In order for any Republican to defeat Barack Obama in 2012, it obviously comes down to performance in key states and flipping certain states that Obama carried in 2008.
I am starting with an assumption that states like North Carolina, Virginia and Indiana are essentially red states that caught Obama fever in 2008 and voted Democratic. However, after four miserable years of Barack Obama's over-reach and expansion of the Federal government's size, scope and spending, most should lean back to the Republican fold in 2012. Going through polls in individual states thus far and using an average of those polls to gain a broader view of Obama's chances in these states, a synopsis was determined. Although nationally Obama's approval ratings are dismal as are national polls of the direction of the country, you have to look to his numbers in teh key states and even among certain demographic groups in those states to get a better look at his chances.
A lot has been made of the changes in the Electoral College as a result of the 2010 Census. Texas, a decidedly red state, was the biggest winner, picking up four electoral votes while blue states like New Jersey and New York lost political clout. In fact, traditionally blue states lost seven electoral votes while traditionally red states picked up those seven votes. Among so-called swing states (Florida, Nevada, Ohio and Missouri) the net gain was zero. Assuming everything goes as planned, the Republican candidate starts with seven more electoral votes than they had in 2008 from the get go. However, demographics dictate that Republicans do not take Georgia or Arizona for granted in 2012.
Based upon that state polling data, it is probable at this early stage of the game (primarily based upon Obama's approval ratings in these states) to predict he will lose states that he won in 2008. They are Indiana, Pennsylvania, North Carolina and New Hampshire for a total flip of 71 electoral votes. Added to the bonus seven points from the electoral vote shifts from the Census, that is a total shift of 78 electoral votes to the Republican Party in 2012. It sounds great, but it is not enough to win the election. Obama would still win the election by 42 electoral votes (290-248). Even flipping Virginia would not create a Republican victory, just make it closer (270-261).
Looking at Obama's popularity rating in states that he won in 2008 that Republicans need in 2012, we see the following:
Colorado 48.5% North Carolina 46.8%
Florida 48.0% Ohio 46.0%
Missouri 43.0% Pennsylvania 44.8%
Nevada 47.3% Virginia 48.2%
New Hampshire 46.0%
Using a 47% approval rating as the make-or-break point to give an incumbent, flailing President the benefit of the doubt, it looks like Obama keeps Nevada, Florida, Colorado and Virginia. Thus the key to electoral victory flows directly through Florida again. Either the Republican nominee takes Florida and their 29 electoral votes, or they have to take surprise states in the upper Midwest like Michigan and Minnesota. Even taking New Mexico, Colorado and Nevada would not create a Republican victory. And forget the Northeast and the Left Coast. The easiest path would be to concede these states along with Virginia and concentrate on Florida in 2012. Also, it would not take too much to take Florida. Obama's numbers among key demographics largely mirror his numbers in key states generally, although they are slightly below average in some key categories- moderates, political independents, Hispanics and the 30-45 year old voter.
The following table shows his Florida numbers compared to the general overall numbers in key states for these categories:
Category Overall/key states In Florida
Moderates 63.1 66.0
Independents 45.9 48.5
Hispanics 59.6 57.5
30-45 year old 47.7 52.5
This indicates that if these trends hold true through 2012 among these key demographics in Florida, Barack Obama wins the state and reelection. The most recent polling shows that Mitt Romney would give Obama the toughest time in Florida (but this was before Perry entered the race), but still lose by 5 points.
Naturally, this is all hypothetical as so much could happen in the next 15 months. This is merely an early analysis.