I’m still not sure you get apples to apples. In 2020, the man that talks through the weave convinced his followers not to vote early or by mail. This time, he’s letting them. And so I am not sure we’ll get an accurate picture on early voting or mail-in votes because of this.
These are the same number from a statistical perspective. 1,377 difference given how many tens of thousands are included is not a trend given only 2 points on the line and the fluctuations sure to follow.
The PA numbers are fascinating. Statewide, Dems have returned 41% of the percentage of Dem ballots sent out, but if you look at the 3 largest Democratic hub counties in the state, Dems are running well over a 50% return rate which indicates clerks offices aren't severely behind in counting Dem heavy counties. Allegheny, Delaware and Philadelphia are running a combined >50% return rate of ballots sent out. This may be indicative that more GOP friendly counties are simply behind in their counting.
Not when you consider GOP added ~3,700 more votes than their previous day gain and that Dem ballots sent out were more than 2-to-1 in favor of Dems. That's also two days in a row that the GOP number grew with greater velocity (% increase) than Dem numbers. It's because GOP voters vote closer to election day. That trend will continue to tilt GOP. Especially given that we know Philly and Pittsburgh are actually running ahead of the state in counting the returned ballots.
Agreed that is will not be apples to apples since it is going to be difficult to figure out if the Trump campaign is cannibalizing their election day vote for early votes. However, Democrats have always worked hard to gain an advantage by banking votes early. That has not changed. The number of votes they are able to bank early versus the levels they have done in the past, could be an indicator of where the state might be going. So if Harris is over Biden's numbers by a 100k to 200k, then she would look to be in good shape to win the state.
That can't happen in the end, because there are only about half the number of mail in ballots that were approved to be sent out in PA versus 2020. Also, consider that early voting was extended to 50 days before Election Day this year in PA, so people there have had more time to vote than they did in 2020. Based on what I'm seeing, the mail in voting doesn't look great for Harris. 50% Dem return rate currently in Philly, for example, when they've had over a month to return the ballots and historically, GOP votes late. 41% return rate statewide. That's not good. (for Dems)
I was not speaking directly about PA. This goes for every swing state that provides publicly available information about their numbers of mail in/early voting by affiliation.
Gotcha and to your point about mail-in voting in general, I'd say the perception is these numbers are not looking good for Harris in most states where they are available. Which is a big reason you've seen the betting odds move so strongly towards Trump. People betting several thousands of dollars on a race are taking a look under the hood before they do it.
It’s funny the divergence between states. I’ve posted on MI previously. PA looks good for the Dems right now in early voting. Wisconsin appears to favor the Dems a bit. Arizona favors the pubs, this one seems really out of line with expectations so will watch this one over the next week. GA had a huge turnout early, hard to know who that will favor in the end. Large turnout usually favors Dems, and it’s Disproportionately women which helps them as well. But AAs are under voting a bit so far, and Republicans generally over voting. Nevada has favored the republicans early, but I read those are mostly military votes. Would like to see a broader sample. NC starts today, those numbers will be interesting. Might decide the election.
How does a party cannibalize their voters? Whether they vote early, late or on election day, it all counts the same.
The question is how do you look at an increase in GOP turnout in VBM and EIPV from '20 to '24. If Republicans had a in person vote on election day advantage of +40 in '20 but only had about 10% of their votes come from VBM/EIPV, how do you predict what the election day in person voting advantage for R's would be in '24 if they had 20% of their votes compared to '20 happened through VBM/EIPV.
You don’t. The best predictors are the polls. In a race this close, early voting stats can only tell you so much. However, I will confess that I have always been envious of Dem early voter turnout and if GOP voters are bucking that trend this year, I’m going to be very happy. Once you vote, you can’t change it. So I’d want to be on the team with the early lead in this. Traditionally, however, GOP voters have tended to vote later rather than sooner. There exists a very distinct possibility that Trump turnout is being underestimated. But the truth is, you can only glean so much from early voting statistics. For example, in Pennsylvania, a democrat returned ballot does not necessarily equal a Harris vote.
The polls are not the best predictors at this point. Every campaign has modeled what their turn out needs to be to win the election. The campaign should have their voters ID'd at this point and be working on getting them to confirm they have either voted or have a plan to vote. They should be doing that down to the precinct level. Once you know you have the high propensity, you try and get the lower propensity voters. The polls all have a MoE of somewhere between 2% to 3.5%, and when all the swing states are within that MoE there is no way to know if their voter model is accurate enough to rely on.
The polls used to be the best predictor, but they’ve been wrong since 2016, both ways. Polls tell us what a certain group do people are thinking and why. They don’t tell us if that person or their family actually voted (or will vote).
And what would you say the MoE is for publicly available early voting statistics? MoE being the assumption that Dem voters are voting Dem, GOP voters are voting GOP and determining actual outcome of election? In some areas, I would agree, it's easier to see how well you are doing. For instance, in Philadelphia county, PA, Biden received 82% of the vote in 2020. It being the most populous county in the state of PA, you can more easily see how Harris will perform based on early voting/absentee numbers. But beyond that, it gets murkier quick. And I still stand by the fact that a campaign can't cannibalize their vote. You cannot simply assume that certain things will happen just the way they did in 2020. Trends? Maybe, but getting concrete data from early voting when the state has been polling so tightly, it's an educated guess at best. That said, given that money is being wagered in heavy doses right now at a 60% clip in President Trump's direction, I'd say gamblers are feeling good about what they see in the EV stats, but still, it's just an educated guess at best. Like I said before, in PA, early voting was extended to 50 days this cycle, but only about half the number of absentee ballots were approved to be sent out in comparison to how many were sent out in 2020. So you'd be making apples to oranges comparisons from the start looking at EV numbers there.
Keep in mind, for Arizona, not all ballots have reached mailboxes yet. My household got one late last week, but the other two didn't arrive in the mail until yesterday. There are still a number that will be delivered today and Saturday. This might explain why Pinal County, with less than half the population of Pima County, has returned 2X the number of ballots. Pima, home to Tucson and the U of Arizona, is generally the bluest of the counties in Arizona. This time next week will be a better indicator for Arizona.