1.) While I do not anticipate voter turnout and polling errors in the state of Ohio to be off by the 2 1/2 points it would need to be to give Romney the state. However, it is very common with elections with this type of volatile numbers for poll data to be off around one and a half to two percentage points in either direction, due to dramatically higher voter turnout then expected from one party over the other. Even if increased republican turnout were to only close the gap by a mere one percentage point there then this could be significant.
2.) Two of the state polls in Ohio yesterday were showing between 4-6 percent of the Ohio population is still either undecided or could easily change their mind. While it is true that undecided voters generally strongly favor the challenger it is just not safe to include these past voter trends into projections of the future. It was for this reason alone that some prognosticators predicted Kerry to defeat Bush in 2004, despite the consensus of the polls showing Bush hanging onto a small lead. However, things ended up going the opposite of expected and polling showed that Bush ended up getting over half of the remaining undecideds. One factor that can be looked at in projecting which way undecided voters will go is the approval rating of the incumbent. Generally, one needs to have at least a 49% approval rating to have a reasonable chance of winning a second term. On election day in 2004 Bush was sitting right at 50%. Right now Obama's approval rating sits right on the danger line with an average rating of 48.6. Looking at the above numbers I do not think it is unreasonable to conclude that late deciders will at least slightly favor Romney. Let's assume once again for the sake of argument that Romney gains about one percentage point among these voters.
Obama's Approval Rating on RCP
| RCP Average | 10/15 - 10/26 | -- | 48.6 | 47.9 | +0.7 |
| Rasmussen Reports | 10/24 - 10/26 | 1500 LV | 47 | 52 | -5 |
| Gallup | 10/23 - 10/25 | 1500 A | 48 | 47 | +1 |
| ABC News/Wash Post | 10/22 - 10/25 | 1382 LV | 49 | 49 | Tie |
| Associated Press/GfK | 10/19 - 10/23 | 839 LV | 49 | 49 | Tie |
| CBS News | 10/17 - 10/20 | 1383 A | 49 | 42 | +7 |
| NBC News/Wall St. Jrnl | 10/17 - 10/20 | 1000 RV | 49 | 48 | +1 |
| Politico/GWU/Battleground | 10/15 - 10/18 | 1000 LV | 49 | 48 | +1 |
3.) Already we have a dead heat if both of the above things were to take place. Now instead of Romney having to gain over two percentage points in the next week and a half he would only need to gain around a half of one percent in the next week to give him a favorable chance at pulling out Ohio.
Now the disclaimers. I am not saying that Romney will or will not gain one point with higher turnout then expected. However, it should be considered encouraging that two of the recent Ohio polls were showing a 7-8 point edge in democrats then republicans. Why is this good news? It is safe to say that republican voter turnout cannot be worse then these pollsters are showing in their lopsided enthusiasm-gap findings. Chances are very good that they will fair at least some better then these numbers indicate.
I am also not saying that undecided voters will go for Romney. The oddity of this election is that the challenger is in a close race against an incumbent despite the fact that he has pretty dismal favorable numbers compared to past challengers.
It is important to understand that a Ohio victory for Romney does not give him the election, while it almost certainly would for Obama. Rommey will almost definitely need Colorado to win on election day and that state is a virtual tie using my formula. Then Romney would have to win Virginia and Florida to ensure victory.
No comments:
Post a Comment