Wednesday, June 6, 2012

Some Early Electoral Math

Someone made the mistake of posting an interactive electoral map on the internet, which meant that I spent significantly more time today than I can possibly justify playing around with it. Long-time readers know that I'm an electoral math junkie, so playing around with it is one of the few joys I will have in what will no doubt turn into a recklessly dreary election season.

Anyway, I fussed around with the numbers a bit and it isn't going to take a whole lot of political acumen to declare that it all comes down to--yet again--Florida. Under pretty much any scenario, the candidate that wins Florida will win the election outright. Thanks to an increase in Florida's population and subsequent decrease in Ohio's, nearly any practical combination of states won't be able to effectively overcome a victory in Florida.

The map above (which I'm sure won't work in a week due to its flash-based nature; I'm sure you can find others like it) has it pretty much right: the only swing states in play right now are Nevada, Iowa, Florida, Ohio, New Hampshire, Virginia, and Colorado. Add up everything but Florida and you get 58 votes. Thanks to Obama's lead in most other states, Romney could win all of the states except Florida and lose, so Florida is a must-win for him. (He'll have to pick up some other states as well, basically either Ohio or all the others.)

Obama is in better shape but not much: he can afford to lose both Florida and Ohio but would have to sweep pretty much everything else.

The good news for Romney is that each of these swing states clearly leans Republican and both Florida and Ohio--the big ones--were won by very thin margins for Obama in 2008. In addition, the census reallotment favored the red states by a good bit. There's also the "rising tide" running against Obama--basically, if Romney wins, say, Ohio, he's probably won the rest of the swing states as well. (The reverse was true in 2008 when Obama won Indiana--the first state declared. The race was over at that point. If Obama was winning reliably rock-red Indiana, previously thought to be out of play, he was going to sweep the other swing states easily. As he did.)

Of course, all of this is predicated on the assumption that the only swing states are the seven listed above. While this is probably the case, save a state or two that might shake loose either way, it's also possible that more are in play as the campaign progresses.

Let's take a look at the past three elections. We can't go much further back than that, since the political landscape has changed too much to read too far back than that. (One can certainly argue that the landscape has changed simply between 2004 and 2008, but given how the polls look I'm not willing to make that bet quite yet--it seems reasonably clear that 2008 was a spike in Democratic support as opposed to a meaningful realignment, especially given the gains Republicans made in 2010.) By looking at the margin of victory of each candidate the top swing states are (in order of swinginess):

New Hampshire
New Mexico

Everything from Florida to Wisconsin has an average of less than 5% margin of victory per year. The math here's a little wonky, I realize, but we have to go with what we have.

The list would look a lot different if we lopped off 2008; in fact, one of the surprising things looking at the list is how close the 2004 election really was. States that are a given for Obama this year were almost lost by Kerry that year--for example, Kerry won Oregon by only 4%, despite the fact that Obama won it by almost 17%. The same stats apply in Michigan--Kerry won by around 4%, but Obama carried it by 17%. Even in Wisconsin, Kerry won by .3%, while Obama ran away with 14%. (A lot of the percentages are going to be deceptively strong for Obama. There's no way he wins North Carolina again, even though he won by a comfortable margin. The violent reaction in the 2010 congressional elections confirmed that the few states who flipped to Obama, like Indiana and NC, aren't going to be easy for him this time around.)

If the factors that made 2008 so different no longer really apply--if the same enthusiasm that propelled Obama to office four years ago fades and everything reverts back to 2004 levels--then the number of swing states will dramatically increase, probably to over 15 or so. States currently assumed to be safe now become in play; If, say, Pennsylvania (a state Kerry won with barely 3% but Obama won with 10%) suddenly is in contention, a whole new set of math becomes apparent, and Florida is no longer necessary.

And that's the difficulty with identifying the states that are truly in play. It's hard to declare many states such as Pennsylvania, Michigan, Wisconsin, or Minnesota as true swing states, since they almost always go Democratic. (Of these four states, all of them have gone Democratic in every Presidential election since 1992.) Yet they are declared swing states each year because the margin of victory each election is so small, usually under 5% (the standard threshold for being "in play"). Every year, there's always a chance that they might tip Republican, yet it never happens.

The good news for Democrats is that they never seem to flip to the GOP, so while they seem like swing states they are probably safe. (Even this year, only Wisconsin seems in jeopardy, although Pennsylvania polls are somewhat worrying for Obama--but then again, it's one of those things where if Romney wins PA, the election is over because he's won all the other swing states as well.) The bad news for Democrats is that there aren't many similar GOP states that might tip Democratic. With the exception of Virginia, all of the states likely to tip from blue to red are smaller states like Nevada and Iowa.

There's a lot left of the campaign season to go, so a lot can change. But my predictions are that the original seven states listed above will be true swing states, save Virginia. The surprises will be that Wisconsin leans surprisingly Republican, while Arizona will be vulnerable for Romney. Pennsylvania could be in play as well. Romney will think that Michigan will be in play but it won't; likewise, Obama will try to woo Virginia and North Carolina, two states he did surprisingly well in, but they will not be interested.

And the riskiest prediction? West Virginia just might--might!--go for Romney.

No comments:

Post a Comment