You Might Like
Polling Forecast: The Senate's Leaning Republican
On Tuesday, Americans will head to the polls to determine 36 seats in the U.S. Senate. Although those contests will likely draw as many as 49 million voters, it’s the outcome of only a small handful that will ultimately decide whether the Democrats or Republicans enjoy majority control of the Senate in 2015.
Currently, Democrats have a working 55-seat Senate majority, including support from two independents. Republicans need to pick up a net six seats to gain control, since Vice President Joe Biden would break a 50-50 tie in favor of the Democrats.
As of Sunday morning, the latest round of polling suggests that Republicans are on the verge of gaining eight seats, and perhaps more. The most recent HuffPost Pollster forecast gives Republicans a 74 percent chance to take control of the Senate.
This year’s election map leaves Democrats largely on the defense. While Republicans currently hold 15 of the seats up for grabs this cycle — only three of which look competitive — Democrats are fighting to hold the other 21 seats, which include a significant number in traditionally Republican states that were won during the left-leaning 2008 cycle.
According to The Huffington Post’s poll tracking models, which are based on all public polling, Republican candidates currently lead by healthy double-digit margins in three states where Democratic incumbents are stepping down this year: Montana, South Dakota and West Virginia.
These deficits leave nine more states that have been the focus of the 2014 campaign — the nine states that will effectively decide the outcome.
In three of these states — Alaska, Arkansas and Louisiana — incumbent Democratic senators currently trail Republican challengers by margins of 3 to 6 percentage points in the Pollster tracking models. While these deficits are not insurmountable, the real-world history of late shifts and polling error suggests that all three incumbents are likely headed for defeat.
Republican candidates are also leading by slim margins in Colorado and Iowa, two states now represented by Democratic senators that Obama carried in 2008. And Democratic incumbents are in danger in two more states, North Carolina and New Hampshire, though both of those senators hold tenuous leads going into the final days of the campaign.
With Democrats’ early hopes of unseating Sen. Mitch McConnell (R-Ky.) seemingly dashed, the loss of a Democratic majority would appear a near certainty if not for two states currently represented by Republicans, Kansas and Georgia, where the Republican candidates are fighting for survival and where the final polling margins have been close.
During this, the final weekend of the election campaign, the best hope for Democrats to retain their Senate majority lies along a narrow path. Democrats would need their incumbents to hold on in New Hampshire and North Carolina and to overcome relatively narrow polling deficits in Iowa and Colorado. Yet even in that circumstance, Democrats would still need one more seat to retain a majority. They could achieve this by picking off Georgia (or perhaps having one of their their incumbents overcome far longer odds in Alaska or Arkansas). Absent that, they would just have to hope that independent Greg Orman wins in Kansas and opts to caucus with the Democrats in January.
Here’s a closer look at what the polling data says about the nine states that hold the key to the outcome.
Alaska: Sullivan (R) vs. Begich (D, incumbent)
Sen. Mark Begich started off the election cycle with a modest lead in the HuffPost Pollster model. Since August, however, challenger Dan Sullivan has been consistently ahead, though never by more than 4 points. As of Sunday, two days before the election, he’s up by more than 3 points. The Pollster model translates that margin into a better than 60 percent chance that Sullivan will win.
The probability of a Republican gain in Alaska would be higher, but the difficulty in measuring public opinion in Alaska, a huge state with a relatively small number of voters, poses a unique challenge. Common methods of polling, such as by telephone and over the Internet, can be difficult in these areas. Many of the pollsters that do attempt to gather data there are sponsored by partisan parties, or are partisan themselves.
Polls give Sullivan the lead, but also suggest that Begich’s aggressive voter outreach may be paying dividends. This race could easily defy the predictions of a Republican victory — but don’t stay up waiting for results. Alaska is known to take a while to count.
Arkansas: Cotton (R) vs. Pryor (D, incumbent)
Most polls and forecasters have reached a consensus: Sen. Mark Pryor is in trouble. Individual polls on the race have gone back and forth, but have shown challenger Tom Cotton leading since late May. Even when Pryor led in the spring, it was never by much more than 2 percentage points, but Cotton’s lead is now about as high as 6 points. That gives him an 85 percent chance of winning, according to HuffPost’s model.
Of 14 polls conducted since September 1, five have shown Pryor leading, but four of those were sponsored by Democrats. Pryor’s chances depend on either a massive polling failure or a shift of virtually all of the undecided voters, neither of which seems especially likely.
Colorado: Gardner (R) vs. Udall (D, incumbent)
Sen. Mark Udall and challenger Cory Gardner spent much of this election cycle locked in a tie, until Gardner opened up a tiny lead in September. HuffPost’s poll tracking model for Colorado has shown Gardner’s edge over Udall slowly widening to about 2 points in the final days before the election, with about a 60 percent chance that Gardner will win.
Colorado has been something of a conundrum for pollsters in the past, though, and this year may be no exception. In 2010, what looked like a win for Republican candidate Ken Buck actually turned out to be a victory for Michael Bennet, the Democratic incumbent. Now, with mail-in ballots being used more widely in the state than ever before, and amid concerns about measuring the typically Democratic-voting Latino population, there’s some doubt about the accuracy of existing polls. Many Democrats insist Udall is performing better than the numbers show. Additionally, in three recent polls Udall has led by 1 point, although two of those polls were Democrat-sponsored. The bottom line: Latino turnout (and turnout in general) will be a big factor in Colorado. This one could go either way.
Georgia: Perdue (R) vs. Nunn (D) vs. Swafford (L)
David Perdue, a former CEO of the sportswear company Reebok, has held a steady lead in the HuffPost Pollster model. His edge has narrowed in recent weeks, however, and stands at slightly less than a 3-point advantage over Michelle Nunn as of this writing. Libertarian Amanda Swafford was pulling nearly 5 percent in earlier polls, but has since seen that figure decline.
The various election forecasts are split on this race, in part due to the possibility of a runoff if no candidate receives over 50 percent of the vote on Election Day. As of two days before the election, HuffPost Pollster finds a runoff about as likely as not, so there’s a solid chance that a candidate — most likely Perdue — could reach the 50 percent threshold and avoid a runoff. The Pollster model gives Perdue about a two-thirds chance of winning outright on Nov. 4 or winning the runoff. Additionally, the quality of polling in Georgia has come under fire, adding to the uncertainty around this race. Many of the pollsters active in Georgia are partisan-sponsored, and others rely on robocalls, which may increase the difficulty of measuring minority votes.
Keep an eye on the 50 percent mark as the votes are counted. Perdue probably has the best path to winning outright, but polling on a potential January runoff is sparse and few precedents exist to predict turnout. If the race goes to a runoff, it could effectively be anyone’s to lose.
Iowa: Ernst (R) vs. Braley (D)
In the final days of this race, which has often been neck and neck, Joni Ernst appears to have the lead over Bruce Braley, although many polls show a close margin. Two days out from the election, the HuffPost model gives Ernst a lead of just below 3 points, and about a 60 percent chance of winning.
Ernst’s probability of winning is relatively low since both candidates have taken the lead at various times, and many voters are still undecided. This is an open seat in a state where Democrats won fairly decisively in 2012, but with the drop in turnout that is typically seen from presidential to midterm elections, Republicans appear to have the advantage.
Kansas: Roberts (R, incumbent) vs. Orman (I)
Former Democratic candidate Chad Taylor threw a wrench into Sen. Pat Roberts’ re-election plans by dropping out of the race at the beginning of September. A fairly clear Roberts lead dissolved into who-knows-what as pollsters and forecasters scrambled to figure out what a two-way race between Roberts and independent Greg Orman would look like. We still don’t have much clarity: As of this writing, Roberts and Orman remain locked in a near tie. HuffPost Pollster’s model considers this race a true toss-up, with nether candidate having a better than 50-50 chance of winning based on the polling alone.
Given Taylor’s September surprise, there is less polling available than there normally would be. Taylor’s exit was complicated by debates over whether his name would still appear on the ballot. (It won’t.) And Orman isn’t saying which party he would caucus with, should he win — his statements that he would caucus with the majority party are not helpful when the Senate majority is itself up in the air. In this case, a coin flip may be as good as any other prediction.
Kentucky: McConnell (R, incumbent) vs. Grimes (D)
Democrats once hoped that Alison Lundergan Grimes might help save their majority with a strong challenge to Sen. Mitch McConnell, but the minority leader has never trailed. Grimes closed in on McConnell in the spring, but McConnell quickly regained his advantage. As of Sunday, according to the Pollster tracking model, McConnell enjoys a lead of better than 5 percentage points and a nearly 80 percent probability of winning.
This race has been closely watched throughout the election cycle. A McConnell loss would be the first defeat for a Senate minority leader in the general election. That doesn’t look likely at this point, however.
Louisiana: Cassidy (R) vs. Landrieu (D, incumbent) vs. several others
Tuesday’s election will essentially be a primary in Louisiana. Not only does a winner need over 50 percent of the vote to be elected, but the race is a so-called “jungle primary,” listing every candidate running for the seat — including multiple Republicans and Democrats. In this primary, incumbent Sen. Mary Landrieu has held a steady lead, but that won’t be enough to win re-election. In the runoff, HuffPost Pollster’s model shows challenger Bill Cassidy leading consistently. Based on the runoff win probabilities, we give Cassidy roughly a 70 percent chance of defeating Landrieu in December.
New Hampshire: Brown (R) vs. Shaheen (D, incumbent)
Up until mid-August, Sen. Jeanne Shaheen held a fairly strong lead in this race of at least 5 points, but her advantage has since significantly narrowed. The HuffPost Pollster model now shows Shaheen leading by just over 2 percentage points, with slightly greater than a 60 percent chance of defeating former Massachusetts Sen. Scott Brown.
It’s worth noting, however, that the only polls to show Brown with a lead in October were from New England College, whose automated polling has been criticized for its volatility. It’s also unclear how much of Brown’s gains have been from true opinion shifting. Some of his polling gains in recent months could be attributed to shifts in polls from registered voter samples to likely voter samples, which tend to move a few points in the Republican direction due to likely voter demographics. New Hampshire will likely keep Shaheen, but by a narrower margin than it once seemed she would.
North Carolina: Tillis (R) vs. Hagan (D, incumbent) vs. Haugh (L)
Sen. Kay Hagan’s path to re-election has been iffy from the start. According to HuffPost Pollster’s tracking model, she started 2014 up by only 1.4 points, and that lead disappeared by mid-March. After following narrowly behind challenger Thom Tillis for a few months, Hagan retook the lead in late May, and now holds an edge of roughly 1 percentage point, with a win probability on the Pollster model in the mid-50 percent range. (Earlier in the year, Libertarian candidate Sean Haugh took close to 10 percent in polls, seeming to draw more support from Tillis than Hagan, but his numbers have since sunk closer to 4 percent.)
There’s been a lot of polling in this race, as North Carolina is home to several academic pollsters as well as Public Policy Polling, the prolific Democratic-associated organization. Two of the last three nonpartisan polls have shown Hagan and Tillis tied. In this close race, Democrats are relying on their massive get-out-the-vote effort to help Hagan hold on. Barring a last-minute shift among the undecided voters, Hagan should have the edge.
www.huffingtonpost.com/2014/11/02/senate-forecast_n_6089424.html?utm_hp_ref=gay-voices&ir=Gay+Voices
You Might Like
