Sunday, May 27, 2012

Sunday numbers, 2012, vol. 2: End of May electoral college

I'm keeping track of the numbers for the general election much earlier this time around than I did in 2008. I feel it's too early. Still, I'm going to do a snapshot of the situation every Sunday for my own edification and publish at the end of the month until the summer is over, at which time I'll publish every Sunday.  I'll also be keeping track of the Senate races once all the primaries are through.

If the election were held today,  there are 200 electoral votes that look to be solidly Democratic, 174 that look solidly Republican and 164 that I will call "toss-up", though you will see that right now one side or the other looks to be a prohibitive favorite in many of these races.  Here is the current toss-up list of fifteen states. The number in parentheses is the electors and the percentage is using the Confidence of Victory method I developed, which will change over time based on the latest polls.

Romney favored 
TN (11) 98%
AZ (11) 94.8%
IN (11) 94%
MT (3) 93.7%
MO (10) 75.7%
NC (15) 69.5%

Toss-up state
CO (9) 50%

Obama favored
WI (10) 89.7%
MI (16) 90.8% (gold)
VA (13) 91.5%
PA (20) 92.4%
OR (7) 94.7%
NV (6) 96.7%
ME (4) 98%
OH (18) 98.2%

The word (gold) is currently next to Michigan.  This means that if Obama can win Michigan and every state where his odds are better than Michigan, he will be president for a second term. Conversely, if Mitt Romney can win Michigan and every easier state, he will win.  Since the state is currently considered to Obama's advantage by a 90.8% Confidence of Victory number, he is favored overall.

Taking these toss-up states and looking at all the 32,678 possible combinations of outcomes, the median result is Obama with 303 electoral votes and Romney with 235 electoral votes. If Romney wants to console himself, that is much better than McCain did in 2008.  Romney has been known to brag about how much money Ted Kennedy had to spend to beat him in the Senate race in 1994.

Second place is still second place, Mitt.  It still means you don't get the job.

Here is a graph of the overall probability of victory for each side. A month ago it was at about 70%-30% for Obama and it improved steadily throughout the month. It changed in Romney's favor this week for the first time in a month, but the probability of victory for Obama is still at a daunting 98.5% to 1.5% if the election were held today.

I'll post new numbers at the end of June.

Here is a link to my methodology.

No comments: