Local News

Actions

EARLY VOTING BLOG: Early voting kills Trump in NV

Posted at 3:15 PM, Oct 21, 2016
and last updated 2016-11-07 12:08:39-05

UPDATED, 11/7/16, 6:30 AM

Before I dive into the down-ballot races, I suppose we should address the elephant in the room: Could I be wrong?

It does seem a waste of time to consider this possibility. However:

I could be wrong about Donald Trump losing Nevada only if voters behaved very differently during early voting than they usually do. That is, unlike 2008 and 2012, many of those Democrats were voting for Trump and Hillary Clinton is losing by a substantial margin among indies.

None of this is backed up by common sense – do you think the Reid machine/HFA is turning out crossover voters? And I have seen nothing to indicate Trump is winning indies in the landslide he would need to keep it close, especially because the composition of non-major party voters is much less white than it used to be.

I’d guess that right now, based on history and my sources:

  • Trump is down by at least 40,000 votes.
  • About 770,000 votes have been cast, likely two-thirds of the vote.
  • Let’s suppose that there is an Election Day turnout of 450,000 voters. Trump would probably need to win Tuesday by about 10 points to win. 
  • This is almost impossible, unless the Democrats decide not to turn out voters on Election Day.

One more hopeful scenario for Trump includes large rural turnout:

In 2012, 165,000 votes were cast in rural Nevada. Romney won the rurals by 40,000 votes, or 25 percent. About 100,000 already have voted in the rurals. Let’s suppose Trump has a 30,000-vote lead already, which is generous. Could he add 20,000 votes on Election Day? That would be spectacular, but not impossible. So say he has a 50,000-vote lead.

That would mean the Democrats would have to win Washoe and Clark by under 50,000 votes. The 2012 numbers: Barack Obama won those two counties by a combined 107,000 votes. So Clinton would have to do less than half as well as Obama did. Less than half.

I repeat: Trump is dead here, barring a miracle or anomalies invisible not just to me, or many other experts. (I surveyed 14 smart insiders this weekend. All 14 said Trump loses the state.)

(The dynamics of the U.S. Senate race are different, which gives Team Joe Heck hope. If he can win his own congressional district, he has a chance. The numbers there don't look good, as you will see below, but Heck should run well ahead of Danny Tarkanian.)

Now for the deep dive:

CD3: The Dems are up 43-38, or about 10,000 votes out of nearly 220,000 already case. Fifty-four percent of voters already have cast ballots. Three points above the registration edge. Danny Tarkanian would need a robust, lopsided turnout Tuesday to win. Some private data says he’s already lost.

CD4: In the Clark part of the district, Democrats have a whopping 25,000-vote lead. That’s out of about 160,000 votes cast, 49-34, or right at registration. There is almost zero chance Cresent Hardy survives that. There are not enough votes left in the rurals to save him.

State Senate: Half of the voters have cast ballots in the three competitive districts, including 56 percent in Mark Lipparelli’s open seat. Democrats have an 800-vote lead in the Gansert-Reese race up north, which is well above registration but may not be enough for him unless there is a Washoe wave. But the Democrats have almost double-digit ballot leads in both the southern districts, giving them a real chance to take both. There are only about 35,000 votes left in those districts, so the GOP contenders need to be getting a lot of crossover and indie votes in early voting to have a chance. That can still happen in legislative races, but the Dems clearly have an advantage here.

Assembly: Here’s how it looks: I assume John Moore and Shelly Shelton are gone (ADDENDUM: AD34 ALMOST SURELY GONE FOR GOP, TOO (VICTORIA SEAMAN'S SEAT), and eight races are in play (I'm being generous to Brent Jones.). The Democrats need only three of them to take control. The GOP folks believe they have done enough on the ground to stave it off and hold onto 22 seats. But the numbers are daunting in early/mail voting:

AD4 : Dems +220

AD5: Dems +1,600

AD9: Dems +2,300

AD21: Dems +1,800

AD29: Dems +1,200

AD35: Dems +1,900

AD37: Dems +300

AD41: Dems +1,700

So you can easily see the GOP holding AD4 and AD37, which gets them to 17. The rest look terrible for them, so 25 seats is not impossible for the Dems, especially because most of those districts only have about 15,000 or so votes left.

UPDATED, 11/5/16, 3:30 PM

Let me overwhelm you with more data.

The Joe Heck campaign has a chart to push back against the notion the turnout is bad for the GOP Senate candidate. That is: It actually is very white and more rural.

The rural numbers, which I have pointed out here, are slightly up. But we don't know yet whether this will increase overall cow county turnout or if it is cannibalizing overall rural turnout. Trump, especially, and Heck, perhaps, are going to have to win the rurals by unprecedented margins to win. I'm talking in the neighborhood of 50,000 votes, which seems a margin too far.

The natural conclusion: Heck is like Dean Heller in 2012, and Sen. Heller held on against the tide by 12,000 votes. That may be so, although the dynamics are very different, some of them good for Heck and some of them not so good. 

One note: The Heck campaign's admission that the turnout is very much like 2012, which I have been telling you in this space for two weeks, is essentially an admission that Donald Trump is dead.

There is other data from the early vote worth mentioning that Heck may not like as much:

Even though most observers, including myself, have used 18 percent of the electorate in 2012 being Hispanic because of exit polls, that is not as foolproof as an actual look at the voter file. The Democrats have always believed it was slightly lower -- 16 percent, perhaps.

Here's what the data shows on early/mail vote:

 

2012 

- 11 percent Latino 

- 77 percent white

 

2016

 

-- 13. 5 percent Latino

--77 percent white

 

The more ominous news for Republicans is how the majority Hispanic precincts looked between the two cycles: Huge increases except for one. Here is the actual data.

And here is some data from two polling places that are almost all Latino, including Cardenas Market, which had those long lines:

East LV Community Center Turnout

2012: 3,510

2016: 5,878

 

Cardenas

2012: 3,854

2016: 4,703

 

One more data point to consider: NextGen Climate, which is trying to increase millennial turnout, reports that on-campus turnout was up a lot during early voting. The raw numbers may not seem large, but in close races, they may make a difference:

 

2012 -- 5,827

2016 -- 8,696

 

That's a 2,869-vote increase, or nearly 50 percent.

 

 

 

UPDATED, 11/6/16, 5 PM

The SOS now has a cumulative turnout page up showing what I told you earlier. (Tiny Esmeralda and Mineral not in from Friday.) 

Since my first post this morning, in which I all but buried Donald Trump's chances in Nevada, my Twitter feed has, of course, teemed with unrestrained ecstasy and unhinged conspiracy.

So be it. 2016.

Let me answer the question both sides are asking: Could Trump still win Nevada?

Well...yes, BUT:

Let me remind you of the math: Trump would need to be holding 90 percent of the GOP base and Clinton would have to be losing 15-20 percent of hers and he would have to be winning indies for him to be competitive. Let me be clear: None of those things are likely. 

The Reid machine and the Hillary campaign did not spend two weeks turning out crossover voters. They know what they are doing. Trump is probably down 12-15 points in Clark County and 65-70,000 votes. You can't make that up unless Election Day turnout is so large and so GOP-heavy that he could. And with two-thirds of the vote in, and with Democrats not simply willing to roll over and not rev up the machine on Election Day, that ain't happening.

The total votes cast so far -- 770,000 -- are well above 2012 after early/mail voting -- 705,000. But because of all the new voters -- 200,000 of them -- the raw vote lead statewide is down from 2012 (Clark is slightly higher) and the percentage edge is about 2 points lower. The GOP has less than a 1 percent turnout advantage, which is on par with 2012.

But votes are votes. And if the Clark firewall is what I think it is, and Clinton has a 70,000 vote lead when those first numbers pop up on Election Night, it's game over. Indeed, 60,000 might be enough. Why? Because the best Trump can hope for is to win rural Nevada by 50,000 votes. And that would be HUGE -- Romney won the rurals by about 40,000 votes. In Washoe, where the Dems have a 1,000-vote ballot lead now, that would mean Trump would have to win Washoe fairly decisively and reduce the Clark margin, which actually is likely to grow, on Election Day.

Look at it another way:

Let’s be conservative and say two-thirds of the vote is in – it was 70 percent in 2012 and turnout is down this year. That means there are roughly about 385,000 votes left. Let’s say Trump did the impossible and won Election Day by 10 points – 50-40. That would be 192,000 to 154,000, or 38,000 votes. He would probably still lose.

And the chances of Trump winning election day by 10 points are about the same as Billy Bush anchoring the CBS Evening News.

In the Senate race, Joe Heck probably would have to win Election Day by only half as much. But winning Tuesday by 5 points is a daunting task. The Heckophiles are on Twitter making comparisons to 2012 and using the lower lead percentages for the Democrats as a sign that Heck can pull a Dean Heller. Maybe. But private data I have seen belies it, and it would be a disaster for Catherine Cortez Masto and the Reid Machine if that occurred.

The Democratic strategy of building up a registration advantage, especially in Clark, then banking votes in in early voting has worked in the last two cycles. There is no reason to think it won’t this time, and there is no reason to believe the Democratic machine won’t work on Election Day, too.

Republicans may be right that they can get higher turnout on Tuesday. But that’s what they said in 2012, and Democrats won Election Day by 12,000 votes.

And consider this: Latino vote was a higher than usual percentage of the early vote – more than 13 percent. But many Hispanics don’t like to vote early, and that means many will vote Tuesday, which is a real problem for Republicans. In five of the six highest Latino precincts, early turnout was up from ’12, too. It's up 30 percent from 2012, data mavens tell me. We will see if it's more voting early or if it continues Tuesday.

I think the GOP will gain ground on Tuesday. But Clinton is too far ahead, I’d guess, and Cortez Masto may be, too, for it to make a difference.

Sometime this weekend or Monday, I will do some math on the down-ballot races and show you why the two GOP House seats in play are almost surely gone and why the Legislature may be, too, for Republicans. 

UPDATED, 11/5/16, 7 AM

Donald Trump will be in Reno on Saturday, but the Republicans almost certainly lost Nevada on Friday.

Trump's path was nearly impossible, as I have been telling you, before what happened in Clark County on Friday. But now he needs a Miracle in Vegas on Election Day -- and a Buffalo Bills Super Bowl championship is more likely -- to turn this around. The ripple effect down the ticket probably will cost the Republicans Harry Reid's Senate seat, two GOP House seats and control of the Legislature.

How devastating was it, epitomized by thousands of mostly Latino voters keeping Cardenas market open open in Vegas until 10 PM? This cataclysmic:

----The Democrats won Clark County by more than 11,000 votes Friday (final mail count not posted yet), a record margin on a record-setting turnout day of 57,000 voters. The Dems now have a firewall -- approaching 73,000 ballots -- greater than 2012 when Barack Obama won the state by nearly 7 points. The 71,000 of 2012 was slightly higher in percentage terms, but raw votes matter. The lead is 14 percentage points -- right at registration. You know what else matters? Registration advantages (142,000 in Clark). Reminder: When the Clark votes were counted from early/mail voting in 2012, Obama had a 69,000 vote lead in Clark County. Game over.

----The statewide lead (some rurals not posted) will be above 45,000 -- slightly under the 48,000 of 2012, but still robust. That's 6 percentage points, or right about at registration. The GOP turnout advantage was under a percent, worse than 2012 when it was 1.1 percent. 

----The Dems eked out a 200-vote win in Washoe and lead there by 1,000 votes. It was even in 2012. The rural lead, before the stragglers come in, is 27,500. It probably will get above 28,000.

----Total turnout without those rurals: 768,000, or 52.5 percent. If overall turnout ends up being 80 percent, that means two thirds of the vote is in -- close to 2012. Republicans would have to not only win Election Day by close to double digits to turn around the lead Hillary Clinton almost surely has in early voting, but they would have to astronomically boost turnout. The goal for the Dems during early voting was to bank votes and to boost turnout as high as possible to minimize the number of votes left on Election Day to affect races. Folks, the Reid Machine went out with a bang.

As an exclamation point to a historic night in Nevada, in which Clinton essentially locked up the state and Hispanics, insulted all cycle by Trump, streamed into the market, here is what the final Cardenas numbers showed (tallied by an on-the-ground activist):

1,904 voted
1,258: Ds, 66%
165: Rs, 9%
481: NPs, 25%

So Cardenas was responsible for adding 1,000 to the Democratic lead.

Trump has almost no path to the presidency without Nevada. He can say whatever he wants in Reno on Saturday and boost rural turnout a lot, but he made his own bed when he announced his candidacy.  

I'll dive deeper into the numbers later to show just how deep the wave could be Tuesday.

 

UPDATED, 11/4/16, 9:30 AM

No new rural numbers yet, but Clark details show troubles getting worse for down-ballot Republicans. First, some math based on what we have so far (will be a tiny bit better for GOP when outstanding rurals come in).

I say again that it's highly unlikely Trump holds 90 percent of the GOP base while polling is showing Hillary doing so. But I have included scenarios where they both hold 90 and 80. Either one looks bleak for Trump. And if he holds only 80 and she holds 90 and he wins indies by 30 (!) points, Trump would be winning Nevada by....1,000 votes. 

Other scenarios:

Both hold 90 percent of base and Trump wins indies by 10: 48-46, Clinton

Both hold 90 percent of base and Trump wins indies by 20: Clinton by 4,000 votes

Both hold 80 percent of base and Trump wins indies by 10: 46-44, Clinton

Both hold 80 percent of base and Trump wins indies by 20: 49-45, Trump

So he can win Nevada. But Trump would need base numbers and indie numbers that seem unlikely right now. That's why the Republicans hope to change the electoral makeup on Election Day, something they were unable to achieve in 2012.

Deep dive into how the Dems big day in Clark affected down-ballot races:

CD3: Nearly 300 votes added by Dems. 7,000 raw vote lead. 3.5 point lead, 41.5-38. 1.5 above reg. Half the voters there have cast ballots now. Not locked by any means for Jacky Rosen, but she is favored.

CD4: Dems added 2,000 votes on Thursday, lead now about 21,000 in the Clark part. 14-pt lead, point below reg. Turnout is 46 percent. I wonder who Ruben Kihuen has in mind for his chief of staff.... Narrow path for Cresent Hardy (huge GOP turnout on Election Day may not be enough) 46 percent turnout so far

SD5: 150 more voters added to lead. 2,800 raw vote lead. That's 8 points, 3 more than reg. Daunting. Almost half the vote in.

SD6: More than 200 added, lead now at 2,750. That's 7 points, right at reg. More than half the vote in. Need a sea change.

SD15: Bad night, lead down to 850. But Dems up 2 points in GOP district.

Bottom line: Have to say Senate leans Dem, but not ready to call it one way or the other until....Sunday.

Those four Assembly seats I have been tracking all got worse for the GOP: I don't see how you are losing by 2,000 votes after early voting in these small political subdivisions, as David Gardner and Derek Armstrong will be, and hope to survive. Stephen Silberkraus has a 1,000 vote deficit to deal with and Brent Jones has 1,600. These are anywhere from 5 to 10 point deficits. Tough.

UPDATED, 11/4/16, 6:00 AM

All rurals not in, so once again subtract a few hundred from these totals, but here's what we have after a Thursday of Dems +5,800 in Clark and GOP +350 in Washoe:

Statewide: Dems +37,700 (Let's say after the other cows come home,  it's 37,000) In 2012 at this time, it was 41,000, and there were 200,000 fewer active voters.

Clark: Dems +61,500 That's essentially the same as '12, but there are 150,000 more active voters.

Washoe: Dems +800 Not much different from '12.

The statewide lead is 5 and a half points -- 42.2 percent to 36.7 percent -- or just under registration. The GOP has an overall turnout edge of a little more than a point -- 51 percent to 50 percent. It was about 1 percent at the end of early voting in 2012.

The overall turnout is 680,000 voters, or about 46.5 percent of all active voters. That means, if turnout is 80 percent when all is said and done Tuesday, about 60 percent of the vote is in. 

So where are we?

In 2012, after early voting was done, 56 percent of active voters had turned out. So they will not get there, even though the last day is always the biggest turnout day.

Democratic and Republican turnout is down by percentage, but the Dems still have the machine whirring in Clark County to build up a firewall. It is the same as 2012 in raw votes, down slightly in percentage. It's still a 13 point edge in Clark, which is 1 point under the Democratic registration margin.

The Democrats would have to add 10,000 voters today to their statewide and Clark leads to get to where they were in 2012 in raw votes. Highly unlikely. But they already are above where I thought they had to get in Clark -- 60,000 -- to feel pretty good.

Remember that Barack Obama won the state by almost 7 points after the Dems built that 71,000-voter edge in Clark. So it probably doesn't have to be that big for Hillary Clinton to win.

Indie turnout is going to end up at about 150,000, and that universe will be important. But it's still only a fifth of total turnout right now.

Trump still has a very narrow path here that goes like this: Hold the Dems to a reasonable win on Friday and, because turnout is down, squeeze extra GOP voters out on Tuesday, especially in rural Nevada, where Republicans now have about a 25,000 vote lead. The Republicans need to win Election Day pretty big -- remember the Dems won it in '12 -- but it remains in question whether it will make a difference at the top of the ticket, much less down ballot.

UPDATED, 11/3/16, 10:00 AM

All the rurals still not in, but a deep dive into the Clark numbers soul leave Republicans feeling...blue. 

I don't see how Cresent Hardy survives, and Danny Tarkanian needs a miracle. The state Senate races look bad for the GOP, as do the Assembly ones. My only caveat on the Carson City contests is the personal touch still matters.

On the presidential (and maybe applies to the U.S. Senate, too), some math still holds:

----Both candidates get 90 percent of base and split indies: Clinton by 4, 29,000 votes

----Both candidates get 90 percent of base, Trump wins indies by 10: Clinton by 2, 17,000 votes

----Both candidates get 90 percent of base, Trump wins indies by 20: Clinton by 3,000 votes

Note: Trump is not going to get 90 percent of the GOP base, and with all of those votes banked even before the Comey letter, it's almost impossible for him to win indies by 20. (Romney won indies by 7.) You see his challenge.

If Clinton holds her base here (data I have seen shows she is, and minority turnout is going up) and turnout patterns don't dramatically shift in the last two days of early voting, she can't lose Nevada. Solid lean Clinton right now.

Deep dive:

Congressional seats are almost gone for the GOP. One almost certainly is.

CD3: Dems won the early vote by 400, lead by 6,800. 41-37, or 2 points above registration.  Turnout is at 46 percent there, which means, I'd guess, almost 60 percent of the district already has voted. bad news for Tark.

CD4: Dems won by 1,300 votes in the Clark portion of the district (that's 85 percent of it), boosting the raw vote lead there to almost 19,000. That's 47-33, or a point below registration. Turnout is 43 percent there. I wonder where Ruben Kihuen will live in DC.

Legislature looking bluer:

SD5 -- Dems added 150 to their lead, which is now at 3,100 raw votes. That's huge in a legislative race. 44-35, or 7 points above the registration edge. Republicans still seem to think Carrie Buck can beat Joyce Woodhouse, but getting harder to see a path. (45 percent turnout)

SD6 -- Dems won by almost 200 votes on Wednesday, giving Nicole Cannizarro a 2,500-vote cushion now among partisans over Victoria Seaman. That's 43-36, or right at registration. Big trouble for Seaman. (48 percent turnout)

SD15 -- Dems lost 200 over their lead Wednesday, but still have a 1,000 raw vote edge. 42-39, or about 6 points above the reg edge. Heidi Gansert may still be the favorite, and the GOP seems confident, but these numbers indicate could be a late night Tuesday. 

It just keep looking worse for the GOP in the Assembly, although they clearly are hoping the ground efforts and person touch pay off. But David Gardner (1,800), Derek Armstrong (1,600), Stephen Silberkraus (1,000) and Brent Jones (1,500) have deficits that in an ordinary year would be insurmountable. 

 

UPDATED, 11/3/16, 6:45 AM

 

All but seven rural counties in -- and they will reduce the statewide Dem lead by a few hundred -- but here's what it looks like after Day 12:

The Democrats had a big day in Clark -- plus 4,800 -- and a bad day in Washoe -- minus 500 -- but that Clark firewall is now at 55,500 -- exactly what it was after 12 days in 2012, when they got to 71,000 by the end of the fortnight. (On this day in 2012, it was: Dems plus 4,000 in Clark and minus 300 in Washoe.) But the 2012 statewide firewall of 48,000 after early/mail voting is still out of reach.

More than 617,000 have voted in Nevada. That's 42.2 percent. More than half the vote is in unless Election Day turnout is well above the norm.

The Democrats have a 34,500 voter statewide lead (probably under 34,000 after the rurals are posted). That's 42-37, or about a point under the registration lead. The GOP turnout edge is 1.2 points.

Clark turnout definitely is down by percentage, despite Team Clinton's boasts of record numbers. Those additional 150,000 registered voters since '12 matter. It was 47 percent at this time in 2012; it's 42 percent now. 

But raw votes are raw votes. And that 55,500 firewall means a lot, indicating Trump will be down double digits when those first numbers pop up on Tuesday evening. Joe Heck also can't like to see these numbers, especially since everything I have heard indicates he is being crushed in the early vote.

Keep an eye on those rural numbers. The raw vote lead for the GOP there is now above 22,000, and probably will be closer to 23,000 once Elko and the six others post. 

Clark turnout down, Washoe and rurals up is not the formula the Dems had in mind. But registration matters, and if they do close to what they usually do in the last two days (they added about 16,000 voters, which somehow seems unlikely this year, in the last two days of 2012), the blue wave not only will wash over the top of the ticket, but will certainly turn those two House seats blue and maybe both houses of the Legislature, too.

We will know a lot more tonight.

 

UPDATED, 11/2/16, 12:45 PM

The full statewide numbers are in, and here's what we know:

----It still looks like 2012, but the Democrats are lagging behind their 2012 percentages thanks to two straight Democrtaic doldrums days -- Days 10 and 11 are usually bad for them. Yes, the Clark firewall is at 50,000, as it was four years ago. But the statewide lead is 30,500, which is about 4,500 behind four years ago. Clark County turnout is about a point below its registration numbers and the rurals are slightly up and providing the GOP with a 21,500 raw vote lead -- that's out of only 72,500 cast (55-25!).

----The additional voters since 2012 -- 200,000 statewide and 150,000 in Clark -- mean that Democrats are underperforming their registration and Republicans are overperforming theirs. Republicans have almost a point and a half advantage now (it was 1 point at the end of early voting in 2012). In Clark, the GOP has a 2.5 percent turnout advantage. Again, he said, sounding like a broken record, this is why registration matters so much. That 142,000 Clark lead and 89,000 statewide edge is why the Democrats continue to build a firewall.

----The Washoe numbers are heartening for the Dems despite two straight days of significant losses. They have almost a 4-point turnout advantage there, and an 1,800 lead (it was 300 this time in 2012).

----The overall numbers still look ominous for Trump: Give him the best of it -- he is getting 90 percent of the GOP base and Hillary is getting 90 percent of Dems and he is winning indies by 20 points -- and he is still losing the state by 3,000 votes. And no one I know believes those percentages will happen. So long as Hillary holds the base, she is in solid shape here.

----561,000 people have voted now. That's 38 percent. If turnout is 80 percent overall when all is said and done, about half the vote is in and things are getting baked.

Numbers:

561,901 voted

D-238,816 (43 percent)

R-202,219 (37 percent)

Others - 114,866 (20 percent)

 

Right about at actual registration margin.

Bottom line: The next three days are critical. I still think the Dems need to do what they usually do -- add thousands to their total every day -- to feel comfortable. It was 48,000 statewide by the end in 2012. They are not going to get there. The GOP lead in the rurals was about 23,000 by the end of early voting in 2012 -- it's going to be substantially higher this cycle. Watch that cow county turnout and those GOP ballot margins. The Democrats could still get to the 71,000 margin in Clark they had in 2012, but I wouldn't bet a lot of money on it.

Other notes:

Both GOP House districts in play look bleak for the GOP, although CD4, where the Dems have a 17,500 raw ballot lead, looks safer than CD3, where it is 6,200. Very hard for Ruben Kihuen to lose, and Danny Tarkanian needs some help.

The Democrats lost Senate District 5 for the first time but won in Senate District 6 and lost in Senate District 15. But they all stayed about status quo, which means Dems have sizable leads in each one. The GOP seems confident in SD15 despite the numbers, but I think all three are in play.

Not much change in any of those competitive Assembly races, either. All four are heavy lean D right now.

 

 

UPDATED, 11/2/16, 5:45 AM

The SOS did not update its site with rural numbers last night, so those have yet to come in. But here's what we know:

The Dems won the urban counties by a net 1,800 votes Tuesday -- 2,200 in Clark and lost 400 in Washoe. So the statewide lead is about 33,000 without the rurals, which means it's probably closer to 32,000.

The Clark raw vote lead is now about 50,000, which means it is almost identical to 2012, although the percentage is slightly less because of the 150,000 additional voters. Washoe is at about 1,900 -- that's six times better what it was at this time in 2012.

What I wrote in 2012 seems apt -- just substitute Donald Trump for Mitt Romney:

There are still three days left for the Democrats to solidfy their firewall, but they will not get to 83,000 (the 2008 lead), nor will they get to the 12,000-voter lead in Washoe County, where the Democrats have a 500-vote early/mail lead (absentees not in yet all counted in Washoe) out of about 80,000 votes cast. I still think the Democrats have an advantage unless Mitt Romney is winning independents by 20 points or more -- and depending on just how large Election Day turnout is for Republicans. My guess is Democratic Party strategists feel good, but not quite secure.

Bottom line: The next three days are huge, starting with today when Hillary Clinton comes to try to juice the early vote in Vegas. Remember the last day (Friday) is always the biggest. If the Dems can add 10,000 votes to their lead, they will feel pretty good. If they add 20,000, which would bring it to the same number as the end of 2012, it's big trouble for Trump and Senate hopeful Joe Heck, and we can start talking about a mini-wave at least.

 

 

 

 

UPDATED, 11/1/16, 1:15 PM:

The Democrats are ahead right at registration now -- 6 percent statewide. 31,000 votes. Very close to 2012 numbers. The GOP has a half a percent turnout advantage statewide -- it was 1.1 percent by the end of early/mail voting in 2012.

511,000 people have voted. That's 35 percent.

In 2012, 418,000 Nevadans had voted, or about 33 percent.

In 2012, 69 percent of all votes were cast before Election Day -- 61 percent early and 8 percent by mail. Total turnout was 81 percent.

If turnout is 81 percent again, that's just under 1.2 million voters. And that means we are at 43 percent of total turnout already if it's 81 percent again. Let's say it gets as high as 85 percent (seems unlikely), that still means 40 percent-plus is in.

By the end of Tuesday, close to half of all Nevadans who will cast ballots probably will have voted.

That's why these raw ballot leads are so important. There may not be enough votes left by next Tuesday to overcome deficits up and down the ballot, no matter how many Republicans turn out. The GOP, as always, counts on Election Day to bridge the gap in races they have managed to keep close during early/mail voting.

Quick thoughts on legislative races:

In the Senate, the Democrats have substantial ballot leads in all three key districts, ranging from 1,300 in the northern race to more than 2,000 in both the southern races. The Republicans remain optimistic in all three, but these numbers are daunting, especially because, I think, nearly half of the vote is in.

In the Assembly, while the GOP has stanched progress in a couple of districts, I don't see how Brent Jones survives, and the three the leaders are trying to save -- Derek Armstrong, Stephen Silberkraus and David Gardner -- all better be getting substantial crossover and indie votes or they are dead, too.

 

UPDATED, 11/1/16, 11:15 AM:

 

So what do we know after 10 days? 

 

Even though this looks a lot like 2012 in the data, we know that Hillary Clinton is not Barack Obama. There is not the same enthusiasm, and Democrats are having to work harder. They knew this going in, which is why so many groups -- Culinary, For Our Future, League of Conservation Voters, Mi Familia Vota -- are on the ground, especially helping in Washoe.

 

Clinton campaign manager Robby Mook told me turnout is up here by 15 percent, which is true in raw votes. But that's how much the increase was in registered voters, so it's a wash.

 

Turnout advantage after 10 days (ended up about even after all 14 days in 2012, with about 60 percent of each party turning out): GOP has a half-percent lead statewide. Dems lead in the raw vote count by 31,000 after the final rurals came in, still with 48,000 lead in Clark and 2,300 in Washoe. GOP up more than 19,000 in rurals.

 

Clark turnout is slightly below registration and rurals and Washoe slightly above.

 

Context: Final early/mail vote in 2012 showed 48,000 statewide lead and 71,000 in Clark.

 

The Day 10 numbers are bad but not catastrophic, a little worse than 2012 and probably worse than what the Dems thought might happen and better for the GOP. That's especially true in CD3, where mobile sites might have helped them hit a new high -- winning the day by 1,000 votes.

 

 

Thumbnail because I am pressed for time this morning and will post more later:

 

In both congressional districts, the Democrats maintain substantial leads despite the hit in CD3: 

CD3 -- 6,600 raw votes, 41-37, 2 points above registration. Nearly 40 percent have voted there.

CD4 (Clark) -- 16,500 raw votes, 47-33, a point below registration. About 35 percent turnout.

 

Democrats bumped up their totals in AD9 and AD21, but lost by a handful of votes in AD29 and AD35.

 

State Senate 5 was close to a draw, the Dems won Senate 6 by 150 and won Senate 15 by a handful.

 

 

 

UPDATED, 11/1/16, 6:30 AM:

The Democrats had a bad day on Day 10, as they did in 2012, winning Clark County by only 900 early votes, but still moving the firewall to just under 48,000 votes. That's very close to 2012, although the percentage is slightly less.

The Democratic lead in Clark is now under 14 points, which is right at registration. GOP turnout in Clark is now nearly 2 points above the Dems. I repeat myself, but: That's why that 142,000 registration lead in the South matters.

The Dems lost Washoe Monday by 350 votes, which reduced their lead there to about 2,300 voters. 

The statewide firewall -- some rurals still out -- is at 33,500 -- it will probably be down to 33,000 by the time the cows come home. The final numbers likely will show a narrow Monday statewide loss for the Dems, which is exactly what happened in 2012.

It still looks a lot like 2012: At this time four years ago, the Democrats were up 14 points in Clark, about 48,000 votes and Republicans had a 2-point partisan turnout edge (it ended at 1.6 percent for the GOP). I wrote earlier that if Clinton holds 90 percent of her base, even if Trump makes progress with indies (and I think he is based on what I have seen), she could still lose them by 20 and win the state by 2 points. That still holds true.

It's simple math -- I'll show you what happens under the current numbers if Hillary wins Dems 90-5, Trump wins Republicans 90-5 and Trump wins indies 55-35 -- this is a very optimistic scenario for Trump, especially because half the vote was banked before the Comey revelations:

Clinton -- 242,000 votes

Trump -- 233,000 votes

None/rest -- 30,000 votes

Half a million Nevadans have now voted -- that's above 34 percent.

The real question, and I will have detailed numbers later today, is what impact the terrible day for Democrats had down ballot.

 

 

 

UPDATED, 10/31/16, 11:45 AM:

More fun with math:

Total numbers so far show slightly more than 450,000 people have voted; that's 31 percent.

Breakdown is: 44 percent Dem, 36 percent GOP and 20 percent indies. That's almost exactly what the final numbers after Election Day looked like in 2012. (Indie turnout is running about a dozen points behind major party turnout.)

7,200 people had voted in Clark by 11 today. Going to be a low turnout day, and it was a bad one for Dems in 2012, relative to other days.

 

UPDATED, 10/31/16, 9:45 AM:

Trump almost surely is losing Nevada and remains likely to lose it, unless voting patterns are very different from 2012 – that is, unless Hillary Clinton is hemorrhaging her base more than Donald Trump and indies are voting big for him. Neither of those things seems likely, and both are contradicted by polls.

Republicans have to hope for a big last five days and a game-changing Election Day. As I have repeatedly said, recent patterns show 60 percent vote before Election Day. Here’s what happened in 2012:

It was 61 percent turnout before Election Day.

Overall turnout after all the votes were counted: 42 percent to 36 percent. Democrats had 34 percent turnout and Republicans had 29 percent of the total. Democrats actually won Election Day, 39-35.

On Election Day, 23 percent of Dems turned out and 25 percent of Republicans. But that 2 percent edge was negated by the 90,000-voter registration advantage, which is why the Dems still won Election Day by 12,000 votes.

In Clark, where Dems built a 71,000-vote firewall, 57 percent turned out before Election Day. President Obama had a 69,500-vote lead among those who voted before Election Day in Clark. His final Clark number: 100,000. So he added 30,500 votes on Election Day.

Obama won the state by 68,000 votes. Think of it this way: If he added zero votes on Election Day in Clark, he still would have won the state. Obama won Washoe by 7,000 votes and lost rural Nevada by 39,000 votes.

Now you see how Trump’s task appears to be insurmountable unless the Republicans reverse historical trends.

Keep an eye on what happens today: Day 10 in 2012 was a bad one (so was Day 11), relatively speaking, for the Dems in Clark. They only won by 1,600 votes.

Updated deep dive with five days to go:

CD3:  Repubs played Dems to a draw on Sunday, so still 7,800 raw vote difference and nearly 140,000 votes cast. Dems have a 2 percent reg lead there and are still up 6 percent in the early/mail vote. Dem turnout is 3 percent above GOP. 35 percent of the vote is in. Tark needs help .

CD4 (Clark part): Dems added 1,000 to their margin in Clark on Sunday, which means they have a 16,000-plus raw vote lead or 15.5 percent, slightly above registration. As I have said, I think Cresent Hardy will get Dem votes and his evisceration of Ruben Kihuen on TV may cause indies to break for him, and he will win big in the rurals. GOP turnout actually is a point above Dem in this area. But if Dems have a 20,000 raw vote lead in Clark going into Election Day, without serious anomalies, this seems impossible.

SD5 (Woodhouse-Buck): The Democrats added more than 150 votes to their lead, so: 2,400 raw vote lead, and a 10 point percentage edge. That’s 5 points over registration. Very difficult for GOP to take this seat.

SD6 (Seaman-Cannizarro): The Republicans lost here for the first time in four days (Sun City voting was fun while it lasted), by about 100 votes. So: 2,000 raw vote lead, 7 points, which is now at registration. Seaman still has a chance if she is getting a lot of crossover and indies.

SD15 (Reese-Gansert): Dems won by 7 votes here Sunday, so: still 43-38 Dem lead and 1,200 raw votes. They are 7 points over registration here. The GOP folks still seem confident here, despite these numbers, which indicate a possible upset coming.

Assembly districts of note:

AD9 (Gardner): Dems picked up more than 100 more votes here, and it is now a 1,400-vote lead. Still not good for incumbent GOPer David Gardner.

AD21 (Armstrong): Dems picked up almost 100 votes here on Saturday, so it’s now almost 1,400 here. Derek Armstrong still needs a lot of help.

AD29 (Silberkraus): Dems by 70 on Saturday,  so at about 1,000. Stephen Silberkraus still in jeopardy.

AD35 (Jones): A Dem gain of about 40, so now about 1,450. I hope Real Water is doing well, Brent Jones.

Bottom line: Those are awfully big leads for Assembly races where relatively few votes have been cast. But I have one caveat: These are still races that can overcome partisan advantages with hard work and grassroots.

 

UPDATED, 10/31/16, 6 AM

The statewide numbers after Sunday, with the rurals closed, show the Democratic lead is a little more than 34,000.  Washoe was essentially a draw -- Democrats by 15 -- and the Dems won Clark by about 2,800 even though Donald Trump was in town and they bused rallygoers (well, about 20 out of 8,000 of them!) to the polls.

So the percentage lead in Clark is a little down from 2012 for the Dems, but the overall firewall is about the same: 47,000 votes. Washoe is at 2,500, which is almost twice what it was in 2012.The rurals remain plus 15,000 for the GOP

Dems are up 15 points in Clark (it was 16 in 2012), which is a point above registration, and that firewall is going to get to 60,000-plus, unless something strange happens. I repeat: Trump appears to have no path here, but the Senate race is not quite over yet.

The Clark percentage of the overall vote is a tick under 70 percent, or about half a percent above actual registration. Washoe is almost half a percent above and the rurals are more than a half point below.

Statewide turnout is now 31 percent of all voters -- Democrats are at 34.6 percent, Republicans at 33.9 percent and others at 23 percent. The Republicans actually have a slight turnout lead in Clark -- 34.9 percent to 34.3 percent, showing why that 142,000 Democratic registration lead matters. The Democratic turnout edge in Washoe is nearly 4 points (37-33).

The 2012 comparison after all votes were in at the end of the two weeks of early/mail voting:

Statewide: Dems plus 48,000

Clark: Dems plus 71,000

Washoe: GOP plus 900

Rurals: GOP plus 22,100

Very, very similar.

UPDATED, 10/30/16, 11 AM

So here's how daunting it is for Donald Trump right now in Nevada, with six days left in early/mail voting, which is usually 60 percent or so of all votes cast:

If you assume he and Hillary Clinton get 90 percent of their bases (and almost no poll shows Trump doing that well), and she loses indies by 20 points (and most here show her up with indies) she STILL wins the state by 2 points.

Let's suppose Republicans end up coalescing and he gets 90 percent and she gets only 80 percent of Democrats -- a highly unlikely scenario, to put it mildly -- he would still need to win indies by at least 10 points to win the state.

That is what the nearly 90,000-voter registration edge does for the Democrats in Nevada -- if they have their usual turnout, it's almost impossible for Trump to win.

Here’s what’s happening in the down-ballot races of interests after one week:

CD3: The Democrats added 500 votes to their lead in Rep. Joe Heck’s district, which means : 7,800 raw vote lead. Just under 43-36, Dems, which is just under 5 points over registration (37-35). Dem turnout is 35 percent and GOP turnout is 32 percent. If GOP can't even the turnout advantage by the end of early voting, hard to see how Danny Tarkanian has any chance.

CD4: The Democrats added nearly 1,500 votes to their lead in the Clark portion (85 percent of the vote) in Rep. Cresent Hardy’s district, which means: 15,000 raw vote lead, 48-33, same as registration margin. GOP turnout is a point higher than Dem turnout, but I repeat: The rurals can’t save Hardy here unless he is getting a lot of Dem votes because of the TV pummeling of Ruben Kihuen.

SD5 (Woodhouse-Buck): The Democrats added more than 100 votes to their lead, so: 2,200 raw vote lead, and a 10 point percentage edge. That’s 5 points over registration. GOP is slowing the daily attrition, but Joyce Woodhouse is looking strong.

SD6 (Seaman-Cannizarro): The Republicans won here for the third straight day, but only by 22 votes. So: 1,900 raw vote lead, 7.5 points, which is now just half a point over registration. But registration advantages matter,

SD15 (Reese-Gansert): Dems won by 18 votes here Friday, so: 43-38 Dem lead and 1,200 raw votes. They are 7 points over registration here.

Assembly districts of note:

AD9 (Gardner): Dems picked up more than 100 more votes here, and it is now a 1,300-vote lead. Still not good for incumbent GOPer David Gardner.

AD21 (Armstrong): Dems picked up almost 100 votes here on Saturday, so it’s now almost 1,300 here. Derek Armstrong still needs a lot of help.

AD29 (Silberkraus): Dems by 50 on Saturday,  so at about 900. Stephen Silberkraus still in jeopardy.

 

AD35 (Jones): A Dem gain of about 150, so now about 1,400. I hope Real Water is doing well, Brent Jones.

So: Some of the gains are not as large in the Legislature, but the smart betting now, with six days left and much to happen, is that it goes all blue.

UPDATED, 10/30/16, 8 AM:

Democrats won Clark by 4,000 votes, firewall now at 44,000.  Washoe was essentially a draw, Dems by 19 votes. They still lead there by about 2,500.

Statewide, the Dems increased their lead by about 3,000 votes, but a few rural counties have not yet been posted (the rural lead for the GOP is now more than 15,000). Statewide, the Dems lead is now about 31,000 votes. 

The statewide Dem lead is now at 7 points, or 1 point above registration. That's where the 89,000-voter registration edge looks so critical. The Dem lead in Clark is 47-32, which is slightly less than 2012, when it was 49-33. The raw vote lead in Washoe is twice what it was at this time in 2012.

In 2012, on the first day of the second week, the Democrats did a little better, winning Clark by 5,000 and Washoe by 120.

So still a lot like 2012, with GOP showing slight improvement.

UPDATED, 10/29/16, NOON:

One week in the books, and it still looks a lot like 2012, although the GOP is doing slightly better by percentage as the Democrats still have built a significant firewall.

With the uncertainty of the new email revelations – and more October Surprises surely on the way – here’s what the data tells us (some rurals are still not posted by SOS, so GOP will gain a few hundred votes, I’d guess):

----Capsule: The Dems lead statewide by 28,500 votes. I still think Hillary Clinton has to be a solid favorite based on these numbers, but based on Dean Heller bucking the tide in 2012, I still think Joe Heck has a chance. The state Senate is looking bleak after a week for the GOP, but not ready to call it yet. On the Assembly side, the numbers also are daunting, but they were before early voting, so the huge field operation and attempt to get crossover/indie votes in key districts from GOP Leader Paul Anderson will either pay off.....or it won't. Remember, 60 percent usually vote before Election Day, but will that be true this cycle?

----Raw vote turnout is up by 20,000-plus votes, but the additional 200,000 registered voters means it’s down by percentage. Democrats are down 2 points and Republicans are down 3. The actual difference statewide after a week: .6 percent

----The Clark Democratic firewall is almost identical to what it was in 2012: About 40,000 votes. Again, the additional voters (150,000 in Clark) mean the percentage lead is slightly down. But this is more about banking votes for the Dems, and they are on track to get to 70,000 if the second week this cycle mirrors four years ago. Friday was a decent day for the Dems – a 3,600-vote win – but the GOP continued to do better than registration.

----Washoe is where the anomalies are occurring. Republicans are where they were in 2012, but the Dems are 3 points above what they were four years ago. That 2,500-vote Dem lead is worth watching.

----The GOP is banking votes in the rurals and has a 13,000-vote lead. That’s slightly bigger than the same time in 2012.

-----The GOP usually ends up with a slight turnout advantage which is why, I remind everyone again, that push by the Dems to register voters and get the deadline extended could be critical, especially if the Senate race is close. The swarming on the ground by progressive groups this cycle in Nevada and the amount of money spent on voter registration an contact by the likes of For Our Future and the League of Conservation Voters has been substantial. Add that to the usual yeoman’s work by THE Latino turnout operation in Nevada – the Culinary union – and the advantage is obvious, despite relentless efforts by Engage Nevada and the RNC to make up for the hollow shell that is the Nevada GOP.

Bottom line: Unless Donald Trump is getting tremendous crossover and indie support, he cannot win the state. That is, unless the second week of early voting, which the Dems won statewide by about 20,000 votes in 2012, turns out to be much different.

Here’s the deep dive into the data: 

2016: 379,000, or 26 percent

2012: 358,000, or 29 percent

Dems in 2016: 29 percent

Dems in 2012: 31 percent

GOP in 2016: 28 percent

GOP in 2012: 31 percent

Statewide raw vote lead for Dems in 2016: 28,508

Statewide raw vote lead for Dems in 2012: 28,652

Percentage lead for Dems in 2016: 44-36.5, 7.5 points, about 1.5 points above registration

Percentage lead for Dems in 2012: 45-37, 8 points, about 1 point above registration

Clark Dem lead in 2016: 39,500

Clark lead in 2012: 39,800

Clark Dem turnout in 2016: 28 percent

Clark Dem turnout in 2012: 31 percent

Clark GOP turnout in 2016: 28 percent

Clark GOP turnout in 2012: 31 percent

Washoe Dem lead in 2016: 2,500

Washoe Dem lead in 2012: 850

Washoe Dem turnout in 2016: 32 percent

Washoe Dem turnout in 2012: 29 percent

Washoe GOP turnout in 2016: 28 percent

Washoe GOP turnout in 2012: 28 percent

Rural GOP lead in 2016: 13,500

Rural GOP lead in 2012: 12,000

Here’s what’s happening in the down-ballot races of interests after one week:

CD3: The Democrats added only 300 votes to their lead in Rep. Joe Heck’s district, which means the first-week numbers look like this: 7,300 raw vote lead. 43-36, Dems, which is 5 points over registration (37-35). Dem turnout is 31.6 percent and GOP turnout is 28.5 percent, which is quite ominous for the Republicans. 23,000 indies have voted: Danny Tarkanian is going to need a lot of them. Overall turnout is 28 percent, which means about a third of the total vote is in already.

CD4: The Democrats added 800 votes to their lead in the Clark portion (85 percent of the vote) in Rep. Cresent Hardy’s district, which means the first week looks like this: 13,500 raw vote lead, 48-32, 1 point above registration. GOP turnout is actually about a point higher than Dem turnout, but the rurals can’t save Hardy here unless those 16,500 indies have broken for him. Overall turnout is about 27 percent.

SD5 (Woodhouse-Buck): The Democrats added 150 votes to their lead Friday, so: 2,100 raw vote lead, and a nearly 11 point percentage edge. That’s 6 points over registration. GOP is slowing the daily attrition, but Joyce Woodhouse is looking strong.

SD6 (Seaman-Cannizarro): The Republicans won here for the second straight day (thanks, oldsters in Sun City!), but only by 75 votes. So: 2,000 raw vote lead, 8 points, which is now just 1 over registration. High turnout here – 31 percent. Still a possible pickup for Dems.

SD15 (Reese-Gansert): Dems won by 75 votes here Friday, so: 44-38 Dem lead and 1,200 raw votes. They are 7 points over registration here, so another possible pickup.

Assembly districts of note:

AD9 (Gardner): Dems picked up more than 100 more votes here, and it is now almost a 1,200-vote lead. Not good for incumbent GOPer David Gardner.

AD21 (Armstrong): Dems picked up 150 votes here on Friday, so it’s now almost 1,200 here, too. Derek Armstrong needs help.

AD29 (Silberkraus): Handful of votes win for GOP on Friday, so still at about 850. Stephen Silberkraus also in jeopardy.

AD35 (Jones): Another gain of 100, so now almost 1,300. Bye, bye Brent Jones.

Bottom line: I’ll take a closer look at some other districts soon, but these four are all leaning Dem, which means the Assembly could be up to 24 or 25 seats for the Dems if the wave builds. They probably only have to win 1 or 2 of these to take control.

 

 

 

UPDATED, 10/28/16, 2:30 PM

Latest statewide numbers after six days, with all rurals save a few small ones in, show the statewide Democratic margin reduced to 26,342. And:

Overall turnout (almost 334,000) is just below 23 percent of registered voters.

Clark is 67.5 percent of the turnout, or 2 points under its actual percentage of registration; Washoe is at 19 percent, or a point above actual; and the rurals are at 13.5 percent, or a point above. So Democrats clearly want to juice Clark turnout.

Overall turnout so far: Democrats, 25.6 percent; Republicans, 24.8 percent; others, 16.2 percent

 

UPDATED, 10/28/16, 9:30 AM

Breakdowns are in for Clark and key legislative races, so the deep dive, which shows despite overnight improvements, the GOP remains in danger of losing....everything -- watch what happens today, a holiday here, and over the weekend:

CD3: Registration is 37-35, Democratic.  Dem lead is now 43-36, 1 point improvement for GOP over previous day. 7,100 raw-vote lead for Democrats, so small pickup. Slowdown in Dem machine reflected here. Only won early vote by 300 on Thursday.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. Dem lead is 49-32, so GOP picked up 2 points, but Dems still gained about 1,000 votes. Almost at 13,000 raw vote lead now. 

SD5 (Woodhouse): Registration is 39-34, Democrats. Dem lead is 46-34, so GOP picks up 1 point overnight but still 7 points above registration difference. Almost 2,000 raw vote lead. 

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. Republicans won early vote on Thursday by more than 100 votes, with help from Sun City, by more than 100 votes. Dem lead is 46-36, so GOP improved by almost 4 points. Raw vote lead now down to 2,000. 

SD15 (Gansert-Reese): Registration is 38-36, R. Dem lead is 44-38. Good Washoe day helped GOP win by 40 votes, but lead still 8 points above reg difference. Still about 1,100 raw vote lead.

Brief glimpses at key Assembly districts, where GOP played to a near-draw in a couple but the leads are still very large:

AD 9 (Gardner): D+1,100 votes (was 1,000)

AD 21 (Armstrong): D+1,050 votes (was 900)

AD 29 (Silberkraus): D+850 votes (was 900-- GOP won EV on Thursday)

AD 35 (Jones): D+1,150 votes (was 1,100)

Bottom line: The midweek slowdown for the Dem machine continues, and the GOP made a little progress. In this case, progress generally means not losing by as much.

 

 

 

 

UPDATED, 10/28/16, 6:50 AM

As the sun rises on the last day of the first week of early voting, Republicans have recovered from the disastrous initial weekend as the Democratic machine has slowed a bit.

The Democrats continue to build a raw vote lead, now at 27,000 statewide (with a few rurals outstanding that will reduce it a bit) and 36,000 in Clark County (it was 40,000 by the end of Week One in 2012). But the Republicans have slowed their daily hemorraghing, so the statewide number is now much closer to actual registration: The GOP is down 8 points (44-36) when actual registration is 39-33. Once again, this shows why that 89,000-voter statewide registration edge the Democrats built could be so critical.

The 2012 Clark comparisons:

Raw vote lead in 2012 after six days: 36,000

Raw vote lead in 2016 after six days: 36,000

The difference is that there are 1 million registered voters this cycle compared to 850,000 in 2012. So it was 50-33 in 2012 and 48-32 now, so Democrats are 2 points under their 2012 performance in Clark and the Republicans are 1 point under theirs.

The Democratic margin in Clark on Day 6 (3,000 votes and 44-34) was the party’s worst day so far.

The Washoe numbers Thursday – GOP plus 370 – show a correction, at least for a day. The Democrats still have a 2,400 raw vote lead and now have a 43-39 lead in a county that is 37-35, GOP.

So where are we?

It still looks a lot like 2012, and the next few days, including Friday, which is a holiday (Nevada Day), probably have Democrats a little leery, as they usually do not perform as well on Days 7-11. 

The benchmark to look at Saturday after one week of data are in is what 2012 looked like at that time:

Statewide: Democrats had a 28,500-voted lead, which they will almost certainly surpass this year, although the percentage will be slightly down because of the 200,000 additional statewide voters. 

Clark: The Dems had a 40,000-vote lead, which they may not get to this cycle, which will mean raw vote and percentage numbers are down from 2012.

Washoe: The GOP had a 1,000-voted lead, which will not happen his cycle. The Democrats will have a lead, probably of at least 2,000 votes.

Rurals: The GOP had about a 10,500-voted lead in 2012 after the first week; it is more than 11,000 already this cycle. Watch the rurals.

331,000 people have voted early or by mail so far in 2016 (a few hundred more after those rural tallies come in). That’s about 22 percent.

At the end of the first week in 2012, 358,000 had voted, or 29 percent. Doesn’t look like it will get to that percentage.

Democrats are turning out about 1 percent more than Republicans right now (it's almost even in Clark). The GOP usually ends with a slight turnout advantage.

Don’t forget indies: There are more of them this cycle. 64,000 have turned out in the first six days. By the end of the first week in 2012, 62,000 had turned out. The GOP will need them to overcome the Democratic advantage. Polling I trust shows this has not happened.

Finally, a note on raw votes and why they matter. Consider why the Clark firewall is so important, or has been in the past:

In 2012, at the end of early voting, the Democrats had a 71,000 raw vote lead in Clark. So how did those people vote?

Here’s what those numbers showed when they were tallied on Election Night:

President-- 273,364-203,818; that’s 69,500 for Obama (56-42)

Senate-- 245,258-194,250; that’s 51,000 for Shelley Berkley (52-41)

So Berkley ran well behind the president. But the raw vote in ballots was very predictive of the actual votes. So watch how close the Democrats can get to that 71,000 number (even with the new voters). If they get there, it’s almost certainly over in the White House race here, and the Senate race could be a nail-biter again.

 

UPDATED, 10/27/16, 10:00 AM

About 4,000 rural votes came in such my last update, knocking the Dem lead down to 25,284 statewide. The rurals are saving the Republicans from a total blowout -- GOP now has a 10,000 raw vote lead there.

The Republicans did slightly better, as they usually do, on the fifth day, but still lost ground. This is where the Democratic registration lead really is so important and has been in past elections.

The similarities to 2012 continue.

In 2012, 151,281 people had voted early (not counting mail), or about 18 percent. The Democratic firewall had begun to build toward a 71,000 total by the fortnight's end.

In 2016, that number is 170,796, or 17 percent.

The deep dive shows some stabilization for the GOP in lower-ballot races (legislative), but the Democrats continue to build substantial leads because of their registration advantage. Democratic turnout statewide (22.1 percent) is still slightly above GOP turnout (20.9 percent).

The deep dive:

CD3: Registration is 37-35, Democratic.  Dem lead is now 44-36, 1 point improvement for GOP over previous day. 7,000 raw-vote lead for Democrats. Tough to overcome if this keeps going.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. Dem lead is 50-31, so GOP picked up a point, but Dems still 5 points above registration advantage. Almost a 12,000 raw vote lead now. Still don’t see how the rurals save Cresent Hardy because they are only 15 percent or so of the district.

SD5 (Woodhouse): Registration is 39-34, Democrats. Dem lead is 47-34, so GOP picks up 1 point overnight but still 8 points above registration difference. 1,850 raw vote lead. That is very large in a Senate district.

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. Dem lead is 48-34, so GOP improved by 2 points. 2,100 raw vote lead now. Again, very large in one of these political subdivisions.

SD15 (Gansert-Reese): Registration is 38-36, R. Dem lead is 45-38. Good Washoe day helped GOP play almost to a draw on Tuesday, but lead still 9 points above reg difference. Still 1,100 raw vote lead. This race is definitely in play. GOP could lose all three seats.

Brief glimpses at key Assembly districts, where GOP played to a near-draw in a couple but the leads are still very large:

AD 9 (Gardner): D+1,000 votes (was 850)

AD 21 (Armstrong): D+900 votes (no change)

AD 29 (Silberkraus): D+900 votes (no change)

AD 35 (Jones): D+1,100 votes (was 1,000)

Bottom line: Slight midweek slowdown for the Dem machine, as is the norm, and GOP made a little progress. Next few days will be interesting to watch as the GOP usually does better. Clark firewall still on track, and GOP not doing what it needs to in Washoe. 

 

UPDATED, 10/27/16, 6 AM

The latest statewide numbersmissing a few rural counties, show a 26,500 lead for the Dems in raw votes. That's 45 percent to 36 percent, the same percentage lead and 3,500 raw votes more than they had at this point in 2012 and about 3 points above the actual registration difference.

The Clark firewall increased by more than 4,000 votes to nearly 33,000 (it was 30,000 at this time in 2012 but with 150,000 fewer registered voters). But it was the GOP's best day so far, and slightly better than its performance in 2012 on Day 5.

The Dems won Washoe by a handful of votes, maintaining a nearly 2,800-vote lead there (Day 5 in 2012 saw the GOP win by 200 votes.)

The rurals continue to go big for the GOP, as expected. Totals there now show a more than 9,000-voter lead for the Republicans.

Actual turnout so far by region shows this split: 68.3 percent from Clark (a point below actual percentage of statewide voters), 19.4 percent from Washoe (about a point and a half above actual) and the rurals at 12.2 percent (or just a tick below actual).

Bottom line: It's still very much like 2012.

 

 

UPDATED, 10/26/16, 3 PM

Some have been asking for some 2012/2016 comparisons. Your wish is my command:

2012: By Day 4, 210,000 had voted early or by mail -- that was 17 percent. Raw vote lead was 21,000 votes.

2016: By Day 4, 230,000, or 16 percent. The raw vote lead is almost 24,000 votes.

2012 in Clark: By Day 4, 146,000 had voted, or 17 percent. The raw vote lead for Dems was 25,000 in Clark.

2016 in Clark: By Day 4, 155,000 had voted, or 16 percent. The raw vote lead for Dems is 28,500.

2012: The lead for Dems statewide was 48-36 after four days. The lead in Clark was 50-33.

2016: The lead for Dems statewide is 46-35 after four days. The lead in Clark is 49-31.

Very similar to 2012, as I have been saying.

In 2012, Barack Obama won Nevada by almost 7 points after Democrats amassed a 70,000 early vote lead in Clark and 48,000 statewide. Dean Heller won by 12,000 votes.

UPDATED, 10/26/16, 10:30 AM

 

The latest SOS update shows a statewide raw vote lead of 23,874 for the Democrats. (It was 21,000 in 2012 at this time, but there are 200,000 more voters, so very close in percentages.) The percentages are 46-35. So Democrats are almost 7 points above their registration, and Republicans are 2 points above. This is why registration matters, and why that huge Democratic push at the end of registration, including getting that extension of time after warning the SOS, mattered so much.

What to watch:

--Enthusiasm gap seems obvious so far. This will kill down-ballot Republicans if it continues.

--GOP candidates will need indies to win, and they will need them big league. No evidence they are getting them.

--Hispanic turnout seems high so far in key precincts.

--The Washoe numbers are key, and Dems doing well there. It’s no accident major GOP surrogates have been going there — Graham, McCain, Pence. They know.

--The GOP usually does better in the middle days of early voting in Clark, not winning but coming closer. Keep an eye out for that.

--The Clark firewall will kill Trump. Can Heck save himself even if it gets to 70,000-plus Democratic advantage? Doubtful.

The deep dive shows the wave still building — I’ve added the northern state Senate seat:

CD3: Registration is 37-35, Democratic.  Dem lead is now 44-35, 3 point improvement for GOP over previous day. 6,524 raw-vote lead for Democrats. Still a lead that keeps getting bigger.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. Dem lead is 51-31, so GOP picked up a point, but Dems still 5 points above registration advantage. Almost a 10,000 raw vote lead now. Don’t see how the rurals save Cresent Hardy.

SD5 (Woodhouse): Registration is 39-34, Democrats. Dem lead is 47-33, so GOP picks up 1 point overnight but still 9 points above registration difference. 1,755 raw vote lead.  Real danger sign: Dem turnout is 22 percent and GOP turnout is 18 percent. Down-ballot bleeding sign? Joyce Woodhouse looking solid.

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. Dem lead is 49-33, same as after Monday and still 9 points above registration difference. 2,000 raw vote lead now. And GOP turnout is 4 points under Dem turnout.

SD15 (Gansert-Reese): Registration is 38-36, R. Dem lead is 45-37. Good Washoe day helped GOP play almost to a draw on Tuesday, but lead still 10 points above reg difference. 1,100 raw vote lead. Remember, this is one early on that the GOP thought was a lock. It ain’t.

Brief glimpses at key Assembly districts, where Dem leads keep growing and some good GOP freshmen (and some bad ones) seem in big, big trouble. It would be a miracle now if the Assembly does not go Dem — the question might be how many seats will flip. Ten did the other way in 2014.

AD 9 (Gardner): D+850 votes (was 700)

AD 21 (Armstrong): D+900 votes (was 700)

AD 29 (Silberkraus): D+900 votes (was 800)

AD 35 (Jones): D+1,000 votes (was 800)

 

UPDATED, 10/26/16, 6:30 AM

Latest statewide numbers, with not all rurals in, show a 24,585 raw vote lead for the Democrats.

UPDATED, 10/26/16, 5:30 AM

Clark numbers from Day 4 show the Democrats adding 5,000 to the party firewall in the Democratic county. The numbers: 15,831-10,971. That's 47-33, or right at registration for both parties.

The totals (I don't have Tuesday's mail yet): 28,500 lead for the Democrats in Clark. That's 50 percent to 31 percent, or 5 points above the registration difference.

In 2012, after four days, it was 25,000, and the difference was 4 points above registration.

So a better raw vote total, although there are 150,000 more voters this cycle registered to vote -- 850,000 vs. 1 million. So the percentage is just a tick under 2012. Then again, raw votes are raw votes.

Looking a lot like 2012.

Indeed, Day 4 in Washoe was very similar to Day 4 in 2012, when the GOP had a 200-voted edge, and Day 4 in Clark was almost the same, too.

After four days in 2012, 120,718 voters had early voted in Clark -- that was about 14 percent of total registration. In 2016, that number is 138,953, or 14 percent.

The big difference so far is Washoe: The Democratic firewall was 1,200 at this time in 2012; it is close to 3,000 now.

Remember, too, that all polls show Donald Trump not doing what Mitt Romney did: Holding the GOP base at the usual 90 percent level. So unless indies are going big for him -- and the evidence shows otherwise -- if these numbers hold, Hillary Clinton is going to have an insurmountable lead coming out of Clark County.

Yes, only four days are in. But that's a close to a third of the data. Also, Clark and Washoe represent about 85 percent of the state's total vote, perhaps more.

It's highly unusual for trends to reverse themselves. Slow down, perhaps. But not turn the other way, which is what the GOP needs to happen to avoid a Clark wipeout.

UPDATED, 10/25/16, 8 PM

Washoe numbers for Day 4: Republicans (4,331) and Democrats (4,326) essentially tie. Let's see if it's a blip or a trend, and don't forget Democrats already have built a 2,800-ballot lead in the swing county (slightly GOP).

One more Washoe note: I have the three-day look at the race for Greg Brower's state Senate seat, which has a 38-36 GOP edge. The numbers look like the two southern districts in play: 47-36, Democrats. 1,100 raw vote lead.

The GOP could lose all three of those seats if this keeps up. (Insert "it's early" caveats.)

 

UPDATED, 10/25/16, 12 PM

The deeper dive shows the wave may be building top to bottom after three days, although I need a few more days to be sure.

The overall statewide lead for Democrats is 13 points, 47-34, with just under 12 percent turnout so far. The numbers look very similar to 2012, when Barack Obama won the state by nearly 7 points and Dean Heller won the U.S. Senate race by a point. The Legislature didn't change much in 2012, but Steven Horsford won the new CD.

But 2014 changed the map, and this is about whether the Democrats can change it back by taking back the Legislature and winning those two CDs, the Senate race and the presidential here. 

Turnout in the two contested GOP congressional districts is higher than the statewide 11.7 percent turnout so far -- 12.9 percent in CD3 and 12.1 percent in the Clark portion of CD4.

Unless there is a lot of crossover voting going on and/or indies are voting for the GOP candidates in great numbers, both of those CDs are going blue, and both houses of the Legislature could be, too. Dems needs to win both of the ones in the South or take the one in the North, where turnout also is favoring the Democrats, to take control.

The numbers:

CD3: Registration is 37-35, Democratic.  Dem lead is 46-34, 1 point improvement for GOP. 5,833 raw-vote lead for Democrats. Up from 5,000 after first two days. 51,967 votes cast. Still looking like 0 for Tark.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. Dem lead is 51-30, so GOP picked up a couple of points Monday, but Dems still 6 points above registration advantage. 8,237 raw vote lead, up from 6,500 after first two days. 38,949 votes cast. Not looking good for a second Cresent Hardy term.

SD5 (Woodhouse): Registration is 39-34, Democrats. Dem lead is 48-33, so GOP picks up 2 points overnight but still 10 points above registration difference. 1,509 raw vote lead, up from 1,200 after first two days. 9,998 total votes cast. Joyce Woodhouse looking strong for re-election.

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. Dem lead is 49-33, same as after Sunday and still 9 points above registration difference. 1,797 raw vote lead, up from 1,400 after first two days. 11,054 total votes cast. Need a sea change to keep this GOP. High turnout, too: 15 percent.

Brief glimpses at four key Assembly districts:

AD 9 (Gardner): D+700 votes (was 600)

AD 21 (Armstrong): D+700 votes (was 600)

AD 29 (Silberkraus): D+800 votes (was 600)

 

AD 35 (Jones): D+800 votes (was 650)

 

 

UPDATED, 10/25/16, 6:30 AM

The secretary of state's office numbers show Democrats building on their lead. It's now just under 22,000 statewide, or 47 percent to 34 percent. That 13 points is 7 points above the actual registration difference,

The urban counties again went big for the Democrats.

Clark had another big turnout day -- 34,000 voters -- and Democrats have almost a 24,000 raw vote lead, including mail ballots. That's 50 percent to 31 percent, or 5 percent above what the registration edge is, a slight improvement for Republicans over the first two days. This is very close to what the edge was in 2012 after three days (52-32) and the 5,000-voter edge on day 3 is nearly identical to Day 3 in 2012.

2012 comparison on Day 3, with percentage turnouts relative to registration, with Democrats adding 5,000 to firewall:

Total 34,603 (3.40%) // 2012 total 30,589 (3.59%)
Dem 16,262 (3.71%) // 2012 Dem 16,853 (3.84%)
Rep 11,437 (3.86%) // 2012 Rep 11,540 (3.89%)

In Washoe, Democrats had an 800-voter edge on Monday, bringing the total with mail (where the GOP has a 600-ballot lead, to about 3,000 votes. Republicans have a 4,000-vote registration edge in Washoe.

The rurals have turned out 18,000 voters, and Republicans have a 4,600-vote lead. 

Total turnout so far is 170,825 voters, or about 12 percent. Democrats are at 14 percent of registration and Republicans are at 12 percent of theirs. 

UPDATED, 10/24/16, 1:45 PM

Detailed numbers from Clark after two days – yes, it’s early, but warning beacons for the GOP abound -- show possibility of deep blue wave:

Totals after all ballots tallied in Clark:

EV+AB: 85,662

Dems: 44,227

GOP: 25,440

Rest: 15,985

That is 51.6 percent to 29.6 percent. 18,787 raw votes difference between the parties

Dems+8 over registration, GOP at registration

Not much different than 2012, slightly better.

 

CD3: Registration is 37-35, Democratic.  Dem lead is 47-33. 5,000 raw-vote lead for Democrats. Up from 3,000 after first day. 36,478 votes cast. If this continues, it’s still going to be 0-for-Tark.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. Dem lead is 54-29. 6,500 raw vote lead, up from 4,000 after first day. 27,757 votes cast. Who did you have in mind for your chief of staff, Rep. Kihuen?

SD5 (Woodhouse): Registration is 39-34, Democrats. Dem lead is 49-32. 1,200 raw vote lead, up from 800 after first day. 7,227 total votes cast. Hard to Buck this trend.

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. Dem lead is 49-33. 1,400 raw vote lead, up from 1,000 after first day. 8,227 total votes cast. Need a sea change to keep this GOP.

Brief glimpses at four key Assembly districts:

AD 9 (Gardner): D+600 votes (was 300)

AD 21 (Armstrong): D+600 votes (was 400)

AD 29 (Silberkraus): D+600 votes (was 400)

 

AD 35 (Jones): D+650 votes (was 400)

Real question in Assembly if this continues is if this is a mirror image of 2014, when races that never should have been in play come into play.

 

UPDATED, 10/24/16, 9 AM

Early vote only
88,331 statewide
46,062 Dem (52%)
26,240 Rep (29%)

Total votes cast (includes absentees)
111,061 statewide
55,001 Dem (50%)

35,516 Rep (32%)

 

UPDATED, 10/24/16, 7:30 AM

Bottom line: After two days, the raw vote lead in urban Nevada for the Democrats is close to 21,000. (Most rural counties closed on Sundays, and those numbers, while they favor Republicans, are tiny in comparison.)

Clark County numbers for Sunday are again disastrous for Republicans, with another record day for turnout:

 

Total: 31,000
D: 16,000
R: 9,119

That's 52 percent to 29 percent. Dems are 9 points above registration, Rs are at registration. That's also slightly above the 2012 second day, which was 50-33.

Two day Clark totals, with mail ballots counted so far (no mail on Sunday): 

Total: 84,468

D: 43,672 (52 percent)

R: 25,022 (30 percent)

Actual registration: 43-29, so Democrats are 9 points above their registration and Republicans are 1 above theirs.

That's an 18,000-plus raw vote lead. It was 13,000 in 2012. But there are 150,000 more voters this cycle, so it's slightly better proportionally.

It's a very blue wave so far, and Republicans have to be worried, especially with those Washoe numbers below and how deep the first day went down the ballot.

More details later....

 

 

UPDATED, 10/23/16, 6:30 PM

Turns out the secretary of state's office made a mistake in tallying Washoe absentees (I thought they were too high), which reduces the overall 13,500-vote Democratic lead for day one to about 12,000. Here are the revised results. That's still 49 percent to 33 percent.

Washoe numbers for Day 2 show Dems had an 800-vote lead, so total lead up there is 2,200 votes.

UPDATED, 10/23/16, 1 PM

I now have the district totals for the first day of early voting and the down-ballot numbers also are awful for the GOP. It is early but:

CD3: Registration is 37-35, Democratic. First day: 3,000 raw-vote lead for Democrats. 47 percent, Democratic; 34 percent, Republican. So Democrats were 10 percent above registration and Republicans 1 point below. Bad omen for Danny Tarkanian and good for Jacky Rosen? We’ll see.

CD4: Registration is 44-29, Democratic, in Clark part of district, which is four-fifths plus of the vote. First day: 4,000 raw-vote lead. 53 percent, Democratic; 29 percent, Republican. So Democrats were 9 percent above registration and Republicans right at theirs. Not like 2014 for Rep. Cresent Hardy, and the first good news for state Sen. Ruben Kihuen in a long time.

SD5 (Woodhouse): Registration is 39-34, Democrats. First day: Nearly 800 raw-vote lead; 50 percent, Democrats; 32 percent, Republicans. What a first-day disaster for the GOP. Dems plus 11, GOP minus 2. 13 points. The blue wave ran deep on the first day.

SD6 (Seaman-Cannizzaro): Registration is 41-34, Democrats. First day: Nearly 1,000 raw-vote lead. 51 percent, Democratic; 32 percent, Republican. Another potential disaster in the making: D plus 10, R minus 2.

Brief glimpses at four key Assembly districts, all of which had terrible first days for the GOP:

AD 9 (Gardner): D+300 votes

AD 21 (Armstrong): D+400 votes

AD 29 (Silberkraus): D+400 votes

 

AD 35 (Jones): D+400 votes

UPDATED, DAY ONE NUMBERS, 10/23/16:

Democrats destroyed Republicans in the first day of early voting in Nevada, amassing a 13,500 raw vote lead and dominating in Clark County. (SEE UPDATE ABOVE FOR ERROR BY SOS THAT REDUCED IT TO 12,000)

The raw numbers, according to the secretary of state, combining early and mail ballots: 41,017-27,606. That’s 49 percent to 33 percent. Democrats are 10 percent above their registration numbers and Republicans right at theirs.

The first numbers in the populous South are very similar to the first day of 2012 while the tallies in Washoe County are significantly larger than they were for the Democrats four years ago.

How much of this is people wanting to end their participation in a relentlessly depressing election and how much is the result of the Democratic machine resurfacing after a 2014 hiatus will become clear after a few more days. But this is an ominous beginning for Republicans, who show little enthusiasm so far, if it mirrors 2012, when President Obama won the state by 6 points and Democrats achieved big gains except for the U.S. Senate race that Dean Heller won by 12,000 votes over Rep. Shelley Berkley. The latter result will still give Republicans hope in the race to replace Harry Reid, especially because Hillary Clinton is no Barack Obama. But this is the kind of break out of the starting gates the Democrats wanted.

There may be less straight partisan voting than in other presidential years, but it’s hard to see how that favors Republicans in a presidential race where Donald Trump is not nearly so strong with the GOP base as Clinton is with the Democratic base. The key is where non major-party voters alight – 9,000 turned out on the first day.

The full statewide totals so far (a few rurals included) are here.

But here are some numbers to consider:

Clark

Record turnout of 39,000, 6,000 above 2012. But with 150,000 additional voters, the percentage (3.8 percent) is about the same.

The Democrats won 21,611-10,701. That’s 55 percent to 27 percent, almost identical to what it was in 2012. That means Democrats are turning out 12 points above their registration and Republicans are 2 points below theirs.

If you add in absentee ballots returned so far, the total is 14,137 returned so far, and the Democrats lead there, too: 6,061-5,202.

So the Clark totals: 27,672-15,903. That’s 52 percent to 30 percent.

Frame of reference: In 2012, the Democrats began this way and ended up with a 70,000 raw-vote lead by the end of the two-week period.

Washoe

It was 51-33 up North, and an 1,800 raw vote lead in early voting: 4,809-3,078. The percentage is almost twice was it was on the first day in 2012. The total turnout – about 9,400 voters – is almost identical to 2012.

Washoe had a huge number of absentees returned – usually a GOP strength: 15,467, even more than Clark. Those totals: 6,992, Democrats; 5,992, Republicans. (SEE ERROR ABOVE BY SOS THAT SHOWS GOP PLUS 500 IN ABSENTEES, NOT DEMS PLUS 1,000)

So the totals for Washoe: 11,801-8,8854. That’s 47 percent to 36 percent.

Registration in Washoe is 37-36, GOP. So the Democrats are 11 points above their registration, and the Republicans are 1 percent below theirs.

Most remarkable stat about Washoe: About 10 percent of people already have voted!

By the end of 2012, early/mail voting in Washoe was about even.

These early northern numbers should worry Republicans. A lot.

More numbers and context when I get them.

----

Democrats lead Republicans by 500 absentee ballots in early numbers from Clark County on the eve of early voting, KTNV has learned.

Sources report that the totals as of noon are 4,782 Democrats and 4,236 Republicans as well as 2,406 others. That’s 42 percent to 37 percent.

It’s a small sample, of course, but the final Clark absentee numbers in 2012 showed: 42 percent Democrats, 40 percent Republicans.

I am reliably told that GOP absentee ballot requests are up from 2012 across the state, but the only ones that count are the ones that are returned, of course. (And the only poll that counts….)

Republicans have about a 9 percent lead statewide in mail ballots returned, about the same as 2012. The Democrats here don’t put much effort into absentees, which made up 7 percent of the total ballots in 2012, and focus much more on driving early voting turnout.

That’s why watching what happens Saturday, the first day of early voting, will be so interesting. In 2012, the Democrats far outpaced their 15 percentage point registration edge (it’s 14 percent this year) on the first day.

The partisan absentee ballots returned are almost exactly even in Congressional District 3, the state’s swing district, and they are close in those two contested state Senate districts, too.

So what do we know on the day before early voting starts? Not much.

But this is the blog to watch as I will update it every day with totals and analysis of numbers up and down the ballot.