There is still a ways to go and a lot can change in three weeks, but as far as PA is concerned, the last few days have been the most revealing thus far when it comes to the POTUS race since the start of election season.
Why? Let's break a few key factors down. ⬇️
Critically, we got a pair of battleground district level polls from local pollsters who are HQed in those respective districts.
PA-07, which tends to be pretty reflective of the statewide vote as a whole due to demographics, was found to be Harris+3 (50-47) by Muhlenberg.
Meanwhile, in south-central PA-10 which is left-trending and growing in population, SP&R found Harris up by 5 (46-41).
While Trump's share is way too low, the 46% mark for the VP is notable. For her to get close to 50% statewide, she needs to hit about that % in the district.
Validating those signals is the bevy of quality nonpartisan polling we've been getting around this time. NYT/Siena today showed her up 50-47, for instance. Fox, Cook, Fabrizio/Impact, Research Co., etc. all have her at 50%.
49s have been common too (Q-Pac, TIPP's RV poll, etc).
Which brings us to something people should be aware of. Back in '16, Clinton was really starting to struggle around this time.
SP&R and Muhlenberg that year only found her at 44% and 45% respectively! A lot of 45-47%s and slipping in district level polls.
This is important because when we talk about polling these days in the context of "oh wow, they missed Trump badly," it usually refers to margin and/or undershooting Trump's share.
They were generally pretty good about getting close to actual Dem shares in the averages.
For example, 538's 2020 PA aggregate was 50.2% for Biden, nearly exactly what he got. In 2016, RCP was 46.8% Clinton (she got 47.5%).
This year, we have the additional factor of Trump's % being pretty close to his 2016/2020 levels. Have pollsters fixed the issue? We'll see.
That segue aside, VBM for Dems is off to a strong start in returns and rates. The GOP is being outspent on the presidential level and a lot more heavily in all battleground districts save for PA-01 (Fitzpatrick) + PA SEN, where parity in personal v. outside spending is different.
Ground game has also been a sticking point. As many of you know, Republican efforts have been increasingly outsourced to outsider orgs and PACs while Dems have a lot more field offices, personnel, and tighter in-house organization.
Let me be clear - this is a pretty tight race and the bottom could get shaky for either side later down the road (in polling, VBM, organization, etc).
But, objectively speaking, *at the moment*, one faction does have a slight advantage. Let's see if it holds.
P.S - A few people may have read this incorrectly - 46% in PA-10 implies Harris getting close to or hitting 50% statewide. My apologies if it was interpreted differently.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Lots of thoughts in the coming days, but first and foremost - this was not a turnout thing. After all the votes are in, PA will be right around 2020 levels.
This was persuasion, and the GOP had their best result here since probably 1980. PA Dems got punched in the face.
The magnitude of what happened will take some time to break down, but incumbents like Wild and Cartwright who have been in tougher situations than this went down in flames while row offices bombed.
The top of the ticket basically cratered everybody because while most Dems…
are outrunning Harris, Trump’s winning the state by enough that it doesn’t matter. For instance, Casey is still running ahead of Harris on margin by ~2, same as Slotkin and Baldwin, but the problem is that Trump is winning PA by the most out of the three.
It's been a long election cycle with ups and downs, but my gut feelings + data are about the same as they were in August.
I think VP Harris is favored to win PA by 2 pts, give or take 1 while Senator Casey wins by 4-5.
Let's break down why. ⬇️
There are limitations to how much more Trump can squeeze out of those rural areas without cracking their small urban centers - IE, State College while facing off against high turnout overall.
Some data points for this include the WA primary, Selzer, the PA district polls, etc.
Concurrently, the suburbs (SEPA and co.) are highly likely to shift left again. They've done so unbrokenly in every cycle since '16 and absent decent racedep, how do you counter additional raw volume from, say, Montgomery?
You look to rurals, which I think stay somewhat static.
Ok, buckle in. Here's how Pennsylvania is going to play out tomorrow. You'll want your notifications on since I'll be live tweeting updates throughout the day regarding turnout in certain areas, new developments, etc.
Once the polls close, this is where things get interesting.
Expect county by county comparisons and analyses throughout the night like I have in years past. I'll cover the top of the ticket, congressional races, and state legislature control.
I don't call races unless I'm sure there's no path w/ private and public data.
Like 2022, you should expect a blue mirage at first when Allegheny drops a large batch of VBM, followed by a smattering of other mail and ED from initial counties.
Some counties will drop batches of mail later in the night or the next day, so...
Now that voter registration in PA has closed pre-election, let's examine what exactly is happening numbers-wise.
The perception is that Dems have lost a lot of ground, but the full picture is a little more more complicated than that. Let's start with the table you see here.
Voter reg is a lagging indicator and we're still in the process of getting to the point of what you'd expect to see in a tossup state.
With how shifts are being described, you'd think a 240K decline in Dems are entirely at the benefit of the GOP. Not quite.
Total RVs are only slightly up from 2020 w/ R's up by ~170K, but what's gone under the radar is a similar increase in the IND/OTHER share.
As you probably know by now, they're not particularly conservative voters. They're younger folks, ppl moving into the state, etc.
🔵 DEM: 881,779 - 73.7% returned
🔴 GOP: 501,736 - 70.6% returned
🟡 IND: 166,613 - 59.5% returned
VBM Splits: 🔵 56.9% / 🔴 32.4% / 🟡 10.7%
🗳️ Ballot Edge: 🔵+380,043
📈 Return Edge: 🔵+3.1
Thoughts⬇️
If you're curious as to what's happening on the GOP side, read this excerpt and the overall thread. This phenomenon of Republicans surging with "in-person" early voting for whatever reason is taxing on a system not optimized for it.
As mentioned before, this simply has the effect of moving some of their ED voters over to VBM, which is fine. But I better not hear complaints about election offices because some people decided to badger those poor workers.
It's a departure from the last few cycles pattern-wise.
We're in the thick of the final stretch now, so allow me to explain in detail what the deal is with how VBM/EV works in PA. This is important to understand due to the tremendous amount of misinformation and comparisons being made recently.
First, the current system we have under Act 77 is actually fairly new - only a few years old. It was passed toward the end of 2019 by the Legislature. This year is the first non-COVID presidential election with it in effect, following the first midterm + off year.
It cannot be overstated enough how 2020 was essentially flying blind with a patchwork concept that almost buckled under the pressure of a high volume of mail and 67 different counties/BoEs running a novel system.
Things are a lot better now, but that first year was rough.