Update: Title fixed. How embarrassing.
At long last, the primary season is entirely over. As you probably already know, on the last day of primary season, the tea party handed Democrats the gift one a Senate seat in Delaware:
Senate Snapshot, September 15th
(With Rasmussen)
(Without Rasmussen)
Now, answering two questions that have come up in the comments:
Q1. Why does the non-Rasmussen snapshot say Democrats win 53 seats, even though they are ahead in enough to make 54 seats? (ATL Dem).
A1. Even though Democrats are ahead in enough campaigns to make 54 seats in the non-Rasmussen snapshot, their narrow leads in Wisconsin, California, Illinois and Nevada actually make 53 seats the most likely outcome. As you can see, in the non-Rasmussen snapshot, there are no campaigns where Republicans have a lead as small as Democrats have in all four of those seats.
Q2. Any comment on the difference between the Senate Snapshot and FiveThirtyEight? (Dismalest Scientist).
A2. There are two major differences. First, Nate Silver is attempting to calculate the odds of victory in November, while I am attempting to calculate the odds of victory if the election were held today. I stick with projecting the odds if the election were held today because it requires introducing far fewer assumptions into the methodology
Second, while Nate adjusts and weights poll results in several, largely intuitive, ways (sample size, recentness, house effect, past accuracy), and also introduces a demographic regression into his formula, I am simply conducting an unweighted simple mean of polls taken over the last 25 days.
I think my methodology has two advantages. First, an unweighted simple mean is easy enough, and transparent enough, for almost anyone to understand. Second, from 2008 to 2010, it was noticeably more accurate across the 52 general election campaigns where we both projected final margins. That doesn’t mean my methodology will be more accurate in 2010, but even I really do like testing the comparative accuracy of an unweighted simple mean against Nate’s complex formula.
Keep your questions coming...
Notes
--This is a snapshot, not a forecast. All of the odds presented here are based on if the election were held today. It is not a prediction of future trends.
--Only campaigns within 11% or less in the snapshot that includes Rasmussen polling are listed. If a campaign isn't listed here, then it is not currently as close as any of the campaigns listed here. In the case of Delaware, it is listed in this specific forecast because of all the hoopla surrounding it yesterday.
--All polls used in the averages are taken from Pollster.com.
--A complete description of the methodology behind this snapshot, along with all the research and a FAQ, can be found here.