Member Log in


Remember Me

Forgot Password?

Not a member yet?

What We Do

PollingPoint conducts surveys over the Internet about public affairs.....

»more

Who We Are

PollingPoint is managed by Polimetrix, a non-partisan polling...

»more

What You Do

Voice your opinion! Tell us what you think about the latest issues...

»more

Back to Headlines

Handicapping the 2006 Election: What are the odds of the Democrats winning control of the Senate?

Palo Alto, CA, November 06, 2006 - During the 2006 election, Polimetrix has conducted a series of surveys, including a consortium of research universities and leading U.K. polling organization YouGov. On the eve of the election, Polimetrix is reporting state-by-state results from the Cooperative Congressional Election Study conducted between October 27 and November 5, 2006

The Cooperative Congressional Election Study is led by Professor Stephen Ansolabehere of the M.I.T. Department of Political Science and includes participants from 34 universities, including Arizona State University, Brigham Young University, California Institute of Technology, Columbia University, Dartmouth College, Duke University, Florida State University, George Washington University, Harvard University, Michigan State University, Massachusetts Institute of Technology, University of Notre Dame, Ohio State University, Reed College, Stanford University, Temple University, University of Akron, University of California (Berkeley, Davis, Los Angeles, Merced, Riverside, San Diego), University of Chicago, University of Illinois, University of Maryland, University of Michigan, University of Minnesota, University of North Carolina (Chapel Hill), University of Pennsylvania, University of Pittsburgh, University of Texas, University of Washington, Vanderbilt University, and Yale University. See http://web.mit.edu/polisci/portl/cces for further details.

The survey contains both a pre-election interview and a post-election follow-up. This release reports the results of 20,331 interviews conducted in 32 states where the sample included at least 300 likely voters. The samples are selected from Polimetrix's PollingPoint panel and matched to a sampling frame of registered voters. The sampling error for a sample of size 400 is estimated to be approximately plus or minus 7%. This includes normal sampling error from the frame plus an adjustment for matching error.

The analysis in this release is co-authored by Stephen Ansolabehere, Samantha Luks, and Douglas Rivers.

The Senate: A 50/50 split is more likely than Democratic Control.

In the final day before the election, control of the Senate remains at issue. Currently the Democrats have 45 seats and need a pickup of six seats to gain control. Based on our poll results, only eight seats remain in play - six held by Republicans and two by Democrats. Democrats need to win six of these eight seats. What are the odds?

The Missing Polls: Montana and Rhode Island
We had fewer than 300 interviews with likely voters in both Montana and Rhode Island, which is insufficient for producing an accurate forecast. However, Republican incumbents in both of these seats are widely believed to be in trouble.

Bottom Line: Since we don't have any data to add for these races, we'll go with the odds given by InTrade: 74.5% probability of the Democrats winning Montana and a 78.0% probability of winning Rhode Island, which gives a 58% probability of the Democrats winning both (assuming the outcomes are independent). Better than even odds, but hardly a certainty.

Vulnerable Democratic Seats: New Jersey and Maryland
Only two Democratic seats are in play at this stage. In New Jersey, Robert Menendez, who was appointed to fill the vacancy when Senator Jon Corzine became Governor, holds a 53-47 lead over State Senator Thomas Kean, Jr. In the battle to succeed retiring Democratic Senator Paul Sarbanes in Maryland, Democrat Benjamin Cardin leads Republican Michael Steele 51-45. Steele, who hopes to attract votes from normally Democratic African-American voters, appears to be coming up short.

Bottom Line: Democrats hold both of these seats. Not a sure thing, but the probability is high - perhaps 80%

Must-Wins for Republicans: Tennessee and Arizona
At the beginning of the year, both Tennessee and Arizona seemed like relatively safe seats for Republicans. The Arizona race has tightened considerably: incumbent Republican Jon Kyl now leads Democrat Jim Pederson by just 50-46, with the momentum in the Democrat's favor. In Tennessee, Democratic Congressman Harold Ford, Jr.'s bid to capture the seat held by retiring Republican Majority Leader Bill Frist appears to be coming up short. Republican Bob Corker now holds a 51-46 lead.

Bottom Line: Although both races are close enough for an upset, it appears that Republicans will hold these two seats. The chance of a Democratic pickup here is no more than 20%.

The Tossups: Missouri and Virginia
The two closest races in our sample are Missouri and Virginia. In Missouri, challenger Claire McCaskill holds a razor-thin 0.6% lead over incumbent Jim Talent, while in Virginia incumbent George Allen holds a 1.0% lead over former Navy Secretary James Webb. Both of these are well within the margin of error and, with rounding, actually end up as 50-50 ties!

Bottom Line: Anybody's guess. These races are as close to even odds as you can get, which means that the probability that Democrats pick up both seats is only about 25%. People do get two heads in two tosses all the time, but you shouldn't bet on it.

Who will control the Senate?
Assuming the parties hold onto their other seats (or any loss in New Jersey or Maryland is offset by a pickup in Tennessee or Arizona), the Democrats still need an inside straight: win all four of Missouri, Virginia, Montana, and Rhode Island. This is about a one-in-seven chance - it could happen, but the odds are against it.
Much more likely, in our opinion, is a Senate evenly divided between Democrats and Republicans with Vice President Dick Cheney breaking ties.


Polimetrix Senate Polls

State Democrat Republican Other
Arizona Pederson Kyl
(n = 798) 46% 50%
California Feinstein Mountjoy
(n = 1015) 61% 29%
Connecticut Lamont Schlesinger Lieberman
(n = 401) 44% 5% 48%
Florida Nelson Harris
(n = 1005) 60% 34%
Maryland Cardin Steele
(n = 802) 51% 45%
Massachusetts Kennedy Chase
(n = 799) 67% 27%
Michigan Stabenow Bouchard
(n = 800) 55% 40%
Minnesota Klobuchar Kennedy
(n = 501) 57% 39%
Missouri McCaskill Talent
(n = 802) 50% 50%
Nevada Carter Ensign
(n = 402) 42% 53%
New Jersey Menendez Kean
(n = 500) 53% 47%
New York Clinton Spencer
(n = 1011) 70% 26%
Ohio Brown DeWine
(n = 1003) 58% 40%
Pennsylvania Casey Santorum
(n = 1005) 56% 40%
Tennessee Ford Corker
(n = 502) 46% 51%
Texas Radnofsky Hutchison
(n = 1004) 29% 65%
Utah Ashdown Hatch
(n = 402) 34% 66%
Virginia Webb Allen
(n = 802) 50% 50%
Washington Cantwell McGavick
(n = 804) 57% 43%
West Virginia Byrd Raese
(n = 301) 67% 33%
Wisconsin Kohl Lorge
(n = 502) 69% 24%


Polimetrix Gubernatorial Polls

State Democrat Republican
Alabama Baxley Riley
(n = 505) 42% 58%
Arizona Napolitano Munsil
(n - 798) 57% 41%
California Angelides Schwarzenegger
(n = 1015) 42% 52%
Colorado Ritter Beauprez
(n = 500) 60% 40%
Connecticut DeStefano Rell
(n = 401) 32% 64%
Florida Davis Crist
(n = 1005) 42% 54%
Georgia Taylor Perdue
(n = 804) 42% 59%
Illinois Blagojevich Topinka
(n = 800) 61% 39%
Iowa Culver Nussie
(n = 301) 54% 46%
Kansas Sebelius Barnett
(n = 501) 61% 39%
Maryland O'Malley Ehrlich
(n = 802) 50% 48%
Massachusetts Patrick Healey
(n = 799) 61% 32%
Michigan Granholm DeVos
(n = 800) 55% 43%
Minnesota Hatch Pawlenty
(n = 501) 49% 45%
Nevada Titus Gibbons
(n = 402) 44% 51%
New York Spitzer Faso
(n = 1011) 69% 24%
Ohio Strickland Blackwell
(n = 1003) 62% 35%
Oregon Kulongoski Saxton
(n = 502) 54% 46%
Pennsylvania Rendell Swann
(n = 1005) 61% 37%
South Carolina Moore Sanford
(n = 399) 44% 56%
Tennessee Bredesen Bryson
(n = 502) 63% 31%
Texas Bell Perry
(n = 1004) 28% 43%
Wisconsin Doyle Green
(n = 502) 53% 43%

Back to Headlines