The absentee/early vote thread

[fldemfunds]

Anecdotally, I know a few female ted Cruz supporters in Texas crossing over at the top of the ticket.

[fldemfunds]

2012 FL exit polls had it at 67-13-17, so not much change except a weird drop in black turnout. What's all the rest supposed to be?

I wonder why that is. I would have to assume that such a drop right now is something that will eventually even out, even if only on election day. Even in 2004, before Obama, African American's share of the FL electorate was 12% (vs 13% in 2008/2012)

A 5% drop would be an unmitigated disaster for Democrats.

There has not been any weekend early voting yet, let alone the big souls to the polls drive. This weekend, Dems should take a decent lead in overall turnout. And in FL, NPAs/Independents tend to be Dem-leaning. So even though GOP ahead by 0.5%, HRC likely up 2-3% in overall voting.

[fldemfunds]

Based on EIP/VBM voting, I would not be surprised to see an HRC win by 5-7% in FL. It really depends on turnout on election day. Election day determines whether FL is bad for the GOP or atrocious for the GOP.

[fldemfunds]

https://electionsmith.wordpress.com/2016/10/30/exclusive-in-florida-no-party-affiliation-npa-voters-and-hispanics-who-skipped-2012-or-werent-registered-are-flocking-to-the-polls/

- Republicans are cannibalizing 2012 Election Day voters at a higher rate than Democrats: 18.7% of the 1.51m Republicans who’ve voted thus far in Florida waited in 2012 to cast a ballot on Election Day; 16.8% of the 1.43m Democrats who’ve voted to date voted on Election Day four years ago.

- Over 29% of the 476k Hispanics who’ve cast ballots thru yesterday are either new to the registration books or skipped the 2012 election. This is true for only 1/5 of the 2.5m white voters who’ve voted, and 17% of the 391k black voters who’ve cast ballots thus far in 2016.

Those NPA Hispanic voters were the group that Nate Cohn said his poll would not be able to catch. 29% of 476K is 138K new voters. They're probably breaking about 3-1 for Clinton too. That's part of how she can lead by 13-15 points in the early vote when party ID stats are even

This is what I've been saying to everyone who is nervous. NPAs are usually unaffiliated dems in FL. The fact that party reg is even, means Hillary is probably up 3-5% right now, at least. She's going to win FL by 5%+.

[fldemfunds]

don't really get the cannibalization theory right now, since republicans are in fact leading in FL.....AND according to polls more trump voters still want to vote.

would need the UNAFF to be real gamechangers this time.

@NV...trump is toast.

Because likely voters are reliable. If someone has voted in 3/3 of the last presidential elections, then with less effort or no effort, they're likely to vote in this one. So if you go into election day and Reps have 50% of their most reliable votes in, but dems have 25% of their most reliable votes, you can guess who is more likely than not to have the better election day.

Now regarding the Republican lead. It's less than 1%. Florida is complicate for party registration purposes, but less so than in 2008. At this same point in 2008, Republicans led in total votes by approximately 6%. They're leading right now by less than 1%. Considering there's been a decent, though not complete, sorting out of dixiecrats changing their registration to R, this bodes extremely well for Democrats. Basically, in 2008 *when Dems barely won Florida* the GOP had a 6% in REGISTRATION TURNOUT advantage, which probably understated their lead because of the higher number of Dixiecrats, but now, with a lot of those same Dixiecrats now registered as Republicans, voter reg numbers are even. This does not include NPAs which tend to be dems in Florida.

[fldemfunds]

According to Tom Bonier, people's self-report about having voted early isn't reliable so you can't take it at face value but...

Early voting:
FL - 36% have voted; HRC leads 54%/37%
NC - 29% have voted; HRC leads 61%/33%

This is based on self-report, I believe.

But even if we were to allow for rooms that more Democrats lied about having voted early, it's safe to say that Hillary is doing fine so far in FL. She's leading, though probably not by 17% like above.

Where did you get these numbers?

[fldemfunds]

Sig bet. Hrc wins fl by 5 or more

[fldemfunds]

Sig bet. Hrc wins fl by 5 or more

I think she'll take FL, but you're out of your mind if you think it'll be by 5. Neither the polls nor even a generous reading of the early vote sees that coming.

Polls have moe. Also I happen to think because of the party shuffling and npas, her lead is being understated in fl.

[fldemfunds]

https://twitter.com/tbonier/status/793438423877124096
Good news from TargetSmart regarding Florida.

1) With over 4 million ballots cast, GOP registrants outnumber Dems by 16k votes cast. That's a 0.4% advantage. However...

2) Almost 700k votes have been cast by Unaffiliated voters. And those voters skew heavily Dem, with a 13% modeled partisan advantage

3) Unaffiliated EV/AV are more likely to be Hispanic (22%), or Millennial (16%), and only 65% are white

4) Tomorrow AM we will release our latest @_TargetSmart and W&M tracking survey of Florida, including full breakouts of EV/AV.

5) I've seen the initial results, and I can say they will challenge the conventional wisdom of where the race stands. Tune in tomorrow!

Interesting

This is why I've been so adamant that democrats will win by 5+.

The registration and performance gap for democrats in 2012 was about 5.5%. Roughly 450,000 reg changes to republican happened. This covers almost the entirety of the gap. At least enough to warrant saying that registration numbers are more reflective to a substantial degree than they were in 2012. So if you just stopped there, Florida looks like it did in 2012 with registration numbers becoming a more accurate indicator of performance.

However, we've also had substantial new dem and npa registrants. 2/3 of new registrants since august have been non white voters and approximately half are non white since 2012. Dems have a substantial npa advantage in Florida. You can't simply say dems need to match reg based turnout from 2012 because that is before the republican primary had a sorting effect on dixiecrats who voted gop, so reg numbers used to overstate performance, but now because of the npa surge, reg numbers drastically understate the advantage.

[fldemfunds]

don't be so hostile of whites...male turnout is stable, white women are up.

This.

Plus a lot of those white voters are college educated in the Research Triangle. Being white =/= being a Trump supporter necessarily.

[fldemfunds]

Obama actually won the entirety of the early vote by 67K when VBM was included, IIRC. He won the state by 64K. He won in person early voting by 200K (1.35M to 1.15M and lost VBM by about 133K or 5%).

Are you sure about Florida? I read Democrats cast close to 250,000 more ballots than Republicans in 2012 while currently we have a Republican edge there.

There was a massive sorting affect in Florida this cycle because of Trump. About 300,000 to 400,000 people switched to GOP to vote in the primary. (Source: http://bigstory.ap.org/article/d1c611a35aa5488789e0ad0cf4f2e818/gop-gains-ground-dems-voter-registration-key-states). However, approximately 3/4 of new registrations have been non-white voters since August, and half have been non-white since 2012. (Source: http://www.politico.com/states/florida/story/2016/08/increase-in-minority-voters-poses-problem-for-trump-in-florida-104717)

The thing with those people who switched from Democrat to Republican is THAT THEY WERE NOT VOTING FOR DEMOCRATS. They hadn't since 1964 (maybe Carter). They weren't Democrats. So even though Democrats would constantly go into election day with a 4-6% lead in turnout according to registration, it was closer to 1% actual votes cast.

But if you look at who has registered (and turned out) among new registrants, whether Democrat or NPA, they are largely either young or Hispanic and lean Democrat. So while the turnout by registration has leveled off for Democrats, when you include NPAs into consideration, Democrats have added probably 5-6% of cushion on their margin from 2012.

[fldemfunds]

Obama actually won the entirety of the early vote by 67K when VBM was included, IIRC. He won the state by 64K. He won in person early voting by 200K (1.35M to 1.15M and lost VBM by about 133K or 5%).

Are you sure about Florida? I read Democrats cast close to 250,000 more ballots than Republicans in 2012 while currently we have a Republican edge there.

There was a massive sorting affect in Florida this cycle because of Trump. About 300,000 to 400,000 people switched to GOP to vote in the primary. (Source: http://bigstory.ap.org/article/d1c611a35aa5488789e0ad0cf4f2e818/gop-gains-ground-dems-voter-registration-key-states). However, approximately 3/4 of new registrations have been non-white voters since August, and half have been non-white since 2012. (Source: http://www.politico.com/states/florida/story/2016/08/increase-in-minority-voters-poses-problem-for-trump-in-florida-104717)

The thing with those people who switched from Democrat to Republican is THAT THEY WERE NOT VOTING FOR DEMOCRATS. They hadn't since 1964 (maybe Carter). They weren't Democrats. So even though Democrats would constantly go into election day with a 4-6% lead in turnout according to registration, it was closer to 1% actual votes cast.

But if you look at who has registered (and turned out) among new registrants, whether Democrat or NPA, they are largely either young or Hispanic and lean Democrat. So while the turnout by registration has leveled off for Democrats, when you include NPAs into consideration, Democrats have added probably 5-6% of cushion on their margin from 2012.

Where is the specific quote of 300-400k switching?

https://electionsmith.wordpress.com/2016/10/15/the-last-word-on-voter-registration-numbers-in-florida/

You must be logged in to read this quote.

Daniel Smith finds that the big Dem drop is mostly old Dems dying or being inactive. There are some switchers, but not a huge amount.

http://fivethirtyeight.com/features/republicans-voter-registration-gains-probably-arent-gains-at-all/

60% of the approximately 300k narrowing is attributable to party switching. I don't think analysis includes NPA to GOP. My estimate of 300k is probably on the higher end, it is probably closer to 200-250k.

Another estimate: http://www.sun-sentinel.com/news/politics/sfl-by-the-numbers-the-trump-effect-in-florida-20160325-htmlstory.html

38 out of every 10,000 (and there were approximately 5 million dems) puts it at roughly 200k.

Even given this slightly lower number, my substantive point still stands. 200k voters left the party for our opposing major party (representing most of the registration-turnout gap from 2012). NPA registration has heavily skewed Democratic. Given these two factors, two things are true: 1) partisan turnout is significantly more reflective of the actual vote in Florida, and 2) NPA votes are probably breaking to Hillary by a pretty decent margin (my guess is 55-45 on the slightly conservative side).

When you combine these two realities, there's not much stopping you from concluding that HRC's lead is probably around 5% right now with the turnout numbers being what they are

[fldemfunds]

https://mobile.twitter.com/_TargetSmart/status/793635374828363776?s=09

It's nice being right

[fldemfunds]

http://www.nbcnews.com/card/hispanics-make-greater-share-florida-early-vote-2016-n676486
Whites down 4%.

With a week of EV to go. I think they'll be around 66-67% this cycle at the end of it all.