, among others) that new registrations since the Dobbs decision are skewing younger, more female, and more Democratic, at least in some states.  Do you think this is true, and if so are you taking any steps to ensure these new voters are adequately represented in your surveys? 

Registration data isn't something I follow closely (just not the crux of my work), but I'm sure this is true. Registration data is public and if Bonier & others were just making things up, they would've been caught by now.

Voter file vendors who provide us the samples we use for polls update and maintain their lists regularly (especially in election years) so new registrants are included in what they sell us and they have just as equal a chance to be selected for a survey sample as any other registered voter. Even still, though, new registrants usually only exist in small amounts (typically only a few hundred, maybe a few thousand per month) so they generally don't exert much influence. There are some notable exceptions to this though - Sanders' 2016 campaign and O'Rourke and Abrams in 2018 registered huge numbers of new voters - which is why people like Bonier keeping an eye on this stuff is so helpful.

[Pollster]

I of course defer to you because you're an expert, but don't you think there is possibly just as much of an issue with pollsters likewise weighting to either or extreme? It certainly feels like on the other end of the spectrum, there are pollsters that are weighting this to a typically "red wave" year who may be missing the type of movement we've seen in all of the specials so far. Like, if you're modeling towards an R+3 environment because you think that's what it will be, versus say, a D+1 environment where it is now in the GCB.

Very good question - you have your order of operations backwards, though. Most serious pollsters weight only on variables that are independent and trackable metrics like gender, age, race, region, and vote history (sometime we weight to party registration in states that have it since that is an independent and trackable metric as well) and then let the partisan metrics fall where they may, which is why nonresponse bias is such a serious problem. The only decisive/non-marginal guesswork most pollsters (well, Democratic pollsters) do is on education since the census doesn't track that and voter file services have no way to provide it (though many have started trying to model it). Many of us have started weighting to vote recall ("who did you vote for in the 20xx election?") to correct for nonresponse bias though this requires guesswork as well, and it is always risky to rely on self-reported metrics like this since people lie, misremember, and a healthy number will refuse to answer (vote recall questions also often show a bias where people are more likely to say they voted for the winner).

This is a good writeup (my firm didn't participate, but I'm friends with a few of the people who did).

[Rookie Yinzer]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Seek help.

[wbrocks67]

I don't want to jinx anything, and of course just having an R next to his name is probably enough for most GOP voters, but if Shapiro does end up overperforming in November, there will be many things we can think back to to understand why very simply and easily

https://twitter.com/jjabbott/status/1568953981648490498

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

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Seems a bit hypocritical, since you are also an active member of Atlas.

[Orwell]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Seems a bit hypocritical, since you are also an active member of Atlas.

Im not too active on the whole election front, mainly because I've ascended past the need for electoral politics and am waiting for the inevitable military coup.

[GeorgiaModerate]

I want both Fetterman and Shapiro to have a comfortable lead and then lose on election night so I can see all you losers meltdown.

Seems a bit hypocritical, since you are also an active member of Atlas.

Im not too active on the whole election front, mainly because I've ascended past the need for electoral politics and am waiting for the inevitable military coup.

I'm sure there are better forums than this one for military coups.

[wbrocks67]

I of course defer to you because you're an expert, but don't you think there is possibly just as much of an issue with pollsters likewise weighting to either or extreme? It certainly feels like on the other end of the spectrum, there are pollsters that are weighting this to a typically "red wave" year who may be missing the type of movement we've seen in all of the specials so far. Like, if you're modeling towards an R+3 environment because you think that's what it will be, versus say, a D+1 environment where it is now in the GCB.

Very good question - you have your order of operations backwards, though. Most serious pollsters weight only on variables that are independent and trackable metrics like gender, age, race, region, and vote history (sometime we weight to party registration in states that have it since that is an independent and trackable metric as well) and then let the partisan metrics fall where they may, which is why nonresponse bias is such a serious problem. The only decisive/non-marginal guesswork most pollsters (well, Democratic pollsters) do is on education since the census doesn't track that and voter file services have no way to provide it (though many have started trying to model it). Many of us have started weighting to vote recall ("who did you vote for in the 20xx election?") to correct for nonresponse bias though this requires guesswork as well, and it is always risky to rely on self-reported metrics like this since people lie, misremember, and a healthy number will refuse to answer (vote recall questions also often show a bias where people are more likely to say they voted for the winner).

This is a good writeup (my firm didn't participate, but I'm friends with a few of the people who did).

Thank you for this - I just want to make sure I add every time that I (and I'm sure a lot of other people on this forum) are very appreciative of your insights and perspectives.

[wbrocks67]

Aha! So this makes me even more convinced that Susquehanna likely kicked their PA-GOV poll.

Also would certainly be something if Oz started outperforming Mastriano as this suggests, but not wholly unsurprising if Oz is able to coalesce GOP better while Mastriano continues to turn off anybody who is not a die-hard Republican.

https://twitter.com/SusquehannaPR/status/1568069201222422528

[Pollster]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

[MT Treasurer]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

I agree with this — opinions of Mastriano are considerably more hardened, with little room for growth beyond the base (which he, unlike Oz, has mostly already consolidated). I was obviously critical of Oz as a candidate (while noting that there was a relatively easy fix to his biggest problems), but credit where credit is due — forcing Fetterman into a corner with the "Either he’s hiding how unfit he is for the office or or he’s afraid to reveal where he stands on the issues" move was pretty smart. I’m not a swing voter, but I’ve warmed up to him over the last few weeks, and I like to think of myself as somewhat adept at the entire "thinking/feeling like a swing voter" thing. I honestly never had that feeling with Vance.

I don’t necessarily expect this to happen, but Oz running the best Republican comeback campaign of any GOP candidate in the home stretch of this cycle would, by far, be the most amusing plot twist of this election season. FWIW (and I’m sure you agree), I do think there is a certain type of candidate who performs well only under pressure and when trailing in the polls, and this might apply to Oz as well. It was arguably true in Rick Scott's case as well, who came from behind in 2014/2018 but nearly blew a sizable lead in 2010. I’d be curious to hear your thoughts on this, but you could even argue that it’s easier to campaign when you have to climb out of a hole than when you’re trying to maintain a lead, but I’m sure that depends on the type of person you’re dealing with.

[RussFeingoldWasRobbed]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

I agree with this — opinions of Mastriano are considerably more hardened, with little room for growth beyond the base (which he, unlike Oz, has mostly already consolidated). I was obviously critical of Oz as a candidate (while noting that there was a relatively easy fix to his biggest problems), but credit where credit is due — forcing Fetterman into a corner with the "Either he’s hiding how unfit he is for the office or or he’s afraid to reveal where he stands on the issues" move was pretty smart. I’m not a swing voter, but I’ve warmed up to him over the last few weeks, and I like to think of myself as somewhat adept at the entire "thinking/feeling like a swing voter" thing. I honestly never had that feeling with Vance.

I don’t necessarily expect this to happen, but Oz running the best Republican comeback campaign of any GOP candidate in the home stretch of this cycle would, by far, be the most amusing plot twist of this election season. FWIW (and I’m sure you agree), I do think there is a certain type of candidate who performs well only under pressure and when trailing in the polls, and this might apply to Oz as well. It was arguably true in Rick Scott's case as well, who came from behind in 2014/2018 but nearly blew a sizable lead in 2010. I’d be curious to hear your thoughts on this, but you could even argue that it’s easier to campaign when you have to climb out of a hole than when you’re trying to maintain a lead, but I’m sure that depends on the type of person you’re dealing with.

I still think they both lose but yeah Mastriano outperforming him never made sense

[wbrocks67]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

I agree with this — opinions of Mastriano are considerably more hardened, with little room for growth beyond the base (which he, unlike Oz, has mostly already consolidated). I was obviously critical of Oz as a candidate (while noting that there was a relatively easy fix to his biggest problems), but credit where credit is due — forcing Fetterman into a corner with the "Either he’s hiding how unfit he is for the office or or he’s afraid to reveal where he stands on the issues" move was pretty smart. I’m not a swing voter, but I’ve warmed up to him over the last few weeks, and I like to think of myself as somewhat adept at the entire "thinking/feeling like a swing voter" thing. I honestly never had that feeling with Vance.

I don’t necessarily expect this to happen, but Oz running the best Republican comeback campaign of any GOP candidate in the home stretch of this cycle would, by far, be the most amusing plot twist of this election season. FWIW (and I’m sure you agree), I do think there is a certain type of candidate who performs well only under pressure and when trailing in the polls, and this might apply to Oz as well. It was arguably true in Rick Scott's case as well, who came from behind in 2014/2018 but nearly blew a sizable lead in 2010. I’d be curious to hear your thoughts on this, but you could even argue that it’s easier to campaign when you have to climb out of a hole than when you’re trying to maintain a lead, but I’m sure that depends on the type of person you’re dealing with.

This is doing a bit of revisionist history IMO. First of all, unless new polling says otherwise, nothing in the evidence that we've gotten from polls or from bits of focus group data shows that voters are really in tune with Oz's health attacks - if anything, the evidence seemed to show that they were backfiring. Not just that, but Oz used this for the whole debate issue, and then Fetterman agreed to a debate, and then Oz was still complaining.

Not just that - but Oz doing an appearance or two in the suburbs with Nikki Haley and Pat Toomey is not something that really shows that it's going to have a major impact given that in that same time period, we've already had numerous other controversies with Oz like his incest remarks leaking or even the stuff about dogs on social media.

This is why I think articles about his trying to appeal to suburbs are so out of touch - you can't argue that Oz is suddenly trying to appeal to the suburbs just one or two weeks after he literally appeared with Trump and Mastriano at a rally, and right after audio leaked where he said abortion is murder. Even as I said before, his current position on abortion appears to be a total ban except for rape, incest, and life of mother which is still wildly out of step with the GE, let alone the suburbs. People in the suburbs are not dumb - just because he comes and shows up with Nikki Haley doesn't mean anything (as if she's some become of moderate-ness either)

If the bar is on the floor, then sure maybe Oz is doing a little less badly than in the summer when he was getting owned every single week. But that is a very low bar.

[kwabbit]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

I agree with this — opinions of Mastriano are considerably more hardened, with little room for growth beyond the base (which he, unlike Oz, has mostly already consolidated). I was obviously critical of Oz as a candidate (while noting that there was a relatively easy fix to his biggest problems), but credit where credit is due — forcing Fetterman into a corner with the "Either he’s hiding how unfit he is for the office or or he’s afraid to reveal where he stands on the issues" move was pretty smart. I’m not a swing voter, but I’ve warmed up to him over the last few weeks, and I like to think of myself as somewhat adept at the entire "thinking/feeling like a swing voter" thing. I honestly never had that feeling with Vance.

I don’t necessarily expect this to happen, but Oz running the best Republican comeback campaign of any GOP candidate in the home stretch of this cycle would, by far, be the most amusing plot twist of this election season. FWIW (and I’m sure you agree), I do think there is a certain type of candidate who performs well only under pressure and when trailing in the polls, and this might apply to Oz as well. It was arguably true in Rick Scott's case as well, who came from behind in 2014/2018 but nearly blew a sizable lead in 2010. I’d be curious to hear your thoughts on this, but you could even argue that it’s easier to campaign when you have to climb out of a hole than when you’re trying to maintain a lead, but I’m sure that depends on the type of person you’re dealing with.

For all of his flaws, no one can say Oz isn’t a very intelligent man. Despite being out of touch, he’ll have some wherewithal that most of the GOP candidates might not have. Besides being an out of touch carpetbagger, it’s harder to hit Oz on many of the issues. Despite veering right in the primary, Oz doesn’t have the image of an crazy extremist nor someone that would totally inept (like Walker).

I think Oz finally caught on that he was heading for an embarrassing defeat and decided to actually try to campaign. If he’s actually putting effort in, he should be decent.

[wbrocks67]

Oz is clearly trying to right the ship (and doing a not-horrible job) whereas Mastriano's aides reportedly isolate him from any dissenting opinions - not hard to see why/how Oz performs a few points ahead of him, especially with Senate control on the line.

I agree with this — opinions of Mastriano are considerably more hardened, with little room for growth beyond the base (which he, unlike Oz, has mostly already consolidated). I was obviously critical of Oz as a candidate (while noting that there was a relatively easy fix to his biggest problems), but credit where credit is due — forcing Fetterman into a corner with the "Either he’s hiding how unfit he is for the office or or he’s afraid to reveal where he stands on the issues" move was pretty smart. I’m not a swing voter, but I’ve warmed up to him over the last few weeks, and I like to think of myself as somewhat adept at the entire "thinking/feeling like a swing voter" thing. I honestly never had that feeling with Vance.

I don’t necessarily expect this to happen, but Oz running the best Republican comeback campaign of any GOP candidate in the home stretch of this cycle would, by far, be the most amusing plot twist of this election season. FWIW (and I’m sure you agree), I do think there is a certain type of candidate who performs well only under pressure and when trailing in the polls, and this might apply to Oz as well. It was arguably true in Rick Scott's case as well, who came from behind in 2014/2018 but nearly blew a sizable lead in 2010. I’d be curious to hear your thoughts on this, but you could even argue that it’s easier to campaign when you have to climb out of a hole than when you’re trying to maintain a lead, but I’m sure that depends on the type of person you’re dealing with.

For all of his flaws, no one can say Oz isn’t a very intelligent man. Despite being out of touch, he’ll have some wherewithal that most of the GOP candidates might not have. Besides being an out of touch carpetbagger, it’s harder to hit Oz on many of the issues. Despite veering right in the primary, Oz doesn’t have the image of an crazy extremist nor someone that would totally inept (like Walker).

I think Oz finally caught on that he was heading for an embarrassing defeat and decided to actually try to campaign. If he’s actually putting effort in, he should be decent.

That's why I think the most effective attack on Oz is still just being incredibly out of touch with PA. You're right, the 'extremist' angle doesn't really mesh because while Oz is a fraud, he seems more willing to do whatever rather than stake out a purposeful extremist agenda. Let alone that, but he's also a TV physician, so it's harder to craft that image.

However, all of that being said, that's why I think the "Oz is a carpetbagger who knows nothing about real Pennsylvanians and is a quack TV doctor" is the best hit you have, and it's the most effective. The whole schtick with Oz is that he is some rich guy who flew in from Jersey who has little to no ties to actual Pennsylvania or the people in the state, and thinks he can buy a senate seat.

I think it was in the article about Fetterman's rally on Sunday where they interviewed a Trump 2016/2020 voter who was voting for Fetterman, and he said essentially the same thing - Oz just doesn't know what it's like for real people in this state, and he can't connect to them.

I think Oz has also changed his positions so many times throughout the years that the whole "Which Oz can you actually trust?" angle is also effective. One day he says abortion is murder, another day (years ago) he's pro-choice, and now he's for a total abortion ban but in certain cases. He's the definition of a terrible flip-flopper.