Join the Email List




Monday
Dec102018

Let's talk about polling, again.  

Let’s huddle up for a second and have a little chat about polling.

In 2018, I spoke to 25-30 groups and without fail, there were two questions always asked:

  1. How is Blake Bortles still in the NFL?
  2. And some variation of Why is the polling always wrong?

Now that the Jaguars have benched Bortles, we can dispense with the first question and focus on the second one.

There are basically two kinds of polls:  the ones you don’t see, and the ones you do.  Candidates who are spending millions and millions on television need good data, and that is the polling you don’t see – or at best rarely see.  In the public line, these are the dreaded “internal polls” – polls that when they see the light of day must be wrong, because they are released with an agenda.   However internal polling is typically pretty spot on – for two very connected reasons:  political pollsters stake their reputation on good numbers, because candidates must have the best information to make decisions – and candidates spend a lot of money for those polls to be accurate.  For example, during the 2018 Democratic primary for Governor, while the public polling missed it, our internal Graham polling was pretty clear that Jeff Greene's negative attacks on Phil Levine and Gwen created significant downward movement for the frontrunners in the race and space for Andrew Gillum to rise -- just as in the same way, the DeSantis private polling, which was released, showed his surge over Putnam to be earlier, more significant, and more sustained than the public polling reported.

Then there are the public polls.

First, longtime readers of this blog will know my issues with public polling did not start in 2016 or 2018.  Longtime political analyst Charlie Cook once called public polling “dime store junk,” a phrase that sometimes, is charitable.  I’ve been particularly harsh on Quinnipiac (I believe I've called their polling a "dumpster fire" and once suggested they couldn't count the topings on a pizza), not because I have any particular dislike for the school, or their beloved mascot Boomer the Bobcat, but because their polling is often cited as a benchmark.   When organizations like Quinnipiac publish polls, given the brand they have created, people take them at face value -- despite the fact their polling in Florida has often been a disaster – like the Jaguars football season.

Let me give you an example.  The 2012 Romney/Obama Florida race was one of the most stable races I’ve ever been around.  Both candidates started with a pretty high floor, and while there was movement, there were never any big shifts in the race, and the race never moved far from even.  Yet, over a four-month period, Quinnipiac had the race go from +6 Romney to +9 Obama – then within a month, back to Romney +1.  Over that period, the RCP polling average moved a couple tenths of a point.  Another university pollster called the race for Romney just weeks after they had Obama +3.    In reality, the race was always very close.

Public polling in the Governor’s race here in Florida in 2018 was, to quote noted linguist Deion Sanders, a total “shibacle.”  Back to the Quinnipiac, I felt like I spent most of October dealing with texts/emails/tweets from activists/donors/supporters wondering why I kept saying the Senate and Governor’s race were close, when the Q poll kept saying it wasn’t.  And they weren’t alone – the bulk of public polling lived in a reality that was separate from the real one.

So where is the disconnect?  Let’s explore a few things:

  1. First, most public polling is done at a fraction of the cost.  That alone will diminish the quality – in a lot of university polling, live callers are students, not trained call centers.  Robo polls are cheap and can’t be used to reach cell phone users (though some use internet panels to supplement).  It isn’t a hard and fast rule about everything, but generally in life, if you spend $1,500-2,000 versus spending $40-60,000, the former product will be inferior.
  2. Florida’s voter file is public, but many pollsters still use random digit dialing off phone lists, which will always lead to a survey that is broader than the electorate at large.
  3. Florida’s electorate is exceptionally stable, yet many public pollsters don’t weigh their models.  For example, some just let party ID “float” – meaning it pegging the turnout model to wherever the random sample lands it, making a race seem fluid, when in fact, the only thing really moving is the public pollster’s model.
  4. Florida is a hard state to poll, particularly with ethnic minorities.  Our state’s Hispanic and Black populations are both exceptionally diverse and missing the mark here can really mess up a survey.  For example, take a survey of 800 Florida votes, and you probably will get about 120 Hispanic respondents.  If that sample is too Puerto Rican, the whole thing will be too Democratic – and if it is too Cuban, it could make it look too Republican. 

And keep in mind, beyond this, polling that is done well typically has a 95% confidence rate – meaning that 5% of polls that are done are going to be off.

The problem with these issues is they create a lot of “noise” in polling – as in the 2012 example of Quinnipiac showing a 15 point move in a race over four months that maybe moved 2 points.

And here’s where it starts to go south. The media reports public polling as fact, typically with very little context, and often with no regard for a pollster’s record – particularly as news coverage looks more and more like sports coverage, with the focus on who is winning or losing.  Put numbers next to a name, and add a little paragraph about the poll, and someone will at least tweet it.  And supporters of different campaigns latch on to one poll or another, to bolster their own arguments. 

I do think there is a place for public polling, and a lot of groups do a ton of fascinating work.  For example, you should read everything the Pew Trusts puts out – not because of their great horserace numbers, but because they engage in fascinating surveys about the political and cultural fabric of America.

But since I don’t think public horserace polling is going away, here are a few ideas on how we should consume it going forward:

  1.  Every poll that is reported should have a very specific methodology statement that breaks down their sample in very specific terms.  All good science – and polling is a science – should be replicable. It shouldn’t take high level calculus to re-engineer a poll.  Who did they call?  What was the party split, ethnic split, gender split, and regional breakdown?  And how was the data collected? None of this is too much to ask.  If this isn’t available, don’t report it.
  2. Individual polls should be reported next to polling averages.  The averages themselves aren’t perfect, but at least provide some context of when a poll is outside of reality – therefore when CNN releases a poll showing the Governor’s race in Florida at a 12-point margin, the consumer can see this is an outlier.
  3. To this last point, journalists need to do a better job of filtering this stuff.  Journalists don’t have to report every poll that comes across the email.  My friend Tom Eldon takes it a step further, and suggests, just like the College Playoff Committee only updates its poll once a week, maybe journalists should aggregate polling, and release it once a week.
  4. Finally, it is not news that Florida is a very close state – top level races are going to be inside the margin of error – so when someone releases numbers that show someone winning by a margin that is outside the norms of political reality, even if they come from an organization with “university” in the title, there is no requirement to publish them.

There are some journalists who are wise in how they report polling, and not every public poll is a mess, but overall, the incentive is for groups to create public polls, because public polls create news – and news creates interest in the organization doing the polling.  And while I don’t believe organizations intentionally publish suspect data, again, there is little incentive to tighten up their internal controls and try to get things closer to right.  Until the news media decides to collectively be more careful in reporting this data, it is up to the rest of us to be skeptical.

Monday
Nov192018

So, about Tuesday night...

OK, it is time to talk about that thing that happened.  No, not the Jaguars blowing a 16-point lead in the 4th quarter – though I have plenty to say about that – I am talking about the election. 

Before I get into it, this election was the third consecutive Governor’s race decided by a point or less, bracketing two consecutive Presidential elections decided by a point.  This drives homes two points:  One, Florida, for all its dynamic growth and demographic changes, is very stable; and Two, when organizations like Quinnipiac try to peddle off polls showing candidates in Florida with 6-point leads, or 9-point leads, you now know what to do with that information (a post/rant on public polling is coming soon).

There are a lot of reasons why Florida is very competitive – you can read my take here – but it is what it is.  Big chunks of Florida cancel each other out, and both parties have large, and quite dug-in bases – and neither have a base that alone gets them to 50% + 1.  Winning Florida (or losing it) is about managing the margins throughout Florida. 

A couple of things – for the sake of this exercise, I am going to look at the Governor’s race, for one, and only one reason:  the consistency of comparing races over time.  In terms of the Senate race, much of the difference between Nelson and Gillum occurred in places like Brevard, where Nelson has a very long history, and Pinellas, where again, I think Nelson’s history on the ballot helped him out.    Gillum did slightly better in Duval, and in Orange/Osceola counties – which is likely a testament to the work Scott did after Hurricane Maria as much as anything.   But overall, the margins are similar.

Also, there are a lot of ways we could view this race, but since it is my blog, I am going to start by breaking it down as GOP base markets (Pensacola, Panama City, Jacksonville, and Fort Myers) versus DEM base markets (Tallahassee, Gainesville, West Palm, and Miami) – and the I-4 markets (Orlando & Tampa). 

Let me start at where I ended a piece in early October, my view at the time on the race: While I am currently bullish about my party’s chances, both from the standpoint of mood, and Dem opportunities for growth, if DeSantis and Scott are able to replicate Trump like share of the vote in the large suburban and exurban counties around I-4, things could get very tight, very quickly.” 

So, let’s break that statement into two buckets – Dem opportunities for growth – and the large suburban and exurban counties around I-4, starting with the former first.

If we go back and look at the previous two Governor’s races, the two parties pretty much matched the other in their base counties.  In 2010, Democrats won their base markets by 33,840 more than the GOP did, and in 2014, the margin was 14,360.   Fast forward to 2018, and Andrew Gillum won the base Democratic markets by 109,809 more votes than DeSantis won the base Republican markets. 

How did Gillum expand the Democratic advantage in the base markets?   One word:  Miami.

In fact, if you add up the three non-Miami Democratic markets, Gillum and Crist won them by almost the same margin (102,390 for Gillum, 102,698 for Crist), but Gillum won the Miami market by some 143K more votes than Crist did.

In fact, the same dynamic played out for DeSantis– outside of Fort Myers, he won the other base Republican markets by almost exactly what Scott won them by, but because Gillum was able keep him in check elsewhere, particularly in the Jacksonville market, DeSantis’ growth in the GOP market came nowhere near matching Gillum’s in his. 

*There is an argument that DeSantis did a good job keeping the Miami, particularly Dade margins, from being even larger, but that is for a longer look at Miami.

So how did DeSantis win?  The second bucket of this exercise, which is the same way Trump did.  Literally, exactly the same way.

As I sat in under the press tent at the Gillum Election Night Party, trying to make my rain-soaked laptop work, I felt the same sense of doom as I felt in 2016 -- twice.  First when the Pasco absentees came in, then around 7:50pm, when it became clear the Gillum/Nelson leads were not good enough to overcome the likely GOP advantage from the counties that would report at 8:00 EST, when the Central Time Zone polls closed.  The exurban counties in I-4 had done it again.

We know that Democrats grew their margins – significantly, in the places where my team historically runs up the score – and more than the GOP did in their traditional markets. But then we get to I-4.  Gillum lost the I-4 markets by 146K votes, or just under 70K more votes than Crist did.  Nelson lost the markets by slightly less – 138K or so votes.  Those margins more than offset the gains elsewhere and added up to a loss. 

But that is not the end of the story – and here is where the 2016 comparison sets in – just like Clinton, Gillum ran up new high-water mark margins in the urban areas, particularly around Orlando.  In total, Gillum won the urban counties of I-4 by 120,000 more votes than Crist did – and this is despite winning Pinellas County by some 30,000 votes less than Crist (the Crist margins in Pinellas were much more about Crist than they were about Democratic party performance).  However, where Crist lost the suburban and exurban counties around I-4 by about 157K votes, Gillum lost them by 355K.

Or more simply:  the counties around the urban I-4 counties delivered DeSantis with more of an increased margin than the Miami media market delivered for Gillum.  In a race where most everything else stayed the same – that made the difference.

Here is another way to look at it:

If you look at 2014, the Scott margin of victory pretty much matched his win in the Orlando media market, with his margins from North Florida, plus Tampa and Fort Myers almost exactly balancing out what Crist won South Florida.  Guess what happened again?  Literally the exact same thing. 

For Democrats, the Orlando math is a good way to highlight the problem – it is very hard to win a pure turnout fight.  When you look at 2018, the margins in urban Orlando were spectacular, and just like on Election Night in 2008, looked almost unmatchable.  But the problem for Democrats – the GOP margins everywhere else grew too, in some cases by a lot more, and cancelled out the gains.   For example, Gillum won Orange County by 85K more votes than Crist did – but if you add up the six republican counties in the Orlando market, he lost them by about 84K more votes than Crist.

Broadening out to the two markets combined, just like Clinton, Gillum won the I-4 urban counties by more votes than Obama did in 2012.  But whereas Clinton lost the non-urban I-4 counties by nearly 450K, and Gillum by 355K, Obama lost them by under 220K votes.  In fact, to prove the point of how important it is for Democrats to be more competitive in these counties, had Obama in 2012 lost the non-urban I-4 counties by as much as Gillum, we would have lost Florida that year.  

Nothing demonstrates this shift more than Volusia County, a place Obama won in 2008, lost by about 3K votes in 2012, and where both Nelson and Gillum lost by more than 22K votes – or the three counties north of Tampa (Pasco, Hernando, and Citrus), where Obama lost by 23K votes in 2008, 37K votes in 2012, but both Gillum and Nelson lost by more than 75K votes.

And here is where my frustration sets in – and not with either campaign – but the mindset generally of approaching the state, this idea that Florida can be won entirely in a few corners, or that Florida is just a turnout state.  Look at the difference between the Ag Commissioner’s race, and the Governor’s race.  In the Ag Commissioner’s race, the Democratic candidate, Nikki Fried, had better margins in 38 counties compared to the Governor’s race, including 33 that both candidates lost.  In other words, losing by a little less in a lot of places added up to the difference between winning by a little, and losing by a little.

This piece is not a criticism of any campaign.  God knows I’ve been on enough conference calls to know campaigns must make hard decisions, and often those decisions, particularly the ones Nelson often faced against a far better financed opponent, were a choice between two sub-prime options -- and that all of those decisions look easier in hindsight, and when you are on the outside.   And to their credit – especially Mayor Gillum, the campaigns did spend a lot of time in some of these communities. 

Instead, I lay this out to answer the question I’ve gotten a lot this week: what can we do differently?  From my view, the answer today is the same as it was when I got this question after Bush beat Kerry in Florida – you can’t take on these kinds of margins in Republican counties and hope to make it up elsewhere.  Sure, you might get a win here or there, but over time, it is just a losing proposition.  And here’s the thing – the same math exists in 2020, and in 2022 – in other words, getting right up to the edge of winning, unless my side is willing to drive a conversation about the math problem that we have. 

There are a lot of things to fix, but step one is voter registration.  In 2008, when our campaign shut down, Democrats had a nearly 700K advantage in voter registration.  Today it is just over 250K.  We need a sustained and permanent voter registration effort, which by its own nature, will keep us to remain engaged in the types of emerging communities where we are growing – as well force us to engage in many communities where we have not.   

I have a few pieces I will work on over the holidays – the aforementioned one on why I think the public polling continues to be a mess in Florida, as well as some more specific deep dives into individual markets. 

Thanks again for reading.  I really do appreciate it.  Happy Thanksgiving.

PS -- At least the Noles stomped the Gators in basketball on Election Night.  :-)

Tuesday
Nov062018

Like the Jaguars Season, the Election is Nearly Over.

To:      Fellow Americans Who Don’t Sleep

From:  Steve Schale, Tired Florida Man

Re:      You know you are going to miss these.

*5 hours until the polls open

*17 hours until the first of 23 phone calls between 7 and 7:05PM asking what I think.

*19 hours until FSU basketball season tips off against the boys from Hogtown, East Florida Seminary

*3 days until we the media forgets this election and goes all-in on 2020.

*5 days until the Jaguars lose again.

*6 days until the Vet Fest 5k in Tallahassee, as I need to burn off the last few week’s diet.

*708 days until someone else writes the first Florida memo of the 2020 cycle

*728 days until the Election Day 2020.

This is the second memo of the last day, and for ease of life, I am going to copy and paste the first part, so you have it, sans the jokes about Democrats voting like Blake Bortles throws touchdowns to the opposing teams – virtually everywhere, and often, and get into this thing.  There is really no reason to repeat it, anymore than there is a reason for Bortles to throw two interceptions in a single half.

Also, I would be remiss if I didn’t mention that I am writing this after a day that began at 5:30 AM, has included multiple GOTV shifts, and about 400 miles in the car.  I can attest for the record that my entire diet has consisted of granola bars, candy I rifled from campaign offices, some tacos, a slice of ? old pizza, and caffeine in a variety of forms, so forgive me in advance for the bad jokes, sarcasm, and etc, that you will find.

As of this evening, Floridians have cast:  5,111,452 votes

Democrats:  2,074,400 (40,58%)

Republicans: 2,049,877 (40.105)

NPA:  987,175 (19.3%)

Total Democratic margin:  24,523 (+0.48)

For comparison, yesterday we were at just over 4.8m voters and GOP +24,689

Friday we were at 4.46m voters, and GOP at 56,902.

Despite many counties not opening on Sunday (Sunday was optional for counties in Florida), nearly as many people voted on Sunday as Saturday, when the polls were open statewide.  The nearly 50,000 voter net gain the Democrats had was twice as big as the last Sunday in 2014, and nearly matched the 55,000-net gain of the final Sunday in 2016. 

Already, 2 million more people have voted as voted before the Election Day in 2014.  The Republicans went into that Election Day with a 97,000-voter lead, or roughly a 3% advantage.  One way to think about this election in comparison – even with the Election Day advantage Republicans had in 2014 and will have in 2018, I have no doubt Charlie Crist would have won in 2014 with the electorate where it is today.  In terms of the partisan difference in the electorate, Democrats start Tuesday morning in a net of about 120,000 voter better position than four years ago.   

My Republican friends like to point out that the electorate in 2016 was very similar, and I agree, it is.  I also agree to their point that the electorate tomorrow is likely to be more Republican than Democratic, possibly by as much as 150,000-200,000 votes (though I think the latter is unlikely given that Democrats will still have infrequent voters turning out to cancel some of their advantage).  But one thing appears in all polling to be different:  crossover voters, and independents, both who broke late to Trump in 2016, and appear to be breaking for Gillum and Nelson.

To give a quick example of what this means, even with the large surge of GOP voters in 2016, the defining feature of Trump’s win was late deciding independents, who overall gave Trump a four-point edge in the exit poll – a number that based on polling, seems quite plausible.  If that same segment of the electorate had given Hillary Clinton a four-point edge, she would have won Florida by a margin like Barack Obama in 2012, who, yup you guessed it, won independents.

Republicans also will point out, and if they won’t, I will for them, that Crist won independents, and lost.  This is also true.  But again, if you go back three paragraphs, you will see that the electorate was substantially more Republican.  There is a point at which the GOP turnout advantage could be so big that Gillum and Nelson would have to win independents by something so Herculean, akin to the scale of imaging Blake Bortles throwing 10 passes without bouncing one off an offensive lineman’s helmet, that the math doesn’t work.  But this is far from that kind of model.  More on this later.

Couple of other cut and paste facts from the morning memo:

The share of the electorate that is Black at 13.6%, which means Black voters are turning out at a higher rate than their share of voter registration (13.2%).  Hispanic is up to 13%, which still lags its registration, but it is moving up.  Overall, the electorate that is about 68% white.  Several of you have asked why I keep mentioning this number, and it is simple:  Democrats in the last few cycles have struggled with white voters, so the greater the percentage of the electorate that is diverse, from a math perspective, the lower share of the white vote required.    It is not likely that tomorrow will see the electorate get more diverse, nor do I think much will happen to make it less diverse. 

The Black turnout is driven across all groups:  2014 voters, 2016 voters, new voters, etc. – consistent across all targets.  The Hispanic and NPA turnout is being driven by newer voters.  The percentage of voters who did not vote in 2014 is up to 33%.  Within that universe, it is more Democratic -- Dems have about a 110K voter lead among the expansion universe, and it is more Hispanic -- over 18%, than the electorate at large.  It is also more NPA, with 26% of expansion voters not registering with either party, which makes sense because it is also younger -- nearly 23% under the age of 34.  

So what happens tomorrow?  Republicans show up, and infrequent voters continue at some level, maybe not as high as in early vote, but still at a steady click  So what does this mean?  I have been pretty set on about 7.25 million for turnout for most of two weeks.  I don’t see it going lower than this, and while it may be a bit higher, I don’t know that it is a lot higher, mainly because while we have seen surge, a lot of the surge is really just convenience voting.  As I told a reporter or two today, the most remarkable thing is just how normal this electorate looks – just with more volume. 

So let’s do some quick math (I NEVER SAID THERE WOULD BE NO MATH).

Dem edge today is roughly 25,000.  It could go up or down a bit by morning, given the absentee ballots that came in today, but isn’t likely to change much.

Let’s say, worst case scenario for Democrats, virtually every likely GOP target shows up tomorrow, and they win the day by 200,000 voters.  Again, I think reasonably, this number is closer to 160,000, but for sake of this model, let’s say it is 200,000,  And let’s say that to get there, turnout is about 1.5-1.6 million for these super voters.  At this level, based on how infrequent votes have been voting another 400,000 infrequent voters are likely to show up and at the current Democratic pace, Democrats probably net minimum of 30,000 voters. 

200,000 GOP edge on EDay

25,000  Current Dem Edge

30,000 Dem infrequent edge.

= 145,000 final GOP edge in turnout. 

At 145,000, this means Republicans would have a 2 point edge in the share of the electorate, and as I showed the other day, there are a variety of ways, with very little crossover, and a very reasonable NPA win, that the Democrats win. 

More on this in a bit.

Let’s go back to an exercise from one of the first memos – how do both parties win?  Well, for Democrats, run up the score in a few places, and keep it between the ditches everywhere else.

The big places for us, Broward, Dade, Palm Beach, Orange, and to lesser extent, Hillsborough.  Let’s start with Broward, where Democrats entered Eday in 2014 with a 100,000 voter lead – today, that number is 165,000, and arguably even more important, the county will turnout a bigger share of the electorate.  In 2014, the total Broward turnout was 44%, and as of just today, it is 40%.  It will exceed its 2014 turnout, quite possibly by a significant number – and that is just volume, volume that adds up in the Democratic column.  Turnout in Broward has made Florida Man's driving on I-95 seem peaceful and tame.

In fact, there are 8 counties that are within 10% points of reaching their 2014 turnout percentages – in other words, the counties that are performing the best relative to their 2014 turnout, and of those 8, four are significant Democratic base counties:  Dade, Broward, Orange, and Hillsborough.   Every single one of these counties will be a bigger chunk of the electorate than 2014, and everyone of them will deliver large majorities for Gillum and Nelson. A fifth county, Osceola, is also in this category and is a Democratic base county.

If you take these five counties, currently the Democratic advantage in turnout is 313,584 voters.  In 2014, the advantage was 134,439 voters – and even if you just factor in the higher turnout numbers, these three counties are still about 90,000 voters ahead of where they were four years ago.   That is not insignificant.

Let’s also look at the two “play defense by playing offense” counties that I mentioned in one of the first of these memos, which feels like 18 months ago now.  In two large counties in North Florida, Clinton outperformed Crist:  Escambia (Pensacola), and Duval – also known as DUUUUVAL, which is Jacksonville, home of Blake Bortles’ fumbles.  The key for the ticket in those two places will be increasing African American participation.  These are also two communities that both DeSantis and Scott will want to look more like they did for Scott in 2014 than they did for Trump in 2016

So what is going on there?  Well in Duval, in 2014, Republicans had about a 3% lead in party share entering Election Day – in 2016, the Dems had about a 1.4% lead, and today, the Democratic advantage is over 3, or roughly 12,000 voters. Democrats are not only denying the Republicans a large margin in a county that Scott won by 34,000 votes (+13%), but they might just win the whole darn place.    I have wanted to win DUVAL since taking the head coaching job for Florida Obama in 2008 – we almost got there in 08, Clinton got closer in 16, and I am going to go bold, and throw a Blake Bortles deep ball into triple coverage and say Gillum go up and grab that pass, and bring DUVAL home in 2018.  As for Escambia, the party advantage was 23% in 2014.  Today it is 17%.  Chipping away at the margin.  That’s how Democrats win. DUUUUUUVAL.

Secondly, Republicans run up the score in a handful of counties, and win a few dozen by decent vote total, to counter balance the growth the Democratic ticket is likely to see in the urban counties.  For DeSantis and Scott, their path lies with the dozen or counties where Trump (2016) outperformed Rick Scott (2014).  While the GOP ticket is unlikely to see the same kind of raw vote margins Trump won in these counties, they will want the final percentage spread to look more like Trump than like Scott.  Most of these counties are in the I-4 corridor:

Tampa market:

Hernando (Scott 47.9%  +2,013 votes - Trump 62.9%, +27,211 votes)

Citrus (Scott 53.7%, +8,881 – Trump 68.3%, +31,667)

Pasco (Scott 46.8%, +2,859 – Trump 58.9% +51,967)

Pinellas (Scott 41% -39,659 – Trump 48.6%, +5,551)

Sarasota (Scott 48.7%, +4,972 – Trump 54.3%, +26,541)

Manatee (Scott 51.7%, +12,356 – Trump 57.0% +30,647)

So how does it look there?

Again, keep in mind the goal in these counties is to change the math, like what the Democrats had done in their base counties.  In 2014, the Republicans went into election day – and today, their margin is 8%, for a net lead of 58,107, which is nearly the same percentage margin as 2014, and a net gain in voters of just about 15,000.  Sure, tomorrow could blow up here, but what has been keeping the GOP markets down isn’t lack of GOP enthusiasm, it is Democrats in these places are voting.  In fact, in Pinellas, Sarasota, and Manatee, Democratic voter turnout rates match, or even exceed the Republicans. 

Orlando and South

Marion (Scott 55.3%, +19,869 – Trump 61.7%, +45,806)

Volusia (Scott 48.8%, +6,434 – Trump 54.8%, +33,937)

Charlotte – Ft Myers DMA (Scott 52.5%, +8,273 – Trump 62.5%, +26,781)

Martin – West Palm DMA (Scott 55.3%, +9,220 – Trump 62.0%, 23,091)

Just like above, in places where Trump blew up the numbers, we are seeing flatter growth – Republicans leading these four counties in early vote in 2014 by 15%, or about 33K votes, and today leading by 14.3%, or about 48K votes (with 100,000 more votes cast).

In fact, generally, while Democratic counties are quickly reaching their 2014 turnout rates, Republican counties are trailing;  25 counties the GOP traditionally win are more than 20% behind their 2014 final turnout percentage, and 40 total are more than 15% behind.  They are simply not getting enough volume – so far.  This will change tomorrow, but there is a lot of catching up to do to get to the kind of turnout the GOP saw in 2014 in contrast to the Democrats that cycle.

Couple of other quick observations – right now, the Miami and Orlando media markets are ahead of their projected share of vote, while most of North Florida is behind.  This should level out tomorrow, though Miami and Orlando will both be a bigger share of the electorate in 2018 than 2014.  

So here is where I am.  While I don’t discount the GOP edge on election day, I do think if the NPA and crossover vote is doing anything close to what pollsters are finding, that edge gets eliminated quickly.  In fact, a 9-point Gillum edge in NPA eliminates a 2 point GOP edge in turnout without having to win a single more GOP vote than DeSantis wins among Democrats – and I think Gillum wins more Republicans than DeSantis wins Democrats.

I know a lot of the polling of late has shown it stretching to a 4-5 – and even 7-point race, but honestly, I’ll believe it when I see it.  That being said, I will be less surprised if he wins by that margin than I will be if he loses – and honestly, if James Comey hadn’t sent a memo, and I hadn’t lived with watching the disaster for my party that was Election Day in Florida in 2016, I probably would have no hesitation in saying where this plane lands.  I think 2016 is the thing that gives most people pause, when in reality, the vast majority of signs pont to Gillum and Nelson.

That doesn’t mean a win is a certainty – if GOP really shows up, Dems turnout stalls, and white independents crash tomorrow, that could be a bad combination.  For Trump, it took all three of these things happening to win, and while I expect one will happen, the other two are a lot  less certain.  Moreover, the polling seems to show the race consolidating towards both Gillum and Nelson, where as in 2016, you could feel the race slide towards Trump late.    Again, I put the odds of DeSantis winning at lower than the odds of Gillum winning by a comfortable margin – but that doesn’t mean it can’t happen.  So if you are on my team, GOTV.  If you are not, I hear there are some Jaguars re-runs on Netflix.  

Tomorrow I will be watching turnout in the counties north of Tampa, and in Pinellas.  This where the real Trump surge happened on Election Day in 2018, and where if we see a repeat, we will see signs during the day.

At 7:00, I will check Pasco.  I can’t say I have a specific trigger on Pasco, other than kind of like when Bortles chucks one up under pressure with no specific offensive player in mind, I more go by feel, but anything close to parity is a good sign for Democrats.

Pinellas will report quickly and will report most of their vote at once.  Up, and I will feel good. Down, and I’ll settle in.  DUVAL is also early, and I want to see my side up.   Dade early reports quickly, and Crist margin was 50,000 in early/vbm in 2014 – I want to see north of this. 

I’ll start checking the urban/suburban counties around Orlando, and again, mostly just want to see margins in line with, or below 2014, as well as Orange and Osceola, and will want to see margins closer to 2016 numbers than to the Crist 2014 margins.

Then heading to 8 – what does the total margin look like?  In 2014, Scott won the Central Time Zone markets by 140,000 votes – so are the Dems clear by at least that?  If so, given late reporting southeast Florida counties, they are probably fine.  If not, well, you can probably turn on basketball.

Florida is pretty fast reporting, except for Palm Beach, which I think is still counting ballots from 2000, so unless this thing is really close, I think we will have a good sense of where this is headed at a reasonable hour.

Real quick, I want to thank a few people.

First, my friend Dan Newman, who for two cycles has been an invaluable resource to these pieces.

I also want to thank the other vote counters – Dan Smith, Caputo, Wiggins, Tyson, and Wayne Bertsch – I appreciate bouncing things off each other – and the thoughts you all share with me, as well as – and I will protect the innocent, some of the national guys who help me check myself.

And to everyone who reads these things, truly, thank you. As I think I have said in the past, in my younger years, I struggled with reading and math comprehension, so I made excel sheets, and wrote out concepts to help me think things out.  To this day, I think by writing. I started writing these pieces as more internal objects in my campaign days, to help me process decisions – and as I migrated out of day to day campaign stuff (getting old sucks), someone suggested sharing these to provide maybe a little insight into how at least one old hack thinks about the state and its trends.    Like everyone else who spends time in this data, it is a time-consuming labor – but a labor of love, and I really do appreciate you reading.

To all the candidates who ran, congratulations, you have more guts than me, who has decided not to run a few times in his life.  Your willingness to step into the Arena is admirable.  In the words of Teddy Roosevelt: "It is not the critic who counts; not the man who points out how the strong man stumbles, or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena...who spends himself in a worthy cause; who at the best knows in the end the triumph of high achievement, and who at the worst, if he fails, at least fails while daring greatly, so that his place shall never be with those cold and timid souls who neither know victory nor defeat."

And lastly, while I have a lot of folks I am rooting for, I want to specifically wish Andrew Gillum good luck tomorrow.  While I initially was on another team, I’ve known Andrew for well over a decade, and I have nothing but genuine admiration for what he’s achieved in his life. His improbable story is one that can give hope to so many for whom hope is a shrinking commodity. I also think back to my first two years working in the legislature, when divided government led to some incredible achievements;  Bright Futures scholarships, KidCare expansion, School construction, and many more, and I believe Andrew will bring a much needed, fresh voice to the process, and help drive some big ideas, and big conversations.  I’d be proud to call my friend my Governor.  Bring it home sir.

PS- Congrats to Miami Dade College on turning out the most early votes, and winning the early voting Team Democracy State Championship.  With a top five finish, look for UCF to declare themselves the Early Voting National Champions.   Thanks to all who voted,  you quite literally help save democracy.

For the 12 of you who read my other posts, see you soon.  For the other 38, see you all in 2020, God willing.

Monday
Nov052018

One more day

To:      Anyone still standing

From:  Steve

Re:      DUUUUUVAL

Date:   November 5, 2018

I greet you on this Monday morning, rejoicing in the fact the Jaguars did not lose yesterday.   The pride this brought to the good people of DUVAL county led to the largest early voting day of the cycle, as nearly 20,000 proud residents of old Cowford came to the polls.   Really proud of my city.

Before we get started, there are a few notes -- this will be the first of two notes -- the second will come late tonight or in the morning, for two reasons:  at the time I am writing this, I haven't had a chance to get access a deep look at the voter file including yesterday's data -- and since I am doing some GOTV volunteering today, it will be tonight before I get another chance.   And secondly, my computer decided today would be a great day to go to the go to laptop heaven, so I will be taking a second run at this stuff tonight on a new laptop.  

Apropos to nothing, what is interesting is my laptops all die on a two year cycle, though God has granted me passage in all previous years to the week or so after the election.  This year, it seems we fell a day short. 

Yesterday was a pretty remarkable day.  Roughly 275K Floridians voted, bringing our total to just under 5.1 million votes.  I suspect that by the time we get through the VBM ballots that will come in today, plus the early vote from the counties in the Panhandle impacted by Michael, we will get to 5.2 million before the election.  This is roughly 40% statewide turnout.  Just nuts.

Democrats had a big weekend -- starting the weekend down just under 57,000 votes, and leave the weekend up about 24,000 votes.  Yesterday was particularly impressive, with the Democrats seeing a net gain of almost 53,000 votes, which is roughly on par with the final day of early voting in 2016.    Democrats were going to the polls like Blake Bortles throws touchdowns to opposing teams -- everywhere, and with great frequency.  

So here is where we are:

Total votes:  5,094,645

Democrats:  2,067,856

Republicans: 2,043,167

NPA:  983,622

Total Democratic margin:  24,689 (+0.5)

One thing on these numbers -- several of you have pointed out that you have seen slightly different numbers day to day, and this is possible.  I pull my data from the state site around 7:30 AM every morning, and counties frequently will update data from the previous day throughout the day, changing the counts slightly.  As I have a job, I view a day as between the last time I put these things in my excel spreadsheet, and when I put the new data in said spreadsheet.  My blog, my process.  :-)

For comparison, yesterday we were at just over 4.8m voters and GOP +24,689

Friday we were at 4.46m voters, and GOP at 56,902.

And to give a sense of the volume, we started Monday at 2.72 million (GOP +59K), meaning that in the last seven days, we've seen 2.4 million votes.  That is stout.

According to my friend josh Geise, who has a functioning laptop (Congrats bro), yesterday was very diverse in terms of turn out, with the Black vote making up 26% of voters, and overall non-white landing at 53%. 

That would put the share of the electorate that is Black at 13.6%, which means Black voters are turning out at a higher rate than their share of voter registration (13.2%).  Hispanic is up to 13%, which still lags its registration.  Overall, the electorate that is about 68% white.  Several of you have asked why I keep mentioning this number, and it is simple:  Democrats in the last few cycles have struggled with white voters, so the greater the percentage of the electorate that is diverse, from a math perspective, the lower share of the white vote required.    From a democracy standpoint, it is also healthy, as we live in a state that is 47-48% made up by residents in America considered to be ethnic minorities.  The electorate should more closely reflect this in a representative democracy, and this one is headed in the right path.

Turnout is way up, which is fantastic, but it is important know this isn't all organic growth. Nearly a million voters so far cast a ballot on Election Day in 2014, and those numbers are basically dead even between the two parties.

The Dems advantage is driven by an expansion from the 2014 electorate.  The percentage of voters who did not vote in 2014 is up to 33%.  Now keep in mind, Florida is an older state in terms of population, and as my friend Dan Smith from UF likes to remind me, it is an older population that refreshes itself from mortality rates due to people retiring here.  So Florida does have significant turnover from cycle to cycle -- but 33% growth is well beyond the norm.  Within that universe, it is more Democratic -- Dems have about a 110K voter lead among the expansion universe, and it is more Hispanic -- over 18%, than the electorate at large.  It is also more NPA, with 26% of expansion voters not registering with either party, which makes sense because it is also younger -- nearly 23% under the age of 34.   

Looking to Tuesday, there are about 2.7 million voters who voted in either the 2014 general election or the 2018 primary, but haven't voted yet - with the vast majority of those being 2014 voters.  If you break this down into tiers -- there are 1.7 million voters who voted in the 2014 general election on Election Day who have yet to vote - and the GOP has about a 160K voter edge. If you take all the 2014 voters, the GOP has about a 200K voter advantage, and the overall universe grows to 2.4 million.  Adding in the 2018 primary voters doesn't really change anything.

This being said, even the worst case scenario for Democrats has this electorate landing at about +2 Republican, which daily readers of this blog will remember, is about where I thought it would land last week as a worst case scenario.  More likely, if the surge voter numbers continue at some level tomorrow, it will be more like +1 or +1.5, and in this scenario, Andrew Gillum and Bill Nelson will win. 

I do think after the weekend that turnout ends up north of 55%.  60% would be 2.8 million votes, which honestly, is quite possible. I want to spend some more time with this tonight, but regardless, we are well north of where we thought we would be.

In the note I write later tonight -- more like late tonight, I will send along my cheat sheet to how I will follow the election, but here's the deal, if the polling is right, and Gillum and Nelson win by 4 points or more, we will know pretty early, because as much as 50-60% of the state is likely to report before 7:30, 8:00 EST at the latest.  Do keep in mind, as I will remind you all tonight, there is a significant GOP vote in the central time zone, so the early numbers will get more Republican as the night goes on, but I will help try to explain this on twitter.

With that, more later.  

 

Sunday
Nov042018

Almost there - 2 Days Florida

To:       Fans of Daylight Savings Time

From:   Steve Schale

Re:       TWO MORE DAYS

As a longtime hack, I always saw “falling back’ as a biennial opportunity to get an extra hour of sleep.  But that is probably where the upside of going back to standard time ends.   If I ever run for President, Make Evenings Light Again will be a key part of my platform.

Florida Man and Florida Woman voted yesterday like there it was BOGO day on flip-flops at the early voting site. 

When it is all in (several counties have not reported yet), total in-person early voting will be around 300,000 votes.  With the VBM ballots that came in, the day will come close to (but not reach), the last Saturday in 2016.   There is no question in my mind that Florida will cast more than five million votes before Election Day, and there is no question in my mind that the margin between the two parties will be less than 1%, and probably more like 0.5%. 

Like yesterday, this memo will be shorter for two reasons:  I want to get some more volunteering in for GOTV, and I will take a much deeper look at everything in tomorrow’s memo.   Also, keep in mind these totals are not perfect – there are about 10 counties, a couple of which are mid-sized, that did not report at 7:30 AM.  

So, here we go: 4,817,062 votes have been cast.

Republicans: 1,964,364

Democrats: 1,936,328

NPA/Minor: 915,370

Republican edge is 28,036 (+0.58%)

The GOP voter advantage has floated between the mid-50s and 70 since the start of in-person early voting, but this is the first time we’ve seen it really move.   In 2014, Democrats won in-person early voting on the final Sunday by 25K votes, and in 2016, the number was 55K, and I do think the Democrats win the day by enough to take a lead.   When the straggling vote by mail ballots come in, given the GOP edge there, the number may pop up close to tied.  Welcome to Florida.

The GOP went into Election Day with about a 90K voter edge in 2014, or about 3%

The Dems went into Election Day, also with about a 90K voter in 2016, which was about 1.3%

Crist won the NPA vote by 6, not enough to overcome the GOP early and EDay advantage.

Trump won the NPA vote by 4, enough combined with a strong EDay advantage to win.

The GOP will almost surely win the turnout battle on EDay – and this isn’t because of anything other than they simply have more certain voters who wait until the final day to vote.  So like everything in Florida, this will come down to who wins more crossover, and who wins the NPAs.   We are, as I often point out, both a turnout and a persuasion state. 

Statewide turnout is now 36.2% - in other words, 2.7% of Florida registered voters cast a ballot yesterday.  Among Republicans, it is 42%, among Democrats, it is 39.2%, and among NPA, it’s up to 25.1%.   Yesterday alone, 3.2% of all registered Democrats voted. 

As we saw in both 2014 and 2016, the NPA share is growing as we get closer to Election Day.  On Monday, it was 17.7%.  Friday it was 18.4%.  Yesterday it is 18.7%. Today it is 19.  It will get over 20, and for Gillum who most polling shows is winning the NPAs, this is how you make up a partisan difference – and the bigger the NPA share, the more his advantage there helps.  How important was NPA to Trump?  If you hold everything else constant, he won the NPA’s by about 90K votes – if they had split the NPA vote, the race would have been in a recount.  If she won the NPAs by four?  She would have won Florida.

The last two days also show how the electorate is getting more diverse. After Wednesday, the electorate was 71% white, and yesterday it is 70%, and this morning it was 69.  Yesterday, non-Hispanic whites made up 54% of voters, with African Americans and Caribbean voters making up 19%, and Hispanics another 14.  The Hispanic number continues to lag a bit, but it is headed in the right direction. Right now, Hispanics are 12.6% of registered voters, and I think this ends up in the low 13s, based on trendlines.   There is a possibility Black voters will exceed their share of registered voters in the early vote – right now, it is 12.96%, and registration is 13.2%. 

Keep one thing in mind, Hispanics are under-represented by the registered voters, namely, not all Hispanics self-select Hispanic on their voter-registration, and some have been registered long enough that it was an option in the county where they live.  In this election, I don’t think this delta is significant, but just worth nothing.

In-person early vote has exceeded vote by mail, which happened in 2016, but not 14.

51% of the vote is in-person early, and Democrats have a 37,585-voter advantage.  49% is vote by mail, and the GOP advantage is 65,621

GOP return rates of VBM ballots continues to out-pace my team,   Republicans have returned 72.7%, Democrats 65.5%, and NPA 60.1%.  The statewide return rate is 67.3%, which is down from 2014. 

Dems have 114K more unreturned...excuse me while I go bang my head against a wall.

OK, I am back. 

Two last observations, and then off to work.

One – DUVAL – wow.  Setting records every day, voting like Blake Bortles played against the Steelers.  I am not going to predict that Gillum and Nelson win DUVAL yet (I might tomorrow), but long-time readers know I am a big fan as a Democrat of playing defense by playing offense in this county, there is a chance that Democrats will go into Election Day with a 10,000-voter advantage.  DUVAL is a place that can provide huge GOP margins, but that won’t happen on Tuesday.

Right now, the regional distribution of votes looks very good for Democrats.  Miami is almost 20% (naturally lands 17-18%), and Orlando is also up.  But it also looks a lot like 2016 did at this point, with a few places that helped my home team higher than normal.  What happened on Election Day was a lot of those counties flatlined, and GOP ones took off – that and NPAs and other swing voters broke to Trump.

I will get more into this tomorrow, but for Democrats, this is rounding into nice shape.  There is very little doubt in my mind that when we go back and look at actual votes, Nelson and Gillum will start Tuesday ahead of their opponents.  How much ahead?  We can’t tell from this.  Is it going to be enough?  Well, that is why if you want them to win, you should now close your laptop, or put the phone down, and go pull a shift.  America’s team, the Jaguars has a bye week, so there is no excuse.

As I used to tell our crew in Obamaland – and this really applies to both sides of this race  it is now in your hands.  Don’t be a Blake Bortles and fumble it away now.

Until then, if you are a journalist in Florida covering an election, watch out, because Florida Man and Florida Woman are coming to vote, and if history says anything, you do not want to get in the way of determined Florida Man or Florida Woman, because you never know when they are packing a small alligator, or spatula.  So be safe out there.

Saturday
Nov032018

3 Days Out - Florida is Careening to the Finish

To:       Fans of Number 3 (Dale Earnhardt of course)

From:   Steve Schale

Re:       3 days to go!!!

Today we celebrate Dale Earnhardt day, as we stand 3 days from an election cycle, that just like the last lap of Talladega, is coming to a close with cars all over the track, a loose tire landing in the stands, and someone upside down on their roof sliding through the infield.  

Yesterday was the largest in-person early voting day, and the overall largest day (ev+vbm) of the cycle.  Right now, the total stands at roughly 375,000, but when the counties that have not reported yet pop into the system, it should be very close to, if not exceed, 400,000.

And yesterday is an instructive day for showing how Florida works.  Democrats absolutely blew it out in a handful of counties that always go Democratic, and Republicans, while not seeing the same kind of margins, did well in a bunch of counties that they do well in.  The result for the day:  nearly a tie.  I’ll walk through some of that later.

Also, this memo will be shorter for two reasons:  I want to get some volunteering in for GOTV, and I want to spend some time working on my Monday memo, which will lay out where we are at the end of early vote, and the scenarios that can happen on Tuesday.

So, here we go: 4,462,042 total votes

Republicans: 1,835,373

Democrats: 1,778,373

NPA/Minor: 832,198

Republican edge is 56,902 (+1.3%)

To give a sense of just how stable Florida is:  look at Monday: it was 2,726,392 (+2.2%) with the GOP holding an edge of 59,048.  1.7 million votes over 6 days, and t was basically a tie for the week.  That’s just how the place works.

At this point, we are going to go to Election Day with 1.9 to 2 million more votes than voted early in 2014.   As a percentage margin, it is better than 2014, worse than 2016, but above all, I don’t think either comparison really works at this point.

Statewide turnout is now 33.5% - in other words, 2.9% of Florida registered voters cast a ballot yesterday.  Among Republicans, it is 39.2%, among Democrats, it is 36%, and among NPA, it’s up to 22.9%.

As we saw in both 2014 and 2016, the NPA share is growing as we get closer to Election Day.  On Monday, it was 17.7%.  Yesterday it was 18.4%.  Today it is 18.7%.  It will get over 20 when this is all said and done – and as I said yesterday, that is an important point for any candidate whose path to win includes NPAs – the bigger their share, the more that vote means in the final equation.

The last two days also show how the electorate is getting more diverse. After Wednesday, the electorate was 71% white, and today it is 70%, and continuing to track into the upper 60s.  What caused that drop?  In the last two days, the electorate has been 15.5% Black, and 13.7% Hispanic.  It is likely this will continue into the last two days of early voting.  Not surprisingly, the share of people who did not vote in 2014 continues to rise.  

And continuing the trend from earlier in the week;  cannibalization (which probably isn't a real word), or the action of parties moving their Election Day vote into early voting, thus making early voting look more robust than it is.  Right now, over 18% of the electorate are cannibals – ok, it is Florida, so be clear, I mean early voting cannibals.  About 25,000 more Republicans fall into this category, but I don’t think that is overly significant.   But this overall stat is important:  while more than 30% of this electorate is new from 2014, the higher turnout is also being driven by convenience voting, which is why I believe the final turnout will be higher than 2014, but not ‘mini-Presidential” in volume, as some have suggested.

But then again, who knows.  People are voting.  That is good.

One other quick thing, to close the loop on Hurricane Michael:  voting patterns have basically returned to normal there.  They are still a tiny bit behind the state, but at this point, there is no evidence that Michael will have any significant impact on the election.  This is very good news for that community, and for democracy at large.

51% of the electorate to date is vote by mail, and the GOP has a 67,540.  Of the 49% of all ballots in that were cast in-person early, Democrats hold a roughly 10,600 ballot lead.  In person early voting will outpace vote by mail this year. 

In total, just over 3.48 million ballots have been requested – again far more than 2014, and more than 2016.  Republicans have returned 70.5%, Democrats 63.2%, and NPA 58.4%.  The statewide return rate is 65%, which is down from 2014.  But there is still time to return your ballot. 

HINT: Return Your Ballot.

Here is what is remaining:

Democratic unreturned ballots: 515,780

Republican unreturned ballots: 399,366

NPA unreturned ballots: 308,109

And, to everyone on twitter who likes to point out that people may have requested a ballot and voted early, and are like “Schale, stop beating us up, our people are voting” – yes, there are Democrats who are requesting ballots and voting in person.  But guess what, there are Republicans, and there are also NPAs doing it.

In total, roughly 150,000 Floridians have, to steal a term, cannibalized their vote by mail ballot, and yes, a few more Democrats have than Republicans.  So, in the sum, there are still A LOT MORE DEMOCRATS with vote by mail ballots sitting under the dog’s tennis ball, or next to that thing you bought at Target but aren’t sure where you are going to put in your house.

Get your ballot. Take it to an early voting site or take it to the polls on Tuesday.  Just don’t leave it sitting under your gym clothes in the backseat of your car.

Last thing for today, I thought today’s numbers were instructive about how Florida works.  As I’ve written in the past, Florida tends to operate like a scale that corrects itself back to balanced.  For as dynamic as the state is, the politics are stable, and quite dug in, which explains why our elections are so close.  In short, Democratic areas and Republican areas tend to balance themselves out, and the state is won or lost on the margins. 

Over the last 6 days, more than1.7 million ballots have come in, and the difference between the two parties is about 2,000 votes. 

Yesterday, out of nearly 400,000 votes, even though Miami, Broward, Orange, and Palm Beach blew it out, the statewide partisan difference was about 1,600, so my twitter engagers ask: ‘Hey man, how did the Republicans turnout out more voters if 125,000 people voted in the big four Democratic counties,”

Well here is how: Democrats yesterday won 18 counties.  Of those 18 counties, 8 are very small, and the total margin in those 8 was 471 voters.  And honestly, we aren’t winning most of these counties.  In these 18 counties, Democrats margin in turnout advantage was roughly 35,000 voters, of which 80% came from the big four counties:  Broward, Miami, Orange, and Palm Beach.

On the flipside, Republicans won 44 (5 haven’t reported) by just over 33,000, but to add up to the same margin as the Democratic big four, you had to go to their top 17 counties.  In other words, they win a lot more places by decent margins, while we win a few by a ton.

So, they cancel themselves out like this:

Broward (Dems +11,303) = Bay, Lee, St. Johns, Collier, and Brevard

Dade (Dems +8.268) = Nassau, Lake, Sarasota, Sumter, Santa Rosa and Okaloosa

Orange (Dems +4,336) = Pasco, Manatee, Charlotte, and Marion

Again, remember, all of this is on the party margins.  If NPAs break hard one way or another, or if there is significant crossover vote, these margins are all impacted.

But generally, who wins or loses will do so, not by where they win, but why they win or lose by everywhere.  It is the little things at the end that determines who wins close elections.

Two last things.

1.  Dear UCF:  You all want people to think you are state champions or national champions or something, but here is the deal:  you are losing to UF, FIU, and FSU/FAMU in terms of early voting totals on campus.  And you have a lot more students than these others.  So until you stop losing to FIU, please stop telling everyone you "want Bama."

And secondly, yesterday was a tough day in our community, as a shooter walked into a yoga studio, killed two and injured others.  A few months back, it was a video game competition in Jacksonville, before that Stoneman Douglas, before that Pulse, and in between that, a whole lot of other shootings that never made the headlines.  During session, i was proud to help some of the parents and students of Parkland pass a significant, and bipartisan school safety bill, but let's be honest, the solution to this isn't hardening everything, and arming everything and everyone.  The new normal isn't normal, nor can we ever let it become so, because when we do, that is when those who seek to do harm win.  if you are reading this, it is because you care about democracy, and here's the thing, people who truly care about the nation can find solutions -- and everyday people -- just like those kids and parents from Parkland -- can force real change.  So yes, we should pray for the victims and their families, but we can all also do something, from small things like just being kinder to each other and creating space for dialogue, to actually sitting down and figuring out how we can slow this epidemic.  

Some more tomorrow, and a lot more Monday.  Thanks again for reading.

Friday
Nov022018

4 Days Out - Florida Man is Crushing This Voting Thing

To:       Fans of Fridays and Dogs

From:   Steve Schale

Date:    Nov 2, 2018

We are on the verge of one of the most exciting weekends each fall:  the bye week – the week where the Jaguars are guaranteed not to lose. 

This year, that occasion coincides with the final weekend of early voting, if there is anyone left to vote.  Florida Man and Florida Woman are crushing this voting thing like it was a Natty Light can run over by a monster truck at the county fair. 

Eclipsing the 4 million vote number, yesterday was the second biggest day in terms of total votes coming in the door, and the biggest day of in-person early voting.  Democrats narrowly won both, leading to about a +5,000-voter day.

Total votes: 4,068,596

Republicans: 1,689,457

Democrats: 1,630,927

NPA/Minor: 749,212

Republican edge is 58,530 (+1.4%)

Yesterday we were at 3.724m votes, with GOP edge of 63,537 (+1.7%)

On Monday it was 2,726,392 (+2.2%) with the GOP holding an edge of 59,048.  So yes, people are voting – but all people are voting.  In other words, about 1.4 million voters have voted, just this week.   This week, the most dangerous place in America is between Florida Man and a voting booth. 

Roughly 1.1 million more voters than this day in 2014 have cast a ballot.  For comparison, in 2014, right at 2.96 million voters had voted, Republicans had about a 125K -ballot lead at this point – and led by around 4.2%.   

Statewide turnout is now 30.6%.  Among Republicans, it is 36.1%, among Democrats, it is 33%, and among NPA, it’s up to 20.5%.

The latter data point is very important – as we saw in 2016, NPA voters are picking up late, and this is important for one big reason:  The bigger the NPA share of the electorate, the less of Democrats need to win them by (as a percentage), because their vote weight becomes bigger.  For example, if there are 7.25 million voters, and Gillum wins NPA by 8.  If NPA equals 18% of all Florida voters, his margin from NPA is about 100,000.  At 21% of all Florida voters, it goes up to roughly 125,000.   Today it is at 18.4, but that has grown from 17.7% on Monday. 

Republicans have now returned 64.3% of all their requested ballots, Democrats 57.2%, and NPA’s 52%.  Statewide, the return rate is 58.9%. 

And therefore, Republicans have a turnout lead.

53% of the electorate to date is vote by mail, and the GOP has a 65,927-voter lead).  Of the 47% of all ballots in that were cast in-person early, Democrats hold a 7,397-ballot lead.

In total, just over 3.48 million ballots have been requested – again far more than 2014, and more than 2016. 

Here is what is remaining:

Democratic unreturned ballots: 554,058

Republican unreturned ballots: 438,996

NPA unreturned ballots: 328,226

115K more of you Democrats need to #BringItToThePolls – because we are getting into the window where the mail starts getting dicey.  You can probably mail it today, but since the future of the state kind of depends on it, you might want to drop that bad boy off in person.

Black turnout continues to drive the shifting demographic make-up of the electorate.  Black share of the electorate is now about 12.2%, well above 2014 and 2016 at this point, and nearly at where 2016 was going into Election Day.  Hispanic share has remained flat all week, at about 12%, and right now, there are more Black voters than Hispanic voters.  What is interesting, Hispanics are disproportionally newer voters:  about 44% of Hispanics to date did not vote in the 2014 election, compared to about 30% among all voters so far.  I do think this will change into the weekend.  White is still just over 70%, though I still think it lands in the 67-68% range.  Overall, turnout among Black voters is 27.6%, and among Hispanics, it is 22.2%.  

One thing on Hispanics – turnout among possible Maria migrants is fairly low.  I was always skeptical of the fallout from the storm really changing the math.  That being said, Hispanic, and specifically Puerto Rican turnout over the next four days could very much impact this race.

Democrats should win the remaining days – though Republicans did have a good Friday last week.  I think this deal gets close to parity – but not quite at parity by Election Day, but reminder, there are a lot of more Republican 2014 voters left to vote on Election Day.

That being said, both sides are cannibalizing their own Election Day vote.  To date, there are slightly more Republican 2014 Election Day voters than Democratic 2014 Election Day voters who have voted, but the bigger story is the scale of things, roughly 17% of early voters this year voted on Election Day in 2014.  This is one reason why I think turnout is going to be record in number, and certainly higher in percentage than 2014, but  not crazy high – as a lot of our growth has come from people just voting earlier.  Minus those voters, Florida is more like 3.4-3.5 million voters to date, which is still forward leaning, but not as much as what the topline number suggests.

I don’t think we add another million voters over the closing few days, but we probably get to 5,000,000 votes going into Election Day.  By comparison, just over 6 million voted in 2014. I still think turnout is around 7.25 million – and maybe a bit more, but honestly, I haven’t had as much time to really dig into that question this week.  That is today/tomorrow’s project.

What is interesting, there aren’t a lot of red flags for either side in the data.  Turnout rates SW Florida are very strong for Republicans, as they are in some of the counties around Tampa and Orlando where Trump really ran up the score. They are arguably a little down in the Panhandle – and not just the Michael counties, but that is normal, as a lot of those smaller counties have more Election Day voting.

For Democrats, turnout in the bigger counties looks very good, though I’d like to see more in Dade from Democratic voters. Broward is closing in on 30% turnout, but only hit 44 in 2014.   Also, Democratic turnout in some of the Republican counties with higher populations of college educated voters is also strong – Dem turnout is higher than Republican turnout in St. Johns, and Sarasota, and close to parity in Seminole. 

So lastly, the question I get on twitter all the time how does Gillum or Nelson win if more Republicans vote than Democrats?

Right now, I think a turnout model where the electorate is 2% more GOP than Democrat is a pretty fair place to peg it, given the number of GOP super voters left to vote.  If we have 7.25 million votes, that is 150,000 more Republicans than Democrats casting a ballot.

So, if you believe the polling, Gillum is winning a few more points of Republicans than Democrats, and winning NPAs.

Let’s assume the final electorate is 41 R – 39 D – 20 NPA (it may be higher than this), and Gillum wins just 2% more Republicans than DeSantis wins Democrats – and he wins the NPAs by 8.  I think both of these assumptions are conservative, and realistic.

What happens:  Gillum wins by about 1.5%. 

There is a long way to go, but if the polling is right, the electorate is molding into shape for Gillum to bring it home.

Now if he, or frankly anyone, could just play quarterback for the Jaguars.

Thursday
Nov012018

5 Days Out

To:       The Blake Bortles Fan Club

From:   Steve Schale

Date:    November 1, 2018

We have reached Blake Bortles Day, the day where we are his jersey number days out until Election Day.  The question – which side will see Playoff Blake drive them to victory, and which side will get Everyday Blake, and turnover any chance of hope and victory.

To this end, a reminder of the data point that I mentioned in my first memo:  The last time that a non-incumbent won Florida, Barack Obama, it happened after the Jacksonville Jaguars made the playoffs and beat Pittsburgh on the road in the playoffs.  Well, last season, the Jaguars made the playoffs, then went on the road and beat the Steelers up there.  So, when Andrew Gillum raises his right hand in January and takes the oath of office, remember to thank the Jaguars.

Yesterday was the third day that total voting topped 300,000 ballots, was the third biggest day for in-person voting, and the second biggest day overall. 

48 counties have now passed their entire 2014 early voting totals.  Miami Dade today will likely end the day 100,000 votes ahead of their entire 2014 total.  People are voting.  Statewide turnout rate is now 28%, and both parties are over 30% among their own partisans.  We are cranking.

Total votes: 3,724,251

Republicans: 1,555,374

Democrats: 1,491.837

NPA/Minor: 677,040

Republican edge is 63,537 (+1.7%)

On Monday it was 2,726,392 (+2.2%) with the GOP holding an edge of 59,048.  So yes, people are voting – but all people are voting.

Nearly 1 million more voters have voted at this point compared to 2014.  For comparison, in 2014, right at 2.77 million voters had voted, Republicans had about a 132,346-ballot lead at this point – and led by around 4.77%. 

In total, 30% of all Democrats have voted, and 33% of all Republicans.  Turnout among NPA is 18.5%, but if history is a guide, this will pick up as we get closer to the election.

Republicans have now returned 64.3% of all their requested ballots, Democrats 57.2%, and NPA’s 52%.  Statewide, the return rate is 58.9%. 

And therefore, Republicans have a turnout lead.

55% of the electorate to date is vote by mail, and the GOP has a 66,170-voter lead (42.3% - 39.1%).  Of the 45% of all ballots in that were cast in-person early, Democrats hold a 3,173-ballot lead.

In total, just under 3.48 million ballots have been requested – again far more than 2014, and more than 2016. 

Here is what is remaining:

Democratic unreturned ballots: 597,745

Republican unreturned ballots: 481,505

NPA unreturned ballots: 351,945

And to the twitter followers who keep asking if some of this delta comes from voters who requested ballots, but voted in-person early, yes, it is.  But the difference isn’t significant between the parties.  Both parties are doing that.  Both parties are cannibalizing their Election Day ballots.  Both parties are turning out record early voting.   So please just tell your friends that to help Mayor Gillum #BringItHome, they need to #BringItToThePostOffice

As usually happens around this time, the electorate starts to look like Florida. Regional distribution is close to typical margins – Panama City and Tallahassee are a little down, some of which can be tied to the Hurricane.  Miami remains very robust, as does Fort Myers.  But it is balancing out. 

It is also starting to round into shape from a demographic perspective.  Black voters now make up 11.5% of all voters and growing daily.  Hispanic is still around 12.  White is now just over 70.  I think it is quite possible the Black share of the vote will reach the Black share of registered voters, which is 13.2%.  Not willing to say it will get beyond it yet, but that will be something to watch this weekend.   35% of non-2014 voters are people of color, and more Hispanic than the electorate at large.  This is all good news for my team.  On the flipside, there remain over 200,000 more Republicans who voted in 2014 than Democrats who voted in 2014 left to vote.   About 2/3rds of the 2014 voters who have not voted yet were Election Day voters in 2014. 

I expect today to look like yesterday, and the weekend to look more Democratic.  Will it be parity going into Election Day?  Probably not.  Will it be close?  Yes.

So, let me close on this one with some bigger thoughts.

This data is just what it is, data.  It means nothing more, or nothing less than it is.  What we learn from this data is who is voting, and for guys like me, it is a chance to measure where things are versus how we saw the election playing itself out.  I for one have never thought there would be more Democrats voting in Florida than Republicans this cycle.  So, to the people who ask me on twitter if I am freaking out?  The answer is no.  This looks like a Florida election, with just more volume than we normally see. 

What does all this mean?  Well, it is a little like asking if Blake Bortles is going to look decent on Sunday or not.  We won’t know until game time, or in this case, until the votes are counted. That doesn’t mean there is no value in this – what the model looks like determines what a winning calculator looks like.  If Republicans have a four-point advantage in turnout share, the percentage of NPA required to win gets high.  If it is 1 or 2, it is modest. If NPA voters grow as a share, the number Dems must win them by also decreases, because their voting power will increase.  This is what we are looking for now – where does this plane look like it might land, and what is required of either to win.

Go back to 2016 – I think I described, based on the model, Donald Trump’s challenge in Florida on election day as akin to hitting a 3 wood over water to a par 5 – not a high percentage or even likely shot, but also not an impossible shot.  In the end, Trump won NPAs in Florida by six, saw a surge of voters, and had more certain voters left to go, and that is how he won EDay by 13 points.  Take away any one of those, and he probably didn’t make up her 240K vote lead.

The latter point is another that goes to how hard it is to predict.  In 2016, Democrats went into Election Day with a 1.5% edge in turnout, and I had her modeled up about 3-4%. Turns out she was up 4, so she in the pre-election day vote, she clearly won a larger share of NPAs, and based on where she was strong, she was likely winning more pre-Election Day Republicans than the other way around.

I think the odds are quite high that the same thing exists now, that both Gillum and Nelson are doing well with NPAs and doing better with Republicans than Scott/DeSantis are with Democrats.  How much better?  It is hard to tell – there are some indicators, and I’ll get to that in a second.  It is also important to remember, as virtually every good pollster will tell you, to poll people who have already voted.  People who have voted are harder to get to take polls, and just like every other universe, pollsters are trying to model what percentage has already voted, versus who hasn’t. 

When I’ve been asked about the “wave” – a term I hate with a burning passion, I think about it this way.  Florida isn’t a monolithic place, and just like trends aren't universal around the country, they aren’t here either.  But there are places in the country that have a seen more of a surge than others – and where those people live in Florida, are likely where will see the same.  What often happens in these moments is three things:  one party gets a bit more jazzed up, one gets a little more depressed, and NPA voters pick up and slide a direction. But these aren’t huge movements.  And down the ballot, in Congressional and Legislative races, the “wave” often means just a couple point shift – the kind of shift that gets you from losing by 2 to winning by 1.   The issue I have with “Wave” is it makes it seem like you should win a bunch of races by a lot, but that’s not what happens.  Wave means a tie goes to your side.

So, what are we seeing?  Both parties are pretty jazzed up, though in fairness, if we compare to 2014, Democrats are more jazzed by comparison than Republicans, hence why the voter turnout difference is closer.  At the same time, the comparison day to day is tricky, because we are a million votes into the election than we were at this day in 2014.   And NPAs and the new electorate?  From NPAs, we have polling, and among the non-2014 voters, we know they are marginally more Democratic and diverse than the 2018 electorate so far.    BUT, we also know more Republican Election Day voters exist.  We also know in Florida, not all swing voters are equal, but that is a longer piece.

There are some interesting things out there.  When I look for things suggesting movement the direction of my team, there are data points out there.  Seminole, and Sarasota both look pretty good.  Both are places that popped up for us in 08, and both have higher percentages of college-educated whites.   I thought when Margaret Good won her State House special election this spring, winning by 3,000 votes, even though 3,000 more Republicans voted in the race – meaning she won both NPAs and crossover voters, it might be the biggest warning flare of the cycle.  We saw a 'turn left' in 2008 as well – in that same county.

Pinellas looks better than 14, but one warning there, Crist really out-performed there, due to his history in the community, so even better in Pinellas may not translate to 2014 totals.  But that is a unique situation there.  The big Dem counties are all up.  Duval looks good.  Alachua is crushing it.   At same time, some of the “Trump surge” counties – Pasco, Marion, Hernando, Citrus, look on paper to be a bit more Republican than last time – though not at the same levels as the trends in the places looking better for Democrats.

If 90% of the 2014 voters who haven’t voted show up, we are looking at another 2.5 million votes.  On the pace we’ve seen with non-2014 voters, that number is probably around 750K, and always land north of it.  But pretty good chance we are more than halfway done.  There is nothing about this data that scares me for Nelson or Gillum winning, and in fact, I believe if we added them up right now, both are.  Both could win by 1 or 2, both could win by 5 or 6, they could split – all are options.  I believe now, as I did after right after the primary, that Gillum will win.  I believe now, as I have for a year, that Nelson will be tight, but that the race would come down to where the country was going late.  If the NPA numbers in public polling are to be believed, he should win. 

That being said, I don’t hold out much hope for my Jaguars going to the playoffs – but that’s OK – they also missed the playoffs after Obama won in 2008.

Wednesday
Oct312018

6 Days Out Florida

To:       Fellow Data Nerds

From:   Steve Schale

*5 Days of early voting (including today)

*6 days until Election Day

*54 days left in the Christmas shopping season

*460 days until the Iowa Caucuses

*712 days until my first 2020 General Election memo

*734 days until the 2020 General Election

The Jacksonville Jaguars traded Dante Fowler to the Rams for a 2019 and 2020 mid-round draft pick, and yesterday, we saw our biggest day of early voting – more than 350,000 total ballots cast, with more than 220,000 coming from in-person early voting.  The day was a virtual tie, with Republicans winning the day by about 2,000 voters.  As I said in a few tweets yesterday, I think this is the likely trend during the middle part of this week.

In fact, voting was so robust yesterday that the voters have surpassed the entire 2014 in-person early voting total, as well as the entire pre-election day total. 

Total Votes: 3,414,365

Republicans: 1,431,655

Democrats: 1,368,718

NPA/Minor: 613,992

Republican edge is 62,937 (+1.8%)

Tuesday it was 3,063,662 (+2.0%) with GOP holding an edge of 60,705

Monday it was 2,726,392 (+2.2%) with the GOP holding an edge of 59,048

Sunday, we were at 2,580,347, with the GOP holding an edge of 70,415 votes (+2.7%).

On Saturday, it was 2,316,413, with the GOP edge at 74,334 (+2.2)

For comparison, in 2014, right at 2.6 million voters had voted, Republicans had about a 133,000-ballot lead at this point – and led by around 5.2%.  In 2018, the GOP lead was about 20K votes at this point out of just under 4.5 million votes.

Republicans have now returned 61.1% of all their requested ballots, Democrats 52.4%, and NPA’s 48.9.  Statewide, the return rate is 55.8%. 

In total, just over 3.46 million ballots have been requested – again far more than 2014, and more than 2016. 

Here is what is remaining:

Democratic unreturned ballots: 636,840

Republican unreturned ballots: 522,086

NPA unreturned ballots: 372,470

There are nearly 115,000 more Democratic unreturned ballots than Republican ones.  Dear Democrats - if you want to help Andrew Gillum #BringItHome, please #BringItToThePostOffice.

The weekend really helped to bring the electorate more in line with how it should look.  Through Monday’s vote (I won’t get to look at through Tuesday until later today), the electorate is 71.8% white, and 11.3% Black (African American and Caribbean).  The drop of the former is mostly driven by a rise in the latter, as Black participation is now above 2014 and 2016 levels at this point.  Hispanic is around 12%, though this number may be deceptively low, as 41% of the Hispanic vote is from voters who did not participate in 2014 (compared to 26% of the white vote, and 29% of the Black vote).  I will be watching this number through the weekend and wouldn’t be surprised to see it jump to 13-14 quickly.  Worth noting – over 20% of the voters on Sunday were Black.

As I said on Monday, when looking at the ethnicity of who is left to vote, if just the people who voted in 2014 show up to vote between now and Election Day, the Black (reminder, Black voters in Florida include African American, Caribbean, and some Hispanic) vote will land north of 12%.  And again, that is if no one new shows up.  I suspect the Black share will end up right around the Black share of total voter registration, which is just over 13.  I think when all said and done, the white share of the electorate probably lands around 67 or 68, which is a pretty good place for Gillum, and Nelson.

Dems lead Republicans with non-2014 voters by about 40,000, which is good news for the home team. On the flipside, and just to stress this for my Democratic friends – there is real work to be done, as there are well over 200,000 more 2014 Republicans than 2014 Democrats left to vote.  My side needs more infrequent turnout to keep the turnout model in a healthy place.

Turnout has been, in a word, nuts.  There are 37 counties who have already exceeded their entre pre-2014 Election Day vote totals, and 35 who have seen more in-person early voting than in all of 2014.

Statewide, already 25.7% of all registered voters have voted. In 2014, turnout was about 49%.  30.6% of Republicans, 27.7% of Democrats and 16.8% of NPAs have voted.

Turnout is highest in The Villages, where 48% of voters have already cast a ballot.  In total, 8 counties have total turnout over 30% of registered voters.    Just as Republican turnout is outpacing Democratic turnout statewide, it is in most counties, though the Democratic turnout rate is higher in some interesting places, namely Sarasota and St. Johns County, two Republican counties with a higher than average population of college education voters.  St. Johns, one of the most Republican counties in Florida (Northeast Florida, just south of Jacksonville), is an interesting place where Trump did worse than Romney four years earlier.  It isn’t a place that will be close, but like all of Florida, it is a battle over margins.

The places where Republicans are most over-performing tend to be small, with one exception:  Miami-Dade.  Republican turnout is running 6.5% higher than Democratic turnout.  There are reasons for this, namely local Republicans have long had a very effective vote by mail program, specifically in older Cuban communities.  There are also several very competitive down ballot races, where the GOP is running field programs.  For Democrats, you should be very happy that President Obama is making a trip there.  There are roughly 120,000 African American and Caribbean voters, and 150,000 Hispanic voters who voted in 2016 and not 2014 in the Miami media market alone, and that is a universe the former boss is pretty good with.   In 2016, we did see a real “Obama bump” in the markets he visited, and I suspect we will again in Miami over the weekend.

I also want to point out Duval County.  As Democrats there have given up hope with the Jaguars, they have poured their energy into turning out the vote.  Six days out, Democrats hold a 2,500-voter lead in Duval, and it growing a bit daily.   The Republican statewide math is easier if Duval is heavily Republican.  Gillum is aggressively contesting the county, which is very smart.

On the flipside, southwest Florida continues to be on a roll for Republicans.  The market should be about 7% of the statewide vote, but right now it is close to 10%.  Republicans in all three major counties in the Fort Myers media market have turnout north of 40%, some 9-13% higher than the GOP statewide average.  Lee County alone will likely hit 50% total Republican turnout sometime tomorrow.  Honestly, I don’t know why President Trump is visiting here, given that turnout is also juiced.

One last thing.  On Sunday, I had a chance to visit Panama City.  If you have not personally seen the damage from Hurricane Michael, it is hard to put it into words.  The counties most effected by the hurricane make up about 2% of the statewide vote, and as one can expect, turnout there was sluggish at first, but it is coming back to life.  In the most affected counties, about 54,000 voters cast a ballot before election day in 2014.  Today it is about 36,000.  But there are 5 days left, and it is catching up. Bay County, for example, has now seen nearly 14,000 in person early voters, compared to roughly 17,000 in all of 2014.   It is almost a sure thing that despite the storm, more people in Bay County will in-person early vote than they did in the last Governor's race.   This may not be great for my home team, but it is good news, and a testament to the resiliency of the community.

Tomorrow I will take a deeper look at some of the demographics driving the NPAs who are voting, as well as the question of youth voters.   Until then, thank you for reading.

Monday
Oct292018

8 Days Out -- Florida Memo

To:       People who can’t stand Mondays.  Everyone else stop reading.

From:   Steve Schale

Re:       8 days out

It finally cooled off in Florida, but the voting got hot.  As of this morning: 2,726,392

This breaks down:

Republicans: 1,151,593 (42.3)

Democrats: 1,092,547 (40.1)

NPA/Minor: 482,252 (17.7)

Republican edge is 59,046 (+2.2%)

 

Yesterday, we were at 2,580,347, with the GOP holding an edge of 70,415 votes (+2.7%).

On Saturday, it was 2,316,413, with the GOP edge at 74,334 (+2.2)

 Democrats made up all the ground they lost late last weekend, winning the weekend vote by roughly 15,500, or just under 4%.  Most of that came yesterday, when Democrats won the day by 11K votes. My home team had a very good weekend.

 Before my Democratic friends on twitter get too excited, there wasn’t much in the way of vote by vote by mail ballots, so unless the 118K or so more Democrats than Republicans who hold an absentee ballot #BringItToThePostOffice, the GOP lead will likely bounce back a bit tomorrow morning when the weekend absentees get added into the mix. 

For comparison, in 2014, right at 2.2 million voters had voted, Republicans had about a 140,000-ballot lead at this point – and led by around 6.4%.  In 2018, the GOP lead was about 9K votes at this point out of 4 million votes.  It was basically tied.

Republicans lead for one reason, and one reason only: Democratic return rates for vote by mail are lagging Republican rates – by seven points in terms of the rate of return. Right now, statewide the return rate is about 49% -- though given what is currently sitting in the mailboxes at Supervisors offices, there is no question more than 50% of Floridians who requested a ballot have mailed it back.  Again, those numbers update in the morning.

Republicans have now returned 54.4% of all their requested ballots, Democrats 47.6%, and NPA’s 42.5.  Statewide, the return rate is 36%. 

In total, just over 3.42 million ballots have been requested – again far more than 2014, and more than 2016. 

Here is what is remaining:

Democratic unreturned ballots: 722,498

Republican unreturned ballots: 604,885

NPA unreturned ballots: 412,736

The electorate is continuing to trend more diverse.  Not including Sunday’s data, which will make these numbers even more diverse, the electorate looks like this (again thru Saturday):  73% White, 10.5% Black, 11.7% Hispanic.  Compare this to Thursday last week, when the electorate was 75% White, 11.5% Hispanic, and 8.5% Black, though as in-person early voting, the electorate is trending more diverse.  Keep in mind, vote-by-mail in Florida tends to be far more white than the final electorate. 

That being said, when looking at the ethnicity of who is left to vote, it is clear that Black turnout is rounding into shape – if just the people who voted in 2014 show up to vote between now and Election Day, the Black (reminder, Black voters in Florida include African American, Caribbean, and some Hispanic) vote will land at about 12.5%.  And again, that is if no one new shows up.  I suspect the Black share will end up right around the Black share of total voter registration, which is just over 13.

If there is a concern I have about turnout from my party perspective – where as the Black share of the electorate has been trending up, Hispanic has been pretty flat.  Outside of Cuban Republicans, Hispanics tend to vote later, and more on Election Day, but this is something to watch this week and into next weekend.

Looking at the electorate that has voted, and the electorate left to vote, there is good news and bad news for my home team.

In terms of non-2014 votes, slightly more than 30% of the electorate did not vote in 2014.   Those voters are younger (30% under 50 – and 17% under 35 compared to 22% under 50 and 8% under 35 for all early voters), more Democratic (Dems are +4%), and more diverse – particularly more Hispanic. 

In terms of people who 2014 voters who have not voted, there are about 200,000 more Republicans. Dems make some of that up if the trends with lower propensity voters keep up, but not all of it.  In other words, after you read this memo, if you are for Nelson and Gillum, go turn out some voters.

In terms of the vote by media market, it is starting to look like Florida.  The two markets most impacted by the storm: Panama City and Tallahassee, are still down.  Miami, as a share of the electorate, continues to grow – and while this will likely flatline or drop a bit as some other areas even out, it is well ahead of where it was at this point in 2014 and 2016.   Orlando, arguably the most important market this cycle, continues to be very competitive – and robust in terms of turn out.  Fort Myers, after blowing up in early vote by mail, is coming back to earth.  Tampa looks like Tampa.

I looked at 2018 versus 2014 by county margins, there are a lot of high points for my home team:

This data reads this way 2014 margin -- 2018 margin; (2014 share -- 2018 share)

Broward: 36,893 -- 74,654 (57-27 -- 58-24)

Dade: -31 -- 22,153 (41-41 -- 44-35)

Orange: 4,611 -- 23,619 (43-39 -- 49-33)

Hillsborough: -1,584 -- 9,556 (40-42 -- 44-39)

DUUUVAL: -8,310 -- 472 (38-50 -- 44-43)

Alachua (UF): 3,989 -- 11,255 (53-33 -- 58-28)

 

And in the swing counties:

Seminole: -9,032 -- -3,728 (32-50 -- 37-43)

Pinellas: -7,921 -- -1,855 (37-43 -- 40-41)

St. Lucie: 748 -- 2,853 (41-39 -- 44-38)

What is interesting, and again is good news – even in many of the bigger GOP counties where they have a higher margin today than four years ago, Democratic turnout is keeping it in check.  For example:

In Lee (Fort Myers), Sumter (Villages), Collier (Naples), St. Johns, Clay (near Jax), while the GOP margin is higher than it was four years ago, the partisan share of vote difference is closer for the Democrats.  Take the aforementioned St. Johns County, one of the most Republican counties in the state.  Republicans have a 3,758 bigger vote margin, but the model has gone from 63-23 R in 2014 to 58-28 in 2018.   Winning Florida for my side is all about those little shifts.

One last little data point, then off to work.  Right now, statewide turnout is about 20.5%.  Among Republicans, it is 24.6%, and among Democrats its 22.1% (NPA is at 13%).  Republican turnout rates are higher in 61 of 67 counties.  I point this out because while I do believe that there is a lot of data from the weekend that looks good, Democrats keep in mind that Republicans are blowing out turnout as well.  More people are voting overall – and there are probably at least 4.5 million more people to vote – and if only the most certain show up, which they will, the GOP will have a solid advantage in turnout. 

I feel better about where my side is today than I did last week, but I don’t feel great.  This still feels closer than the polls, and while I absolutely believe NPA voters will decide it – as they typically do in Florida, how close things are from a partisan perspective dictates by how much our side has to win the NPA vote. 

And while it may be a bye week for my hapless Jaguars, it is far from a bye week for those of you all who want to bring it home.

One last note:  Last week, my side lost one of its brightest rising stars.  Tyrone Gayle was a kid from Jacksonville – we went to the same high school (though he was much younger) and took his charge to be a change agent seriously.  In his short 30 years on this planet, he touched more people than most of us could in a lifetime.   He battled stage 4 cancer for more than two years, and while most of us bitch about having a cold, Tyrone went through treatment while still during a Presidential campaign, never losing his spirit.    A colleague of him said: “Tyrone could fly, but that was never enough for him – he wanted us all to fly with him, so he elevated every person he touched, lifted each and every one of us.  His body failed him last night, but his spirit soars on, flying above us, forever lifting us up.”

While I can’t say we were all that close, he was my friend & fellow hopelessly optimistic Jaguars fan -- and more importantly, I was just an admirer and fan of his, a fellow EHS and Duval kid who did good, and one whose limitless potential was sadly cut short.  F Cancer.  #GayleNation