Because We Have a (Serious) Problem: 2016 Battleground States

[emzie]

Interesting stuff. Shows what I get for not Googling. I suppose it could feasibly become a swing state if things get really close. Schweitzer could probably get it if he winds up on the ticket.

Obama lost Montana by like 3% though in 2008. I think its current demographics would make it like a hightide state for Dems. In a wave election, it'd be the last state to flip to Dems. But at that point, they're probably pushing 400 EVs so it's moot.

Honestly, where I "live" in the US is dependant on how things go this June. If I can afford to miss 2 weeks of work and drive out and visit my mom, I'm going to change my residence on the trip.

 

[Gucumatz]

I can see Texas shifting to a 51-49 state this next election, considering there are active campaigns starting to flip the state. I just don't think there is enough time yet to do so, I say the next presidential election after 2016 [2020] that it would become a real battleground state.

 

[mangxema]

What about the effect of all the changes to voting rules the GOP is making/trying to make?

Hey I wish I lived in a battleground state.

I thought that too once. Then I moved to Florida. Dear God make it stop.

I can see Texas shifting to a 51-49 state this next election, considering there are active campaigns starting to flip the state. I just don't think there is enough time yet to do so, I say the next presidential election after 2016 [2020] that it would become a real battleground state.

There was this PPP poll that suggests Clinton could take Texas:

50% of Texas voters have a favorable opinion of Clinton to 43% with a negative one. She's universally well liked by Democrats (91/5) and a majority of independents view her positively as well (52/41). She holds narrow leads in hypothetical match ups with Marco Rubio (46/45) and Chris Christie (45/43) and a wider one in a contest against Rick Perry (50/42).

That would certainly be interesting.

 

[choxorn]

I would prefer swing state Texas to Republican Texas.

 

[Gucumatz]

What about the effect of all the changes to voting rules the GOP is making/trying to make?



I thought that too once. Then I moved to Florida. Dear God make it stop.



There was this PPP poll that suggests Clinton could take Texas:



That would certainly be interesting.

Polls mean absolutely nothing at the moment, but Clinton is liked. With a decently run campaign I see her easily beating any potential Republican candidate at the minute, regardless if Texas comes into play or not in 2016

 

[Antilogic]

That's true, but I don't think Michigan has been significantly more elastic than Georgia in recent years (don't have the data in front of me at the moment). And at least in Georgia there is a significantly growing urban population (the Atlanta metropolitan area has nearly half the population of the state and growing, although its suburbs are still split between Democrats and Republicans) and a significantly growing Hispanic population that should at least warrant some national coverage (if not equivalent to Michigan). And yet the only thing I can find on it are the occasional blog postings and a handful of clips of MSNBC's Chuck Todd putting forth the hypothesis it could be competitive in 2016 and 2020. As far as I can tell, it hasn't been reported on Fox, CNN is largely ignoring it, and so on.

So, I've been following up on this by calculating the standard deviation of the deltas between the state's vote and the national average in each election since 1988, and made two charts with it (graphic attached). Here's what I found:

So my earlier criticism of the Michigan-Georgia comparison, if you adjust for the national vote, wasn't as well-founded as I thought. Michigan averages around +5.3D, while Georgia somewhere around +10.9R. Michigan is more steady long-term, with a SD of 2.39 against Georgia's 3.22, but that's nowhere near enough to make up for the difference in averages.

Montana is still plain weird and defies analysis. It averages +14.2R yet has 3 major statewide offices held by Democrats, including recent wins that everybody incl. Nate Silver thought would go the other way (the last Senate race).

From this analysis, you might think that Democratic hopes to put Texas in the swing category by 2020 are crazy, but remember the favorite son effect for the Bushes. Of the three Southern/Southwestern states we talked about potentially becoming tossups, Arizona is the best bet. Well, that and North Carolina, but we are already considering that a tossup state.

Virginia is the most steady pro-Democratic trend in this analysis. Unfailingly since 1988, it's gotten more Democratic. The Republican inroads in the Midwest haven't paid off yet, but by this analysis it looks like Colorado and Pennsylvania are about equal at +1.5D and if trends hold Colorado will be a solid-D state. Pennsylvania, though close to the tipping point, is one of the least variable states on the list. At this rate, it might take the Republicans until 2024 or later to make it a true swing/tossup state. Wisconsin might be a better strategy just based on the variability.

What about the effect of all the changes to voting rules the GOP is making/trying to make?

Some analysts (I think they were on CNN when I saw them, I don't have an article reference at the moment) suggested that these efforts make maybe 1-2% difference at most. The explanation given was that any real effort at reducing voting hours by one party is going to incite a group to vote at no matter the cost against the guys who are doing it, so the effect is mitigated.

Myself, I'd like to see some verification of the theory too. I'd like to think there is strong popular support for insuring everyone can cast their ballot, but you'd never know.

I thought that too once. Then I moved to Florida. Dear God make it stop.

I'll ask my 8-ball then make the wrong decision because it will lie to me.

I would prefer swing state Texas to Republican Texas.

I would too, but not because I'm a hyper-partisan who wants the Republicans to fail at every opportunity. Historically, the most populous states tended to be swing states and so campaigns would focus on the larger states. Right now, of the 5 most populous US states, only Florida is seriously contested. California, New York, Texas, and Illinois are all (mostly) safe for one party. If you look at the next 5, Pennsylvania, Georgia, and Michigan are mostly safe, while Ohio and North Carolina have at least voted for different candidates nationally in the last few elections.

Long story short, of the top 10, only 3 have voted for different parties in the last few elections. It's relatively static.

(Side note: I found and fixed one error with the spreadsheet before uploading this pic, I hope there aren't any more. If you guys see something odd, point it out and I'll double-check the calculation.)





EDIT: I didn't refresh the page while writing this and I didn't see metatron's post until now. I'll try to respond to some of the points made there later tonight, have some work to take care of in the meantime.

 

[Antilogic]

So I think my follow-up post already addressed some of the issues that metatron raised. This will be a short follow-up, and then hopefully a discussion will start again rather than a relentless anti-monologue.

Several problems:

1) This leaves a Democratic column worth 253 votes and a Republican worth 191.

253. 191.

No argument from me. I think the electoral college, unlike in the 1980s, favors the Democrats.

2) North Carolina has virtually no chance of justifying its inclusion in this list. It may - potentially - swing. Yes. But so does Nevada and most of the states on your second list.
There is no reasonable chance that North Carolina would wind up being the tipping point state.
Just as a reminder, the necessary conditions would be:
a) Democratic win in NC and nationally, with all of the above states (with the exception of either CO or IA and NH) going the other way (including Florida and Virginia, most oddly).
b) A Republican win in NC and nationally, with most of the states on your list (for a total of 79 out of those 92 votes) being won by the Republican candidate at a more comfortable margin.
This is incredibly unlikely.

If being a "swing state" is supposed to imply any significant (non-indiannesque) relevance regarding the outcome of the election, North Carolina is very much not a swing state.

I was using the term swing state in that post in the media-friendly term as a state that might potentially vote for either candidate. Or, alternately, states that might not be called until a couple hours after the polls close. As far as tipping points go, which I think is what you are focusing on with your 2nd point, I agree completely; Colorado was the tipping point in the last election, and if it continues its leftward swing than it might change to Wisconsin or Pennsylvania. If the Republicans play their cards right, they might have a serious chance at winning them in a close race.

I should go back and incorporate Nate Silver's lingo dividing between swings and tossups.

3) The treatment you gave the rust belt is somewhat optimistic. If i wasn't this lazy i'd dig up some Republicans saying Virginia was out of reach in 2006 (and you'd still have 2 additional years worth of blissful oblivion).
Yes, Georgia roughly as "close". But a) it's not an equal opposite, as others have pointed out and b) the mother of horrifically botched campaigns is not the standard.
Heck, the guy essentially campaigned on "Nah, screw the rust belt, what do i care?!"
Democrats lost one House seat on the entire west coast in 2010 (two if you insist to include Nevada). Rural Oregon? Nah, they all stuck.
They've lost a combined 19 in Ohio, Michigan, Wisconsin, Pennsylvania and upstate New York. You know all this, of course.

If one thing got stuck with voters of all stripes and creeds in this last election it was that the President saved Detroit (for better or worse) and that his opponent has a car elevator and would not look believable with a wrench in his hand (for better or worse).
And you are fine with concluding that the one election (largely void of clearly defined issues) is an anomaly yet the other (ridiculously lopsided on several levels) is a standard upon which one can examine these states?

PPS: I'm sorry. I still can't get past this:

You are still sounding like an 06 Republican believing in the safety of Virginia.
Kerry won Wisconsin by 11,000 votes. Gore by 5,700.
They've won Minnesota by 2.4 and 3.4 and Michigan by 5.1 and 3.4.
They lost Georgia by 12 and 17.

It's true that it might sound the same, but if the data warrants the conclusion than I have to make it. There is nothing in the data at the moment (emphasis here) to warrant Republican optimism in the Rust Belt besides in Wisconsin, where it was neutral to slight Republican in the 1990s. I need to research to see if that effect was caused by Ross Perot or some native state trends. Compare that to the solid trend in Virginia. I don't think we want to fall into the false-equatism trap either.

I'm also betting that local state issues will favor the Democrats (emergency manager laws, union-busting, etc.). It's an assumption, and I'll be up-front about making it. If the Republicans can overcome that, then my analysis would need to be re-examined.

PPPS: I hope this didn't sound too harsh. Just reminding you not to tempt the wrath of the whatever from high atop the thing. ...

I doubt it will be relevant to the outcome of the election.
But i agree with you in so far that it's probably the one place where i'm more optimistic than Antilogic, particularly if Clinton is the nominee.

If I'm risking divine retribution and I'm not even part of their party, I think the Dems should be yelling at me to shut up right about now.

I should be clear with regards to AZ, TX, and GA--I don't think they will be serious swings in 2016. I think they will be talked about as potential swings, but ultimately stay Republican after the conventions and debates. My posts include some guesswork out to 2024, and those honorable mentions are there for watching trends to see if the guesswork is right.

PS: Note that #2 and #3 are somewhat contradictory (if i'm right about #3, things could shift around as far as giving the inclusion of NC on the swing state list some merit). I'm aware of that.

I put this last because it's an important point--we are all projecting out based on current trends, and obviously things can change going forward.

 

[Antilogic]

Minor necro here, but I think there is a good reason:

Nate Silver and the NY Times crew have put together a cool little applet and an article describing it that enables you to play around with the demographics, party breakdown, immigration figures, and parameters on immigration citizenship and percent voting. It will then predict the state-by-state popular vote winners and show you the electoral map (assuming generic Democrats and Republicans are on the ticket, no home state advantages or anything like that) for an election given your assumptions from now until 2048.

I'm still playing around with it, but I figured I'd post it here and solicit some assumptions and the maps they make from the crowd.

 

[downtown]

Geez, that just goes to show had badly this demographic time bomb is. Even if 0% of all illegal immigrants earn citizenship, and even if the Democratic share of the black vote drops below 90% (possible), and even if the democratic white share of the vote drops to around 38% (also possible), assuming typical candidates, Republicans WOULD STILL LOSE.

 

[Antilogic]

Geez, that just goes to show had badly this demographic time bomb is. Even if 0% of all illegal immigrants earn citizenship, and even if the Democratic share of the black vote drops below 90% (possible), and even if the democratic white share of the vote drops to around 38% (also possible), assuming typical candidates, Republicans WOULD STILL LOSE.

Yeah, it's a big problem. If the Democrats retain at least 35-36% or so of the white vote, and do as well as they have been doing amongst minority voters, they are solid. And given the women's vote is favoring the Dems at the moment, it's unlikely they will drop below 35%, I think.

I'm looking at scenarios where immigration and racial divides fade away--say the Republicans get 20% of the black vote, and they do as well as G. W. Bush did with Hispanics, splitting 45% to 55%. Say the Dems pick up 1-3% of the white vote. Florida and Texas won't flip, North Carolina and the South is still competitive, and the blood-letting in the West is stalled.

If you assume the immigration deal fails and there is a continued polarization--0% gain citizenship, the Hispanic vote sways 85%+ to the Dems, the white vote goes 65% to the Republicans, then you got a very interesting trend. 2016 and 2020 are good for Republicans, especially in the Midwest, but then the Dems start making a comeback, flip Texas and Arizona, and then start dominating into the 2040s.

 

[Cheezy the Wiz]

I don't want to live in a battleground state!

Come to Maryland. Republicans at the national level have all but forgotten about us, and the local party squabbles are minor.

 

[Dachs]

Come to Maryland. Republicans at the national level have all but forgotten about us, and the local party squabbles are minor.

On the other hand, you're living in Maryland.

 

[Cheezy the Wiz]

On the other hand, you're living in Maryland.

I know! It really is that good. Love me some Merlin!

 

[madviking]

On the other hand, you're living in Maryland.

I concur with Dachs.

 

[dutchfire]

Building off downtown's thread, let's talk about which states we think will be the swing states, tossup states, and the tipping point in the 2016 presidential election. I'm noting the EVs in parentheses for ease of reference.

Tipping Point

The tipping point in 2008 and 2012 was Colorado (9), but given the demographic trends and strengthening local Democratic party, the tipping point could fall to Virginia (13), the runner-up in 2008, or Pennsylvania (20) or Ohio (18), the runners-up in 2012. If the Republicans hold the line with Hispanics or improve their standing, Colorado will likely hold its position.

The Current Swing States

I think Nevada (6) is likely to fall off the list of serious swing states unless the Republicans make serious progress in bridging the divide with Hispanic voters. Colorado (9) is also in this category--it's worth marginally more to contest since they lost it by slimmer margins and it has 50% more EVs. With limited resources, the Republican strategic choice is clear.

Iowa (6) will always be a swing state because campaigns get their start there--until that changes, it will be contested despite its social-moderate attitude and going Democratic in something like 6 of the last 7 elections--yes, it even went for Dukakis. Very similar case with New Hampshire (4), only it's 5/7 since it didn't vote for Dukakis. A socially moderate Republican has a shot at both of these, though.

Ohio (18) will likely remain a perennial battleground state, as will Florida (29). The parties both have strong local operations. Not much to say here besides duh, obvious.

Virginia (13) looks like it is trending more Democratic, leap-frogging other classic swing states like OH and FL. But it's still close and the Republicans can't afford to lose it; it's definitely in the mix for 2016.

North Carolina (15) will remain a swing state--it's the only serious swing state Obama lost in 2012, and he only lost it by 2ish points. The Democrats will try again, I'm sure.

Michigan (16), Wisconsin (10), Minnesota (10)... why were these considered swing states in the last election when they went for Democrats by more than six points? And Michigan nearly 10? Seriously, Georgia (16) was closer than Michigan and yet nobody called it a swing state.

New Swing States & Outside Possibilities

Pennsylvania (20) has long been the target of Republican last-minute campaign stops, but it's been reliably Democratic except in landslides. However, the Dems are starting to lose ground in the suburbs and the rural areas are voting even more strongly Republican, so if it is contested from the start in 2016 by a moderate Republican, it could be a legitimate swing state. Wisconsin (10) and Minnesota (10) are similar to the Pennsylvanian fools' gold, only with a stronger Democratic margin and thus a steeper hill to climb. I wouldn't count on all three, but PA and WI could be more swingy if the Republicans have a strong candidate and the Dems a weak one.

New Jersey (14) might make it to the swing state list if Chris Christie is the Republican nominee. Otherwise, I think it stays in the Republican camp. And even if it makes it to the swing list, I think it will lean Democratic.

Arizona (11) and Texas (38) are Democratic wet dreams in 2016. Although Arizona has a rapidly increasing Latino population and a growing urban population where the Dems should make gains, they also have a lot of old retirees moving into the state and a strong local Republican party that stave off the trends that flipped their neighbors. So long as Arpaio can win elections, it's staying in the Republican camp. Demographic trends may make Texas a good target for the Dems by 2024, but 2016 is too soon. They will be talked about early in the campaign, but will ultimately stay Republican unless it's a more massive landslide than in 2008.

Indiana (11) was a shocker in 2008, and with the exception of a rape gaffe looked like it was trending strongly Republican in 2012. I wouldn't count on the Senate upset as indicative of a shift, especially since the Republicans should avoid poor-quality candidates for national offices in the future. Likely to remain Republican in 2016 unless there is big landslide (although slightly less massive than for AZ and TX). Missouri (10) is similar.

My Final 2016 List

So here's what I think we will be watching election night 2016, if the race is reasonably close:

The Key States:
Florida (29)
Ohio (18)
North Carolina (15)
Virginia (13)
Colorado (9)
Iowa (6)
New Hampshire (4)

Basically, take off Nevada and the Great Lakes, and the last 3 are a little more iffy.

Additional States for Trendline-Watching:
Texas (38)
Pennsylvania (20)
Georgia (16)
Arizona (11)
Wisconsin (10)
Minnesota (10)

These are my honorable mentions, so to speak, and could also be considered 2020/2024 swing state candidates.

Soooo....
The turning point in 2016 was either Wisconsin or Pennsylvania (pending definitive results. All analysis here is based on the current Wikipedia numbers). Michigan, New Hampshire, Florida and Minnesota were also close to the tipping point.
Nevada and especially Colorado seem to have moved over to the safer side for the Democrats, and given demographic trends are likely to continue in that direction.
Virginia has moved from safely Republican to lean Democrat side since the Bush era, and is likely to continue that way. North Carolina is travelling down a similar path. On the other hand, Iowa went from voting for Obama twice to basically being hopeless for the Democrats (Texas was closer!) . The same holds to a lesser degree for Ohio (Georgia and Arizona were closer). Maine might follow to the lean GOP side, and Maine CD#2 is probably already there. Nebraska CD#2 remains a possible pick-up for Democrats, although it is unlikely to matter.

Given current control of state offices, restrictions on voting rights and on unions are likely in Iowa, Ohio and some of the Rustbelt states. That could move the vote a percent or two to the Republican side, which might be enough to flip the Rustbelt states, and is probably enough to make Iowa and Ohio safely Republican. Conversely, if Democrats can remain strong in Virginia and North Carolina on the state level, this will greatly help their future electoral ambitions.

In the 'Demographics are Destiny' series (y'all don't believe in that anymore, I hope), Arizona [11], North Carolina [15], Georgia [16] and Texas [38] are moving towards the Democrats, and they're pretty big prizes. However, they're still not there, and the question is if the Midwest is moving the other way more quickly.

edit: This thread was started when the abbreviation DT still referred to downtown, not to Donald Trump

 

[onejayhawk]

I am surprised I never posted in this one. That said, it's three years old and had a short live-span before powering down. I miss arguing stats with antilogic.

J

 

[Bootstoots]

Goddamnit. How did Gabbard manage to screw this up? She was supposed to be popular in the Midwest. Granted she only lost MN by 0.4%, but it's not like the recount is going to win it for her.

I'm picking on the Democrat I find most promising right now, because that's how you un-jinx things. The other ones are worse. Mostly much worse, Clintonist-Blairist types. The sooner the Clintonist-Blairists lose power, the better.