Rasmussen is the only poll to have Obama down in Virginia over this past week.
Go Mitt, go!
Do me a favor and go back to our 2008 polling page. Look at Rasmussen's final polls:
Colorado: Rasmussen (O +4 ) Final Result: (O + 9)
Nevada: Rasmussen (O + 4) Final Result: (O + 12)
Wisconsin: Rasmussen ( O + 7) Final Result: (O + 14)
Ohio: Rasmussen (Tie) Final Result: (O + 5)
Florida: Rasmussen (M + 1) Final Result: (O + 3)
Virginia: Rasmussen (O + 4) Final Result: (O + 6)
North Carolina: Rasmussen (M + 1) Final Result: (O + 1)
New Hampshire: Rasmussen (O + 4) Final Result: (O + 9)
What was that about reliable??
Again, I take these state numbers with a grain of salt. I know there are assumptions in them that do not match the national shift and the early vote data, which basically confirms this shift.
Where Rasmussen went wrong in 2008 is one the state level. Why? There were turnout assumptions baked in there based on the 2004 data. Similarly, some of these D-leaning pollsters are using similar 2008 breakdowns when they adjust the polling based on race and geography or in some instances party ID.
I trust the National polls. That shows a dead heat. The early voting in most states confirms a tightening on the state level. Obama is down in absolute terms in early voting and Romney is up. That number is probably in the 6-14 point range depending on the state.
When I see that the absentee numbers are down in the Dem-leaning counties in Virginia and up in the Republican counties, it tells me that there's a shift off of the five-point Obama win in 2008. Polls like PPP show a 2-point swing from '08. The early numbers -- actual data -- say otherwise.
At the end of the day, I am not saying that Romney has this one in the bag. He doesn't. The race is close on both the state level and a number of swing states. However, these comfortable Obama leads in states like Ohio and Virginia defy logic.
