An analysis of possible outcomes

[Hoverbored123]

At this point, I thought I'd do a run-through of all the possible situations we could see on November 6th.  I'll go in order, from best for Obama to best for Romney.  

First we have an Obama blowout, 357-181.  Obama sweeps the swing states; virtually a repeat of 2008, except that Indiana and Nebraska-2 flipped.  For the sake of argument we'll give Missouri to Obama.  

Next we have what is basically the absolute minimum Obama victory, 271-267.  There are other scenarios where Obama gets exactly 270, but they're not plausible.  Between this and the first map, there's tons of combinations in between.  

Next we have a minimum Romney victory.  Romney can theoretically win with 269, but this is the minimum outright victory.  

Here's the map from the updated Berry/Bickers projection.  It seems rather far-fetched, but that's what they say.  

According to the study, their average error is five states.  That means that, optimistically, Romney could flip the next five closest states in the Obama column: Nevada, New Jersey, Maine, Oregon, and Michigan.  It's worth noting that the study has these states at Obama 51% or less; if Romney can move the needle by a point, he can pick up all five.  That gives us this map:

Let's go even further.  Let's say Romney makes a late play for California.  With news of surging gas prices and even a possible shortage, Romney may have an opening.  If he can manage to flip the state at the last minute (as Obama flipped Indiana in 2008), we get this map:

This represents Romney's theoretical ceiling.  It's a remote possibility, to be sure, but one worth considering.  

UPDATE: Based on your response, and for the sake of fairness, I'm including some new maps depicting a possible democratic landslide.  This was a little harder for me to figure out, since Obama doesn't seem likely to win over a lot of states in "flyover country."  But, it's an interesting mental exercise to see how Obama could absolutely run the board:

This map is identical to 2008, except for one tweak: I've thrown in the next three strongest states for Obama in the Berry/Bickers study:  Missouri, Montana, and Arizona.  Missouri and Montana are actually fairly plausible, as Obama was very close to winning those states in 2008.

Even at this level, Obama is only leading 383 to 155.  It's a great victory, but not that much better than he did in 2008.  What would it take for Obama to pull off a true landslide?  After thinking about it for a second, I got an idea: why not throw in the states Bill Clinton carried?  

That gives him Louisiana, Arkansas, Tennessee, Georgia, Kentucky, and West Virginia, for a total of 437 to 101.  As with all landslide scenarios for 2012, this is the absolute maximum within the realm of remote possibility.  It's a great thought exercise, though.  To be honest, this started out as a "how might Romney win by a landslide" post, but after hearing your input, I added some food for thought for you Democratic posters.  This way, if something really weird happens that no one was expecting, I can say I called it in advance.  

[Oakvale]

worth considering. 

cc: Politifact

[7,052,770]

If you're going to include that last one, you should also include this one:

[J. J.]

This is my current map.

Obama:  281

Romney: 257

I'd give it a 50%, with about 20% roughly 50 EV either way  (10% is a Romney victory)

This is my realignment map:

I give it a 30% chance with Romney carrying at least one of the toss up states.

Romney:  395 (396)

Obama:  131 (142)

Overall, today, I'd Obama a 60% chance of winning.

[7,052,770]

30% chance of Romney getting 396 electoral votes?  IIRC you have a background in math or stats, so the only possibility is that you're trolling us.

[J. J.]

30% chance of Romney getting 396 electoral votes?  IIRC you have a background in math or stats, so the only possibility is that you're trolling us.

I'm looking at a 30% chance of a realigning election.  That deals with macro trends.  There is a whole thread on it that was started in January 2008.

When I first wrote about it, I was expecting the Republicans to win 2008 and this to be a Democratic year.  I was actually looking for something along the lines of 1932 in 2012. 

[Donerail]

"I saw their paper and I think there are glaring problems with their methodology. The U. of Colo. model fits the equivalent of 7 unknowns to 8 elections. That's not a good idea. The Colo. model also assumes huge effects from unemployment if incumbent is a Dem., but none if he's GOP. Hard claim to defend. Also, it's false advertising to claim CU model has predicted the last 8 elections right. It's a new model. Hasn't predicted anything yet."--Nate Silver

[Antonio the Sixth]

If you're going to include that last one, you should also include this one:

There are much more rock-solid R States than rock-solid D States. Adjusting for the national margin, North Dakota in 2008 was more Republican than Connecticut was Democrat.

[後援会]

Also, in the extremely remote, extremely unlikely possibility of a total economic collapse, Romney could pull a landslide. I don't see there being any possibility of the opposite happening.

[Hoverbored123]

Also, in the extremely remote, extremely unlikely possibility of a total economic collapse, Romney could pull a landslide. I don't see there being any possibility of the opposite happening.

To be fair, I consider the last map to be the absolute best possible outcome for Romney within the realm of possibility.  A number of things could push Romney into a decisive lead, including:

-a sudden economic downtown or financial collapse
-an abysmal jobs report on November 2 (the last before the election, if I'm not mistaken)
-a major foreign policy reversal (further embassy attacks, Eurozone collapse, etc.)
-a big surge in gas prices
-a major Obama scandal

If any of these things (or some combination of things) happens in the next month, a Republican landslide is possible.  Conversely, if the economy picks up in a big way, or if Romney has a big scandal himself, Obama could just as easily pull ahead.  I've updated this post to include the possibility of an Obama landslide. 

[Hoverbored123]

If you're going to include that last one, you should also include this one:

I'll do you one better: my updated post has a possible landslide map for Obama that's even better.

[Oldiesfreak1854]

At this point, I thought I'd do a run-through of all the possible situations we could see on November 6th.  I'll go in order, from best for Obama to best for Romney.  

First we have an Obama blowout, 357-181.  Obama sweeps the swing states; virtually a repeat of 2008, except that Indiana and Nebraska-2 flipped.  For the sake of argument we'll give Missouri to Obama.  

Next we have what is basically the absolute minimum Obama victory, 271-267.  There are other scenarios where Obama gets exactly 270, but they're not plausible.  Between this and the first map, there's tons of combinations in between.  

Next we have a minimum Romney victory.  Romney can theoretically win with 269, but this is the minimum outright victory.  

Here's the map from the updated Berry/Bickers projection.  It seems rather far-fetched, but that's what they say.  

According to the study, their average error is five states.  That means that, optimistically, Romney could flip the next five closest states in the Obama column: Nevada, New Jersey, Maine, Oregon, and Michigan.  It's worth noting that the study has these states at Obama 51% or less; if Romney can move the needle by a point, he can pick up all five.  That gives us this map:

Let's go even further.  Let's say Romney makes a late play for California.  With news of surging gas prices and even a possible shortage, Romney may have an opening.  If he can manage to flip the state at the last minute (as Obama flipped Indiana in 2008), we get this map:

This represents Romney's theoretical ceiling.  It's a remote possibility, to be sure, but one worth considering.  

UPDATE: Based on your response, and for the sake of fairness, I'm including some new maps depicting a possible democratic landslide.  This was a little harder for me to figure out, since Obama doesn't seem likely to win over a lot of states in "flyover country."  But, it's an interesting mental exercise to see how Obama could absolutely run the board:

This map is identical to 2008, except for one tweak: I've thrown in the next three strongest states for Obama in the Berry/Bickers study:  Missouri, Montana, and Arizona.  Missouri and Montana are actually fairly plausible, as Obama was very close to winning those states in 2008.

Even at this level, Obama is only leading 383 to 155.  It's a great victory, but not that much better than he did in 2008.  What would it take for Obama to pull off a true landslide?  After thinking about it for a second, I got an idea: why not throw in the states Bill Clinton carried?  

That gives him Louisiana, Arkansas, Tennessee, and Georgia, for a total of 424 to 114.  As with all landslide scenarios for 2012, this is the absolute maximum within the realm of remote possibility.  It's a great thought exercise, though.  To be honest, this started out as a "how might Romney win by a landslide" post, but after hearing your input, I added some food for thought for you Democratic posters.  This way, if something really weird happens that no one was expecting, I can say I called it in advance.  

Clinton carried Kentucky twice and Georgia once.

[Hoverbored123]

I knew I forgot something.

[izixs]

The many years I've lurked and the few that I've posted have left me with one conclusion about prediction maps on the forum here: Most predictions on the forum are poorly conceived. Aka, the average poster, left or right, is pretty bad at making predictions individually. The question is... will the average prediction get close to the truth or not? If it does, that suggests something kind of neat.

[Antonio the Sixth]

The many years I've lurked and the few that I've posted have left me with one conclusion about prediction maps on the forum here: Most predictions on the forum are poorly conceived. Aka, the average poster, left or right, is pretty bad at making predictions individually. The question is... will the average prediction get close to the truth or not? If it does, that suggests something kind of neat.

The 2008 final prediction was godawful, fwiw.

[Hoverbored123]

I modified the final map to include Kentucky and West Virginia in the Clinton column.  I forgot he carried both states in 1992 and 1996.