Since the last update on November 4th, a few new polls have resulted in some minor alterations to the overall Senate forecast:
California has tightened up a bit, but that is because of polls that fell out of the average, not because of any new polling showing Republican improvement.
Connecticut remains at "solid Republican," but the rise of Linda McMahon in the Republican primary almost moves it to "lean Republican" or even "toss-up." A second poll will be needed for confirmation before a change is made, however.
Delaware moves from "lean Republican" to "toss-up"
Illinois moves from "toss-up" to "lean Republican
Ohio has slipped a bit closer to Republicans.
North Carolina has new polling, but remains in the same position as the last forecast.
All of these changes actually make no change in the overall forecast. Now, as with two weeks ago, the most likley outcome if the election were held tomorrow is a Republican net gain of four seats.
However, the election is not held tomorrow. It will be held in 50 weeks, making this prediction by Nouriel Roubini all the more worrying:
Based on my best judgment, it is most likely that the unemployment rate will peak close to 11% and will remain at a very high level for two years or more.
If that is the case, Democrats will be lucky to only lose four seats.
With Mike Castle set to run for Senate in Delaware, Republicans have rounded out an excellent season of candidate recruiting for them. Their current strength in not only Delaware, but also Colorado, Connecticut, Florida, Illinois and New Hampshire is the result of top-tier recruiting efforts. By contrast, Democrats have seen their chances in a number of campaigns--Arizona (Napolitano), Kansas (Sebelius), and North Carolina (Cooper)--take serious blows due to failures to snag top tier recruits. The situation in reminiscent of 2006, when Democrats held the upper hand in Senate recruitment, and used it to make significant gains that year.
Here are some possible reasons to be optimistic, despite this forecast:
Rasmussen, which skews significantly toward Republicans, is over-represented in current polling averages.
A shift of 5% across the broad in these campaigns, some of which can already be accounted for in Rasmussen's over-representation in these averages, would bring Democrats even in the forecast.
Republicans are not within 15% of any other potential pickups, so it can't get much worse for Democrats.
For the next year, the two biggest factors in Senate campaigns will be the outcome of primary elections, and the state of the economy. More than ever, it is clear that Democrats will take a real electoral hit if there is not a significant electoral turnaround.
Current 2010 Senate Outlook: Net Republican Gain of 3-5 Seats Republican-Held Seats Democratic Pickups: 1-2 (Ohio, and possibly Missouri)
Keep an eye on Florida as well, if Republican Governor Charlie Crist faces trouble in his primary against Marco Rubio. Texas may also have a special election for Kay Bailey Hutchinson's vacated seat.
Democratic-Held Seats Democratic Losses: 5-6 (Colorado, Connecticut, Delaware and Nevada, plus one or two of Arkansas, Illinois and Pennsylvania)
(Barbara Boxer's seat in California, Russ Feingold's seat in Wisconsin, Patty Murray's in Washington, and the open seat special election in Massachusetts are some other campaigns to watch.)
Notes:
*= Faces primary, but heavy favorite
**= Faces serious primary challenges, but only candidate for which general election polling is available.
***= Not an announced candidate at this time
--Illinois is rated behind Pennsylvania, despite the better polling average for Republicans in Illinois, because there is only one recent poll of Illinois.
The forecast is based entirely on polling conducted after May 31st.
A wave of new polling over the last week has prompted an update in the Senate forecast. The overall seat change prediction remains the same, with Republicans netting 2-3 seats resulting in a Democratic caucus majority of either 58-42 or 57-43. The two category changes are Arkansas moving from "toss-up" to "lean Dem," and Colorado moving from "lean Dem" to "toss-up."
The recruiting phase of the 2010 elections is nearly over. With the exception of the two special elections (Massachussetts and Texas), the remaining holdout on the recuriting front is Delaware. In the first state, much still hinges on whether Beau Biden and / or Mike Castle ends up running, the major candidates. the latst indications seem to be that Beau Biden is, in fact, running.
Primaries are the next stage in the Senate picture. Here are the best case scenarios for each party:
Democrats: In Delaware, Biden runs, Castle does not; Paul is the Republican nominee in Kentucky; Norton is not the Republican nominee in Colorado, Crist is not the Republican nominee in Florida; Hendren is the Republican nominee in Arkansas. This would result in a forecast where Democrats net 0-1 seats.
Republicans: In Delaware, Castle runs; Grayson is the Republican nominee in Kentucky; Norton is the Republican nominee in Colorado; Crist is the Republican nominee in Florida; Baker or Coleman is the Republican nominee in Arkansas. This would result in a forecast where Republicans net 3-4 seats.
Of course, beyond primaries, much still depends on the overall political environment. If a good health care bill passes and a real economic turnaround are on the way by the spring or summer, then Democrats could very well net seats no matter the outcome of primaries. However, it is possible that everything could go wrong for Democrats in terms of primaries, recruiting and the political environment, thus resulting in a Republican net gain of 6-7 seats.
Current 2010 Senate Outlook: Net Republican Gain of 2-3 Seats Republican-Held Seats Democratic Pickups: 2 (Missouri and Ohio)
Russ Feingold's seat in Wisconsin, and Patty Murray's in Washington, are two other campaigns to watch.)
Notes:
*= Faces primary, but heavy favorite
**= Not an announced candidate at this time
***= Faces serious primary challenges, but only candidate for which general election polling is available.
Open Left's Senate forecast for August shows Republicans picking up seats for the first time since, well, ever. Although the 2-3 seat Republican gain I forecast is a far cry from the 6 seats Arlen Specter thinks Republicans will gain, it still allows us to maintain our inglorious title as the most favorable to Republicans among all forecasters.
While I must sound like a very boring broken record at this point, almost everything still depends on the state of the economy. There isn't a huge amount of recruiting left to be done, so changes in the campaigns from this point forward will largely depend on the national political environment. If the economy starts to improve in the Spring of 2010, then Democrats should at least draw even, and possibly even gain seats. If the economy does not improve, it is possible Republicans could ring up a 5-7 seat gain.
The fate of health care reform represents another major factor. If health care reform either fails altogether, or only an extremely weak reform is passed, expect Democratic turnout to drop sharply. This happened in 1994, too. New Jersey and Virginia should make good test cases for just how serious the threat of lowered Democratic turnout actually is.
This forecast is based entirely on polling. In the 2004, 2006 and 2008 elections, simple poll averaging proved as effective in predicting the final margin as any other methodology (as long as there were two or more polls in the final week, and as long as the outcome wasn't a foregone conclusion of 20% or more).
Current 2010 Senate Outlook: Net Republican Gain of 2-3 Seats Republican-Held Seats Democratic Pickups: 2 (Missouri and Ohio)
Russ Feingold's seat in Wisconsin, and Patty Murray's in Washington, are two other campaigns to watch.)
Notes:
*= Faces primary, but heavy favorite
**= Not an announced candidate at this time
***= Faces competitive primary, but general election polling not available for other candidates.)
Political analysts predict Democrats will lose six Senate and 20 House seats next year, said Specter. He said likely Republican nominee Pat Toomey, a former congressman who is president of the anti-tax Club for Growth, has raised a lot of money for the Senate campaign.
"It's going to be a tough general election," Specter said.
Lose six seats in the Senate? What political analysts are saying that? Here are the four major election forecasters who have published complete Senate analyses for August. On average, they actually show more Republican-held seats threatened than Democratic-held seats:
Cook: 6 Republican-held toss-ups or leans, 5 Democratic-held toss-ups or leans
CQ Politics: 8 Republican-held toss-ups and leans; 5 Democratic-held toss-ups and leans
Rothenberg: 6 Republican-held toss-ups or leans; only 4 Democratic-held toss-ups or leans
538: 5 Republican held seats among top 12 campaigns; 7 Democratic-held seats among top 12 campaigns
Rather than predicting a Republican gain of six seats, as per Arlen Specter, these four forecasters average slightly more Republican-held Senate (6) seats as potential flips than Democratic-held seats (5).
Thanks for the pep-talk Arlen. "The party is screwed in 2010, so choose me or be even more screwed." It is always fun when party higher-ups decide to become concern-troll, Village pundits in a attempt to cajole Democrats into accepting the status-quo of powerful institutions.
"I am very concerned about one-party rule in Washington," Specter says in the spot. "That's why it's so important to elect Chris Hackett to Congress. With Chris Hackett, we'll have a check against massive Democratic control. And that's vital for our country."
Perhaps Specter talking about substantial Democratic losses in the Senate is simply his fear of "massive Democratic control" coming to the surface again.
While glimpsing at my blackberry during vacation, I was eager to put up a new Senate forecast upon arriving back in Philadelphia. Senate forecasting isn't the biggest news right now, but all of the developments over the past three weeks made this an essential piece of housekeeping.
The current forecast shows no net partisan change in the Senate, as Democratic gains in Ohio and Missouri are offset by losses in Connecticut and Delaware. Further, the number of close campaigns is increasing. As the chart below shows, about a handful of campaigns on both sides could also potentially switch. Due significantly to better candidate recruiting, Republicans have a slight advantage in the number of close, potential pickups they have created.
The complete forecast, including colorful charts, more causes for Republican improvement, and implications for ongoing legislative fights, can be found in the extended entry.
As of today, we are more than one-third of the way between the 2008 and 2010 general elections. In order or this arbitrary milestone, here is a look at the 2010 Senate picture.
The charts have been expanded to account for competitive primaries, and hopefully to provide greater clarity. The overall forecast, showing a net Democratic pickup of one seat, has not changed. The methodology, focusing only on polls, has also not changed.
2010 Senate Outlook: Net Democrat Gain of one seat, for a total of 61 Republican-Held Seats Democratic Pickups: 3 (Missouri and Ohio, plus either Kentucky or New Hampshire)
(Russ Feingold's seat in Wisconsin, and Patty Murray's in Washington, are two other campaigns to watch.)
Notes:
*= Not certian to enter campaign yet
**= Faces primary, but no polling for challenger
***= All Pennsylvania polling taken May 15th or later
****= There is no polling in Arkansas for Lincoln against any of the three announced Republican candidates. So, polling on Lincoln vs. other candidates, who are probably equally unknown, is listed instead.)
North Carolina and Nevada are listed at the bottom of their respective tables because, while the incumbents in those states appear vulnerable, no potential candidates are polling particularly close at this time.
This week's Senate forecast update is the best one for Republicans all year. However, that isn't saying much, since I currently forecast Democrats to make a net a net gain of one seat, for an overall total of 61. Further, that is the forecast even if all of Representative Mike Castle, former Governor George Pataki and former Senator Jon Sununu all run in Delaware, New York, and New Hampshire respectively. If all three were to decline, then Democrats would be forecast to net 3-5 seats, for an overall total of 63-65.
In short, even after Charlie Crist announced in Florida, and Roy Cooper declined to run in North Carolina, Democrats still have the upper hand in the 2010 Senate picture. That might change, but only if Republicans can not only score the recruits I listed above, but also find top challengers in Colorado and Nevada. One other worrying point could be Arkansas, that is if you actually care whether Blanche Lincoln wins or not (I don't).
Here is the detailed forecast:
2010 Senate Outlook: Net Democrat Gain of one seat, for a total of 61 Republican-Held Seats Democratic Pickups: 3 (Missouri and Ohio, plus either Kentucky or New Hampshire)
(Note: Senate elections where there is no polling showing the incumbent party ahead by less than 10.0% are not examined in detail. Pennsylvania included due to expected popular demand.
*= Faces primary, but currently the heavy favorite against other announced candidates
** = There is no polling in Arkansas is not for Lincoln vs. Hendren, so previous polling on Lincoln versus other candidates is listed in its place.)
Notes on how to read the chart in the extended entry.
One ray of hope in the Florida Senate campaign is that the poll showing Charlie Crist ahead by 21% also showed him with 88% name recognition, compared to only 25% for Meek. It is possible that as Meek's numbers rise as he closes the name ID gap. Unfortunately, it is also possible that Crist will have much more money than Meek, and be able to define him in any potential advertising war.
Here is the latest Senate chart. With Florida removed from the equation, Democrats are now projected to gain only one seat:
2010 Senate Outlook: Net Democrat Gain of one seat, for a total of 61 Republican-Held Seats Democratic Pickups: 3 (Missouri, plus two of Kentucky, New Hampshire, North Carolina and Ohio)
(Note: Senate elections where there is no polling showing the incumbent party ahead by less than 10.0% are not examined in detail. Pennsylvania included due to expected popular demand.
*= Faces primary, but currently the heavy favorite against other announced candidates
** = There is no polling in Arkansas is not for Lincoln vs. Hendren, so previous polling on Lincoln versus other candidates is listed in its place.)
The only Republican incumbent who appears to be threatened is Richard Burr in North Carolina (although it is possible that David Vitter might find himself vulnerable, too). Depending on recruiting, Republicans might end up with more legitimate pickup opportunities than Democrats in 2010. Should Republicans score a "moderate" wave of former Governor George Pataki in New York, Representative (and former Governor) Mike Castle in Delaware, Representative Mark Kirk in Illinois (although that looks increasingly unlikely), and if former Representative Rob Simmons in Connecticut (he faces a primary and hasn't raised any money), then they will be projected to win seats in 2010.
Perhaps we are running up against the limits of how many Senate seats one party can have these days. Notes on how to read the chart in the extended entry.
Quite a bit has happened to the Senate picture since the most recent update back on April 13th. Here is the latest forecast, based entirely on polling information. The chart could change quite a bit depending on which candidates end up running:
2010 Senate Outlook: Net Democrat Gain of two seats, for a total of 62 Republican-Held Seats Democratic Pickups: 4 (Florida and Missouri, plus two of Kentucky, New Hampshire, North Carolina and Ohio)
(Note: Senate elections where there is no polling showing the incumbent party ahead by less than 10.0% are not examined in detail.
*= Faces primary, but currently the heavy favorite other other announced candidates
** = There is no polling in Arkansas is not for Lincoln vs. Hendren, so previous polling on Lincoln versus other candidates is listed in its place.
*** = Pennsylvania included despite not meeting the "less single-digit incumbent party poll lead" requirement due to the high level of interest and state of flux in the election there.
Details on how to read other aspects of the chart can be found in the extended entry.)
Key changes since the last forecast in the extended entry.
Five days ago, Open Left published its first Senate outlook for 2010. As with the 2006 and 2008 forecasts, it is entirely based upon polling. Since that forecast, I have been made aware of additional polling from Delaware, Missouri, North Carolina and Ohio.
In light of this new information, I have worked to clean up the forecast, and consolidate its information into a single, quick reference, easy-to-read (hopefully) chart.
2010 Senate Outlook Republican-Held Seats Democratic Pickups: 5 (Florida, Missouri, and New Hampshire, plus two of Kentucky, North Carolina, Ohio and Pennsylvania)
(Note: Senate elections where there is no polling showing the incumbent party ahead by less than 10.0% are not examined in detail. Details on how to read the chart can be found in the extended entry.)
Now, an electoral forecast, while interesting in and of itself, is not sufficient for the purposes of creating a more progressive governing majority. As such, for the 2010 electoral cycle, the Senate forecast will be combined with a progressive legislative forecast. That is, Democratic candidates will be rated not only by the closeness of their election, but also by what pieces of legislation their election would help produce.
Over the next two years, our electoral efforts should not be based on simply trying to elect as many Democrats as possible, but on trying to elect the right amount of a certain type of Democrat who will pass progressive legislation that failed during 2009 and 2010. So far in 2009, two progressive priorities have been all but defeated: bankruptcy "cramdown" legislation (passed by the House, but it will not be meaningfully addressed by the Senate) and the Employee Free Choice Act (sure-fire in the House, but at least two votes short in the Senate).
This means, for example, that Blanche Lincoln's seat in Arkansas is not a priority at all, given that she is opposed to both pieces of legislation. Whether she is re-elected or defeated, there would be no forward movement in the degree and amount of progressive legislation that is passed (as long as there is no overall, net decline in the number of Democrats in the Senate.) By contrast, the Florida and New Hampshire Senate campaigns, featuring Representatives Kendrick Meek and Paul Hodes, would be top priorities. This is because both have already voted in favor of cramdown and EFCA. Each of those two pieces of legislation would receive two more votes in the Senate with the election of those two members.
With this in mind, the next step in the Senate outlook / forecast is to develop a comprehensive list of where all Democratic candidates for the campaigns listed in these charts stand on progressive legislation that was, or will be, defeated in the 2009-2010 legislative calendar. Over the coming weeks and months, I will be slowly adding information on where all Democratic Senate candidates stand on defeated progressive legislation. With such a list, we will be able to track exactly which seats we need to win in order to pass better legislation, and thus better understand where we should target our electoral efforts. This will be far more useful than a mere Senate forecast because, in the end, this isn't about electing more Democrats, but about passing better legislation. We need to know where to spend resources in order to pass our legislative priorities.
How to read the charts can be found in the extended entry. Any suggestions on how to make the chart more comprehensible are welcomed in the comments.
Two days, ago, I compared the average rate of error from final predictions to final election results for Pollster.com, fivethirtyeight.com, and my own predictions. Looking at 65 elections on November 4th, 2008, where all three sites made public final predictions / estimates, it turned out that Pollster.com and fivethirtyeight.com were equally accurate, and that I lagged about 8-10% behind.
While I was a bit further behind, I still wanted to see where I was less accurate. It turns out that when blowouts (final margin over 20%) and rarely polled elections (only one poll in the final eight days) are removed, my simple, rudimentary methodology was actually the equal of Pollster and 538. As long as there were at least two polls in the final eight days, simple poll averaging was just as good at predicting election outcomes as any other methodology around. Data in the extended entry.
Now that all of the counting is finally done for the 2008 elections, it is possible to compare how different election forecasters fared. The three I have long been most interested in comparing are:
My method, which takes the simple mean of all non-campaign funded, telephone polls that were conducted entirely within the final eight days of a campaign. My rationale for this method is described here: No Special Sauce Needed For Electoral Projections. This is an intentionally rudimentary "election forecasting for dummies" method that anyone can reproduce.
Pollster.com, which uses all polls ever conducted in a state, and creates a regression line based on those polls. This is the ultimate "don't cherry pick polls and don't argue with polls" method. It was developed by a professional pollster and a political scientist, and is explained here.
Fivethirtyeight.com, whose complicated methodology is essentially the opposite of Pollster.com's: adjust every poll based on demographics, previous house effects, and previous error rate.
How did these three distinct prediction methods fare against each other? Results in the extended entry.
Since final results have not, as of yet, been certified by the fifty Secretaries of State, it is still too early to compare my final poll averages of my Presidential and Senate forecasts to the final results in those states. The initial estimate seems to be that polling averages performed very well in states where there were a lot of polls, but did not do well in the less frequently polled states of Alaska (President, Senate and House showed massive, double-digit error), Iowa (6% error), Nevada (6-7% error) and North Dakota (6% error). Polling averages also seem to have been about 4-5% off in Arizona and New Mexico. Everywhere else, the averages seem to have nailed the final targets by 2.0% or less, even though inaccurate winners were projected in Indiana and North Carolina. Basically, it seems like the more polls in your averages, the more accurate the averages become. Makes sense.
While we wait for final results in the polling average states, I am happy to say that my House Forecast has, once again, done extremely well. If, as appears likely, MD-01 and VA-05 go to Democrats, and with CA-04, CA-44, LA-04, OH-15, and WA-08 still undecided, then Democrats will net 21-26 seats. My final projection was 21-27 seats, so I am feeling like I did pretty darn well. I also did well for each of the category projections (more in the extended entry):
President Electoral Vote: Obama 338--200 McCain
National Popular Vote: Obama 53.1%--45.4% McCain
You can see my final percentage projections here. I decided to go with Obama in North Carolina even though the state was exactly tied. The reason is that most of Obama's vote is already in, while McCain still has to get his voters to the polls. That's enough of a tie-breaker for me. If a final North Carolina poll comes out showing McCain ahead by any margin at all, I reserve the right to change my forecast for the state (Update: ARG poll of the state shows Obama up 1% in NC. It functions as a tie-breaker. Update 2: Zogby final tracking poll moves 2 points to McCain, so that breaks the tie in the other direction). Oh--and the national popular vote is just a guess based on the Pollster.com national average. I'm only banking my methodology on the state results.
Senate
Democratic Pickups: Alaska, Colorado, New Hampshire, New Mexico, North Carolina, Oregon and Virginia.
Run-off in Georgia where we fight for the Employee Free Choice Act
Extremely narrow loss in Minnesota
If I have some free time tomorrow, I'll spruce up the final percentages, and post them here. More likely, I will finish them after the election, to test how well my methodology worked.
House Best Guess: Democratic Pickup of 24 seats, for a total of D 260-175 R in the House.
Guess Range: 21-27
Read the entire final forecast here
Well, that was a fun year of forecasting elections. What have you got for your final numbers?
* = The Libertarian candidate in the Georgia Senate race averages 4.5% across the six polls. If no candidate reaches 50% on Tuesday, there will be a run-off on December 2nd. Right now, the polling indicates a run-off is the most likely outcome.
Democratic Held, Uncompetitive Democratic Locks (12): Arkansas (Pryor), Delaware (Biden), Illinois (Durbin), Iowa (Harkin), Louisiana (Landrieu), Massachusetts (Kerry), Michigan (Levin), Montana (Baucus), Lautenberg (New Jersey), Rhode Island (Reed), South Dakota (Johnson), West Virginia (Rockerfeller)
***
I will continue to monitor incoming Senate polls, but unless one of these seats shifts noticeably, I do not anticipate making another Senate forecast. This is tentatively my final forecast: Democrats gain seven seats.
Methodology and analysis in the extended entry.
Update (11/4):: Final poll numbers added. Nothing changes, really.
