There are only 9 states (at most) that are considered “Battleground States (BG)”, meaning that the polls in those states are “close” (integer values are the Electoral Votes (EV)):
- Colorado 9
- Florida 29
- Iowa 6
- Nevada 6
- New Hampshire 4
- North Carolina 15
- Ohio 18
- Virginia 13
- Wisconsin 10
TOTAL=110 EC for BG
(All other states are known to be either Obama’s (EC=237) or Romney’s (EC=191) without any possibility of change.)
There are 538 total EC votes and thus one needs at least 270 to win.
Here is how it thus looks:
OBAMA=237 + his wins in BG
ROMNEY=191+ his wins in BG
If we were to assume that each BG state is a “50-50” coin toss, then on average each candidate will get (110)/2=55 EC votes:
So, here is the outcome (average):
None of the above analysis requires any computers, etc.
However, suppose we want to know THE PROBABILITY (P) that OBAMA wins under the “50-50” independent coin toss assumption above? (This requires some more serious computation.)
Well, we computed that exactly (via computer simulation) and here is the answer:
Now: looking at the BG polls it is NOT TRUE that each state is a “50-50” coin toss; the probability in each such state is different from 50% in general, and all but two states (NC, FL) have probabilities that favour Obama; it is as if we should use a “60-40” coin toss or a “75-25” coin toss for some of these states instead. (Consider Wisconsin as a good example: the probability of Obama winning is more like 90%, not 50%.)
Well, we estimated the correct coin toss probability for each state and then re-did the simulation and this is what happens:
OBAMA on average gets 310 EC and P=0.99
As a theoretical experiment, suppose we even give Romney both Florida 29, and North Carolina 15, and then re-do the simulation (with only 7 BG states):
OBAMA on average gets 297 EC and P=0.99: still, Obama wins; he does not need to win Florida and North Carolina in order to win.
As a theoretical experiment, suppose we even give Romney 1% more in each BG state poll, while taking away 1% from Obama (e.g., giving Romney an additional 2-point spread in each BG state), and then -re-do the simulation:
OBAMA on average gets 275 EC and P=0.66
Obviously, our model is very simplistic; but it looks very good for Obama at this point.
Business Insider Emails & Alerts
Site highlights each day to your inbox.