Talk Elections

First of all, I highly recommend reading this piece by Nate Cohn on why the polls may be off by more in Georgia than in other states, due to the changes in the population, the way many pollsters peg estimates to the Census, and how VR can impact this as well:

Why Georgia May Be Bluer Than It Appears (http://www.nytimes.com/2014/10/09/upshot/why-georgia-may-be-bluer-than-it-appears.html?smid=tw-share&abt=0002&abg=1&_r=0)

---

While I did an exposé using the VAN a while back - which suggested that VR rates were actually below 2010 levels - it turns out that the system simply hadn't been updated. The reality couldn't contrast more.

Georgia has added in 2014 alone 212,000 new registered voters - with estimates suggesting 40,000 - 60,000 more still waiting to be processed (some being submitted months ago and still not showing up in the system). The broader trends in the state have been strong enough to have shrunk the white share of RVs from 63% in 2010 to 58% as of October 7th.

AJC analysis: Georgia sees surge in voter rolls (http://www.myajc.com/news/news/state-regional-govt-politics/ajc-analysis-georgia-sees-surge-in-voter-rolls/nhdgL/?ecmp=ajc_social_twitter_2014_sfp#f70d2ad0.3828699.735513):


How does this stack up to historical trends? There is simply no comparison when looking at changes in voter registration between a presidential year and the following mid-term elections. Let's look at the first graph I've made, which shows the number of active registered voters in the state immediately before each election (1998-2014):


While not the single biggest increase in registered voters between any two election cycles, it is still the biggest increase in registered voters between a presidential and a mid-term in the 50 years of recorded stats on file with the SoS - by a metric ton. Speaking of which, let's take a look at changes over those 50 years in the voter rolls:


There have only been four instances in the past 50 years in which the number of RVs increases in a mid-term cycle when compared to two years prior (1966, 1998, 2006 and 2014). In every other case, voter rolls contracted in Georgia.

And finally, the most visually-stunning graph that drives the point home. How many net active registered voters in total have been added over this two-year period when compared to every other cycle?


It isn't even close. Georgia has literally added twice as many voters during this mid-term cycle as any other mid-term cycle in modern history (310,000). The Dem wave is building!

Wonderful news!

THE DREAM IS ALIVE

Surf's up!

Indeed, my friend.  The wave of progress cannot by stopped. ;D

Seriously though, this is fantastic news for democracy in general, and bodes very well for Democrats in the future, regardless of the elections this year.

I just hope we'll see a similar surge in Texas and Arizona in 2016. :)
I guess Florida could seriously need one as well, right?

I think Arizona should be the main focus next.

http://politics.blog.ajc.com/2014/10/21/only-one-third-of-183416-new-voters-describe-themselves-as-white/

How come Hispanics aren't registering?

I'd venture that Hispanics are most likely to decline to state their race

Excellent! The people are speaking.

Off topic, but how come you left out Franken and Durbin from your sig?

Primarily because they are not in trouble. Durbin isn't even campaigning (except for other candidates) and Franken's situation is similar. Dayton is only at Lean D now and the rest are toss-ups.

Ah, I thought so. Aren't Franken and Dayton set to win by pretty similar margins though?

Yes, but polls vastly overestimated Dayton's margin in 2010.

Most sites have Franken at Likely D while some have Dayton at Lean D.

GET A LOAD OF THIS:

These numbers are looking insanely good for Democrats, and defy almost every poll in terms of the level of black and/or Democratic turnout. Between my last update yesterday and this one today (which covers early voting through Tuesday), the black share of the electorate jumped by one-half of a percentage point (up until yesterday, it had been increasing around 1/3 of a point per day). Even better is that trends during early voting will likely continue to increase in favor of Democrats all the way through the final day of early voting. Not to mention that my calculations suggest that we could have 750,000-800,000 early voters when all is said and done (was 678,000 in 2010).

And to put into perspective what we'd expect the "drop-off" to be in terms of racial composition of the electorate in early voting versus all voters, it dropped from 29% in 2010 to 28% for the entire electorate. This would suggest that we very well may have a black electorate that is a larger percentage of the vote than it was in 2008 or 2012. This is also the first day in early voting (and I'm sure in at least three or four election cycles) where the number of dominant-Democratic primary voters is outnumbering the number of dominant-Republican primary voters.

Dominating!

Nunn will probably still need more than the nominal 23% of the white vote that Kerry '04/Obama '08 got to win the election day vote, and she'll definitely need more than that to avoid a runoff. But, the more black voters GA have, the lower the 'Nunn white vote threshold' gets.

If Perdue wins with a runoff, or loses, republicans really need a 'Georgia autopsy', because they will have gotten the same results or worse as they did in the dem wave year of 2008.

They both need 30% of the white vote to be in a good position to win a majority, assuming the black vote goes 90% D. Most polling puts both of them within that range when considering MoE, and black support always starts out underpolled and rises through the cycle - and usually seems to end up higher than it is pegged in polling (right now, it's bouncing between 80-85% D). I calculated the EV totals earlier and I was getting 51% D with those figures (and these numbers should continue to improve throughout the rest of the week).

Voter file just updated with Tuesday's totals:

607,569 voters
Black share steady (32.0%)
White share steady (62.3%)

Democrats slightly down (40.0%)
Republicans steady (39.2%)
Unknowns rising strongly (14.5%)

Females up (54.9%)

Large increase of 31-50 year-olds (18.8%)