VA-GOV 2017: It's a Wave (GE: Nov 7th - Thread #2)

[The Other Castro]

Monmouth will release their final poll tomorrow. Previous poll was G+1. Some tea leaves:

Patrick Murray‏ @PollsterPatrick
Wapo & Roanoke show tightening.  Qpac & Suffolk show widening. "Trend" driven by Rass & TPC being late game entries in the average.

Nate Silver @NateSilver538
Replying to @PollsterPatrick
So you guys have Northam rebounding? 😬 The polls showing tightening are more recent than the ones showing widening.

Patrick Murray‏ @PollsterPatrick
Replying to @NateSilver538
We're still in field, so not clear yet. But bottom line, is it's still adhering to our overall trend that race has always been tight.

https://twitter.com/PollsterPatrick/status/926851496578297857

So, based on this I would guess that Monmouth will still show a close race, though maybe anything from G+2 to N+2.

[The Other Castro]

The tweets from the Monmouth pollster Patrick Murray also made me think. What would we see if we looked at only the changes in margins of recent pollsters? Excluding partisan polls (assuming I hopefully did all these numbers correctly), the most recent pollsters with relatively recent previous polls to display a change were: Emerson, Roanoke, Suffolk, WashPost, Quinnipiac, CNU, and Fox. Changes were:

Emerson: Gillespie +2 (Northam +5 ---> Northam +3)
Roanoke: Gillespie +6 (Northam +6 ---> Tied)
Suffolk: Northam +4 (Tied ---> Northam +4)
WashPost: Gillespie +7 (Northam +13 ---> Northam +5)
Quinnipiac: Northam +3 (Northam +14 ---> Northam +17)
CNU: Northam +3 (Northam +4 ---> Northam +7)
Fox: Northam +3 (Northam +3 ---> Northam +7)

So, what do we see when we take out the newcomer pollsters to this race like Rasmussen, Trafalgar, and The Polling Company? An average of (G+2 / 7 polls) = G+0.29 average trend, or what we would consider in poll trends as virtually negligible. Now, is this a fair way of calculating poll trends? Not exactly, but it does help show what the changes are from poll to poll when you ignore new entrants that potentially muddy the average and cast false trends. Perhaps Gillespie has been tightening the race, or perhaps it has been mostly static.

[The Other Castro]

Worth remembering tomorrow night:

You must be logged in to read this quote.

And in 2014, Gillespie lost his lead when 92% of the precincts were reporting.

So McAuliffe 2013 took the lead with 76% reporting, Clinton 2016 took the lead with 83% reporting (with a 5.4% final margin!), Warner 2014 took the lead at 92% reporting, and Herring 2013 took the lead in a recount.

Anyone know when Obama 2012 (3.9% final margin) took the lead?  Probably earlier than Clinton and McAuliffe because less purely dependent on NOVA?

I couldn't find the exact percentage in reviewing old election overage, but Obama trailed by ~16K with 80% in, and led by ~15K with 84% in, so I'd guess he took the lead with about 82% in.

[The Other Castro]

So it’s looking like if it’s a 2-3 point win for Northam, this could be called between 9:30-10ish. Closer than that, and this could take a while.

Edit: Though even in VA 2014, Warner took the lead around 10:40, and once the Democrat takes the lead at that point it’s pretty much over, so at least we’ll likely know if Northam wins before 11 (if he wins by more than 1%).

[The Other Castro]

Fairfax too:

Vote Fairfax City @VoteFairfaxCity
As of 8 a.m. 1,365 voters have marked their ballots in today's election. With absentee, that's more than 15.2% #Election2017

https://twitter.com/votefairfaxcity/status/927885831221719040

I sense a trend...

[The Other Castro]

Jonathan Martin‏ @jmartNYT  6m6 minutes ago
Precinct check-in!

Rosslyn in Arlington Co (#19)

‘13:1326 votes
As of 3:15 today: 1132 votes

Steady line of voters in rain, mostly women

https://twitter.com/jmartNYT/status/927995226865766406

[The Other Castro]

Yeah I don't think I'm going to rely on the guy on Twitter that went from wow Northam is going to win by the biggest margin since 1985 to he's relying on just absentees now over the course of a few hours.

[The Other Castro]

Breaking: early exits right here

https://twitter.com/sahilkapur/status/928023885299281920

[The Other Castro]

Steve Kornacki‏ @SteveKornacki  18s18 seconds ago
Virginia exit (first wave) -- composition of electorate
White college degree 43%
White non-college 26%

'16 exit:
White college: 38%
White non-college 29%

https://twitter.com/SteveKornacki/status/928025635787853825

[The Other Castro]

Fairfax County Votes‏ @fairfaxvotes  4m4 minutes ago
5 p.m. estimated voter turnout is 39.1% + estimated absentee at 7.7% = 46.8%. About 90 minutes to get to poll by 7 p.m. Drive safely please!

https://twitter.com/fairfaxvotes/status/928025871079833600

[The Other Castro]

Reminder, Fox's experimental exit poll comes out at 6.

[The Other Castro]

Honestly the most encouraging news I've seen in the past few hours is Fairfax County hitting its 2013 levels.

[The Other Castro]

Not sure if this was posted earlier, but in the first wave of VA exits:

"Democrats accounted for 39% of voters in Virginia, vs. 31% Republicans and 29% independents, per preliminary exit polls."

https://twitter.com/evanmcmurry/status/928027732000296961

[The Other Castro]

So this is indeed looking like 2013 redux?

Yeah, a 2-3% Northam win is probably the most likely outcome.

I'm kind of thinking 5-6 still.

[The Other Castro]

Melissa J. Hipolit‏ @MelissaCBS6  37m37 minutes ago
Richmond Registrar says turnout so far is at a "respectable" 37% @CBS6

https://twitter.com/MelissaCBS6/status/928023056789983232

[The Other Castro]

In the hypothetical scenario that Gillespie wins, would this be a bigger upset than 2016?

No.

[The Other Castro]

I calculated based on the exit poll, Northam leads 51.86% to Gillespie's 46.63%.

[The Other Castro]

So far so good for Northam.

[The Other Castro]

This is looking really good for Northam. I'm liking what I'm seeing.

[The Other Castro]

Calling it. Northam wins.

[The Other Castro]

These House of Delegates races are insane!

[The Other Castro]

The wave is here folks. Get your surf boards ready.

[The Other Castro]

Why the f would you coat diamond with titanium? Friedrich Mohs is rolling in his grave.