I have written articles pertaining to Ohio and Virginia's polling numbers, on how turnout differential determines elections in a vast majority of the cases. However, the state which may show us this dynamic is not Ohio or Virginia, but Pennsylvania (and possibly Iowa, Michigan, and Minnesota).
Mitt Romney made a late push for Pennsylvania. Personally, I think it was a huge mistake to wage wars of attrition with President Obama over Ohio, Wisconsin, Colorado, Nevada, and New Hampshire. Romney should have had a greater reach and focused more intensity across the Rust Belt from Wisconsin to Michigan, to Ohio to Pennsylvania, to Upstate New York and New Hampshire.
Romney campaign advisors picked to win two huge wars in Ohio and Wisconsin instead of spreading out, infiltrating, and trying to put all the Rust Belt into play fighting many small battles. President Obama's team said Romney's push for Pennsylvania was just a decoy and desperation, which is only half truth and all spin. If Romney would have hit all these states hard, after the first presidential debate in Denver, this election might have been a Romney landslide. One only needs to look at composite of Pennsylvania polls (PPP, Gravis, and Morning Call) in the last couple of days before the election to see the president's former 6% to 10% lead is cut to 2% to 3% based off internal numbers in only 7 days. I doubt Romney has the momentum to turn those numbers into a lead based on D+4 to D+6 projections, which is the near optimal scenario for Romney in Pennsylvania turnout differential.
For Romney to win Pennsylvania now his team would have to cut the turnout differential to D+2 to take a lead based on the aforementioned polls. If Romney gets the differential to D+2 he wins the state based on momentum. The problem is getting a D+8 to D+10 state which was trending D+6 to D+8 10 days ago into a sub D+4 scenario. It is plausible and conceivable for such a change in turnout momentum, but if it was plausible in Pennsylvania, it was also plausible in upstate New York, in Michigan, and Minnesota. From a militaristic and systems dynamic point of view it would have been much harder for Obama and his team to counter momentum from the first debate in many multiple new battlegrounds as opposed to the known ones (Ohio and Wisconsin).
Kudos to President Obama and his campaign team for being able to block Romney's massive play for Ohio and Wisconsin. They were playing defense from the get go and will probably pull this election off, because they correctly guessed their opponents moves. Team Romney's late plays for Pennsylvania and Michigan were not expected by Team Obama. While these late play's will likely not payoff for Romney they were hardly desperate or a decoy, they were viable plays all along and under utilized.
Lastly, here is my call on the election results for Ohio, Virginia, and Pennsylvania — based on computation of polls in my model:
Obama +1.5% to -.5%
D+4 Turnout: Obama 48% Romney 47%
Romney +2.5% to +1.5%
D+4 Turnout: Romney 50% Obama 48%
Obama +3.5% to 2.5%
D+5 Turnout: Obama 50% Romney 48%