Are We Heading Towards The 2012 Map?

[Shameless Lefty Hack]

Well?

My #gut tells me that Nov.'s results are going to be anything but strange, especially given Manafort's steering of the Trump campaign into something resembling GOP orthodoxy. But as I mentioned, it is only July 11.

[Dr. Arch]

Well?

My #gut tells me that Nov.'s results are going to be anything but strange, especially given Manafort's steering of the Trump campaign into something resembling GOP orthodoxy. But as I mentioned, it is only July 11.

Yes and no. I think it'll be 2012+NC.

[For Trump, everything. For immigrants, the law]

The 2012 map is certainly not out of the question. If Clinton wins by 3-5% nationally, it's a distinct possibility.

[Ronnie]

At the risk of sounding like Dick Morris, I'm still holding out hope for a Hillary landslide.  The Pew poll gives me reason to believe it will happen, particularly if Hillary does better among Latinos than the poll suggests.

[Mr. Smith]

2012 with NC...maybe AZ or GA after that.

[Dwarven Dragon]

If we assume that the election will be competitive, which is a reasonable assumption to make based on history and fundamentals, then a Hillary win in a competitive election is either 2012, 2012-OH, or 2012-OH-VA.  Ultimately I think whoever wins the election will win both Florida and Pennsylvania.

[Devout Centrist]

Perhaps, but I think it's going to either be Clinton by a bigger margin than 2012 or Trump by a narrow margin

[MK]

I swear if Trump wins half of atlas posters are going to be on suicide watch.  The more posts like this is just making the gut punch worse if in fact Trump ends up winning. 

[Sir Mohamed]

No, Hillary will expand the map.

[Protect Trans Hoosiers]

No.  As things look today, in an election roughly as close as 2012 was, the following states are at least 40% likely to flip:

Ohio
North Carolina
Colorado
Iowa
Florida

The following are at least 25% likely to flip:

Arizona
Pennsylvania
Virginia

The chance of all of those states' outcomes being identical to 2012 is therefore (0.6^5) * (0.75^3) = 3.28% chance of producing an identical map.

[Nyvin]

No.  As things look today, in an election roughly as close as 2012 was, the following states are at least 40% likely to flip:

Ohio
North Carolina
Colorado
Iowa
Florida

The following are at least 25% likely to flip:

Arizona
Pennsylvania
Virginia

The chance of all of those states' outcomes being identical to 2012 is therefore (0.6^5) * (0.75^3) = 3.28% chance of producing an identical map.

That's not a very reliable way to look at it though...states don't just spontaneously flip on their own, they get caught up in the national tide more often.    An example would be if Hillary is winning OH, she's probably winning PA and IA as well (most likely).  

It's more of a matter of where the national tide starts and ends, in 2012 it ended at North Carolina, and Romney won the state by about 2 points.

Using your method, you could have a map where Arizona, Pennsylvania, Ohio, and North Carolina go to Hillary, while the rest go to Trump.   We know that's probably not likely.

[Mallow]

The map will probably be very close to 2012, but there are so many ways for things to be slightly different that it's highly unlikely it will be exactly the same. Obvious considerations are OH, NC, and AZ.

[Protect Trans Hoosiers]

No.  As things look today, in an election roughly as close as 2012 was, the following states are at least 40% likely to flip:

Ohio
North Carolina
Colorado
Iowa
Florida

The following are at least 25% likely to flip:

Arizona
Pennsylvania
Virginia

The chance of all of those states' outcomes being identical to 2012 is therefore (0.6^5) * (0.75^3) = 3.28% chance of producing an identical map.

That's not a very reliable way to look at it though...states don't just spontaneously flip on their own, they get caught up in the national tide more often.    An example would be if Hillary is winning OH, she's probably winning PA and IA as well (most likely).  

It's more of a matter of where the national tide starts and ends, in 2012 it ended at North Carolina, and Romney won the state by about 2 points.

Using your method, you could have a map where Arizona, Pennsylvania, Ohio, and North Carolina go to Hillary, while the rest go to Trump.   We know that's probably not likely.

The states don't move fully independently, but Colorado, Ohio, Florida, and North Carolina are likely to move nearly independently of one another.  Also, if you have five states where the polls are within MoE, those five states are going to be independent events - flip a coin for each of them.  Even if they are "closely related" like PA and OH, there could be weather events that affect turnout in one state and not the other.

States do move in concert, but that will be reflected in the polls.

If we have a race roughly as close as 2012, expect North Carolina, Florida, Ohio, and Colorado to be independent events which right now I would consider tossups.

Current Clinton +3.86% battleground:

Clinton: 270 Trump: 191

That alone gives only a 1/16 chance of duplicating the result in 2016.  There could be movement between now and then which could, for instance, place Ohio firmly in Trump's pocket, open up Pennsylvania to him, but also put Colorado and Florida in Clinton's pocket and open up Arizona to her.

Hypothetical October Clinton +3.86% battleground:

Clinton: 288 Trump: 198

Same PV margin, but now there's even less of a chance of matching 2012.  Because there's that chance for additional drift, matching 2012 is an even less chance than 1/16.

Note that this is only if we're duplicating the 2012 popular result.  If it's more than 1% off of that, a 2012 map is a virtual impossibility.

Even if there's very, very little change in alignment from 2012, it's still pretty hard to duplicate the previous election's map.

[Hammy]

I'll shamefully admit being too lazy to check and ask if we've ever had two consecutive elections with the same map. (originally posted in wrong thread and moved here.)

[DS0816]

No.

Given that no presidential election’s map has later been duplicated…it is because the voting electorate does not stand still.

Historically, the modest shifting (say, no more than five percentage points; but commonly between 2 and 4 points)—in either party’s direction—tend to play out in elections in which the incumbent president wins re-election to a second full term. (1996 Bill Clinton gained 2.96. 2004 George W. Bush gained 2.98. 2012 Barack Obama declined –3.40.) Presidential elections, with a term-limited president who won two consecutive elections, are usually with dramatic national shifting (in either party’s direction; usually it’s in the direction of the opposition party.)

It actually would not be out of the ordinary for the 2012-to-2016 national shifting of the percentage-points margin, in the U.S. Popular Vote, to end up between 7 and 10. In Donald Trump’s direction that would, for potential, be a Republican pickup of the popular vote with as much as +6. I posted earlier in the year a map of holding all of Mitt Romney’s 2012 carried states and with pickups in Ohio, Florida, Virginia, Colorado, Iowa, New Hampshire, Wisconsin, Minnesota, Nevada, New Mexico, Pennsylvania, and Michigan—for a total of 352 electoral votes. In Hillary Clinton’s direction, for potential of winning nationally as much as +14, that would be Barack Obama’s 2012 re-election map with pickups in North Carolina, Georgia, Arizona, Missouri, Indiana, Nebraska’s 2nd Congressional District, Montana, South Carolina, Texas, Utah, and Nebraska’s 1st Congressional District—for a total of 453 electoral votes. Not that I necessarily would predict all of them—New Mexico or Utah—but just an envisioning of the best-case scenario (either way).

[Higgs]

Probably not in my opinion, as I see Trump performing better than many on Atlas think, but the "it's never happened before" certainly is no defense.

[AGA]

I'll shamefully admit being too lazy to check and ask if we've ever had two consecutive elections with the same map. (originally posted in wrong thread and moved here.)

That has never happened.

[Move Forward]

[HisGrace]

At the risk of sounding like Dick Morris, I'm still holding out hope for a Hillary landslide.  The Pew poll gives me reason to believe it will happen, particularly if Hillary does better among Latinos than the poll suggests.

I still have a hard time believing that all this Trump nonsense has gone on as improbably long as it has. It seems like eventually people have to realize that the emperor has no clothes and Clinton will win by a similar margin to Bush in 88.

If the election were held right now, I'd say NC and Arizona would flip Clinton's way