[Update: An analysis using some newer polls can be found here.]
Since the previous analysis in this race, five new state head-to-head polls representing four states have been released.
start | end | sample | % | % | % | |||
---|---|---|---|---|---|---|---|---|
st | poll | date | date | size | MOE | O | R | diff |
FL | Mason-Dixon | 24-Jan | 26-Jan | 800 | 3.5 | 44 | 48 | R+4 |
FL | Quinnipiac | 19-Jan | 23-Jan | 1518 | 2.5 | 45 | 45 | tie |
MI | EPIC/MRA | 21-Jan | 25-Jan | 600 | 4.0 | 48 | 40 | O+8 |
MN | PPP | 21-Jan | 22-Jan | 1236 | 2.8 | 51 | 41 | O+10 |
PA | Keystone Poll | 17-Jan | 22-Jan | 614 | 4.0 | 41 | 30 | O+11 |
Two new polls come from Florida, where the media markets are currently flooded with Republican primary ads. The newest poll from Mason-Dixon gives Romney a small +4% lead over Obama. The slightly older Quinnipiac poll has the race all tied up at 45% a piece.
In Pennsylvania, Obama leads Romney by an impressive +11%. That’s even better than Michigan, where Obama leads Romney by +8% in the new poll. The Pennsylvania poll is more favorable to Obama than the newest Minnesota poll that has Obama up by +10%
In the previous analysis, Obama would have won with a 71.9% probability, and his average electoral vote total was 284 to Romney’s 254.
Now, a Monte Carlo analysis using 100,000 simulated elections gives Obama an expected electoral vote total of 306 to Romney’s 232 for an election held now. Obama is at a 95.1% probability of winning that hypothetical election to Romney’s 4.9%. By traditional statistical inference, we would say Obama’s lead is “significant”—that is, it’s unlikely to be due to sampling error.
Obama | Romney |
95.1% probability of winning | 4.9% probability of winning |
Mean of 306 electoral votes | Mean of 232 electoral votes |