Welcome to the first Daily Kos edition of my Senate forecast. For those unfamiliar with my Senate forecast, it is both very simple and highly accurate.
The forecast is produced by, in every state, taking the simple mean of almost all public polls with the majority of interviews conducted over the past 25 days. The only public polls not included are those conducted by Zogby Interactive (because they are so inaccurate), Strategic Vision and Research 2000 (because they are probably not real polls), and Columbus Dispatch (because they are conducted by snail mail). Every other poll in the public domain, including campaign polls and partisan polls, is included.
The “Current Dem Win %” is a snapshot, based on if the election were held today. It is not a prediction of future trends. It is calculated based on the error rates of this methodology in predicting the final margin of the 145 closest Presidential, Senatorial and Gubernatorial general election campaigns from 2004 to 2010.
In terms of accuracy from the final predicted margin to the final actual margin, this methodology significantly outperforms Pollster.com, fivethirtyeight.com, and Real Clear Politics. However, it is by no means a replacement for those excellent websites, as it only produces a snapshot of current polling. It does not show trendlines, contain polling archives, predict future campaign trends, or any of the other cool things those websites do.
In terms of presentation, I only show the campaigns where both parties are given a chance to win the seat if the election were held today. In the case of this specific forecast, Delaware is also included, because it is closer than Connecticut’s margin in the non-Rasmussen forecast. Oh, and I also present the forecast both with and without Rasmussen polls, because I imagine many people are curious to see what the Senate picture looks like without Rasmussen spam. I know I am.
So, here are snapshots of the 16 closest Senate campaigns right now. Without these campaigns, the partisan balance of the Senate would be 47 Democrats, and 37 Republicans. With Rasmussen polling included in the forecast, the most likely outcome would be 52 Democrats (including Lieberman and Sanders), 47 Republicans, and 1 Charlie Crist. Without Rasmussen polling, the most likely outcome would be 53 Democrats, 46 Republicans, and 1 Charlie Crist.
Senate Forecast, August 18
A * indicates that the candidate has a primary challenger, but is the heavy favorite to win the nomination. All polls used in the averages are taken from Pollster.com.
A complete description of the methodology behind this forecast, along with all the research and a FAQ, can be found here.