Sabato's Crystal Ball

2014: A Tale of Two Elections

New 2014 Senate and House ratings

Larry J. Sabato and Kyle Kondik, Sabato's Crystal Ball October 16th, 2014

facebooktwittergoogle_plusredditpinterestmail

As we approach the home stretch, 2014 has turned into a tale of two elections. On the one hand, this is a classic sixth-year itch election where the incumbent president’s party is going to suffer losses in both houses of Congress. We’re just arguing about exactly how many. Overall, it is indisputable that Republicans will have more critical victories to celebrate than Democrats when all the ballots are counted, and they have a strong and increasing chance to control the next Senate.

On the other hand, there are unusual and even a few bizarre features on the landscape. Some Democratic incumbent senators have been hanging tough in heavily Republican territory; months ago, many observers thought they’d be quickly swept out to sea in a red tide. The GOP is having a difficult time making inroads in competitive “purple” states and districts, and very little progress at all has been seen in Democratic blue areas. Contests where Republicans should have been dominant (Alaska, Georgia, Kansas, and South Dakota) have teeter-tottered back and forth, necessitating more competitive rating changes and, in some cases, even becoming Toss-ups — or worse.

The “hot” trend, though it may fizzle in the end, is the emergence of independent candidates and fusion tickets to compete aggressively in deeply Republican territory where a fully labelled Democrat would have a devil of a time winning. Many people are enjoying seeing the incumbents and majority party squirm in places like Alaska, Kansas, and South Dakota, but the tug of party loyalty increases as Election Day approaches, and so the pleasure may be short lived.

Senate: Purple, red-lite, or redder?

Democrats know they will lose net seats, and they hope somehow the math works out to a narrow 51-49 or a 50-seat “Biden majority.” Stranger things have happened, but almost all the lucky breaks would have to go blue.

While Democrats have the one-up bonus of just needing 50 seats for a Senate majority, Republicans have many more practical pathways to 51. We’ve compared it in a general way to 2012, when President Obama had many more ways to accumulate 270 electoral votes than did Mitt Romney. In the zero-sum game of politics, maybe it is payback time.

But Republicans aren’t there yet, and the hardest 19 days of the election stretch before us. Early miscalculations, shortcomings, and gaffes can be corrected or forgotten; late ones can be fatal in a heated battle.

For now, the Crystal Ball is right where we have been for a long time — thinking that the math and the map look likely to produce a modest Republican Senate majority of one to three seats. Exactly which seats could constitute this majority remains unsettled.

At this point everyone has consigned West Virginia and Montana to the GOP gains column, yet we’d all thought South Dakota was in that category, too. Former Gov. Mike Rounds (R) is still the likely winner, but there will be no landslide in a three-way race that Rounds appears to be leading with less than 40% of the vote. Assuming Rounds finally brings his “A” game, the GOP will be up +3. Arkansas is probably the fourth state to fall, with Rep. Tom Cotton (R) still favored to beat Sen. Mark Pryor (D). The fifth seat might well be Alaska, where Republican Dan Sullivan has had a consistent (of late) lead over freshman Sen. Mark Begich (D). This is a rating change for the Crystal Ball: We are moving Alaska from Toss-up to Leans Republican. The Begich forces have to hope their unprecedented organizing efforts prove the polls wrong.

It’s the sixth seat that proves elusive. If Louisiana’s election were on Nov. 4, we’d likely designate it the seat that makes the GOP majority. But as we all know, Louisiana likes to do things differently, so its likely runoff won’t be held until Dec. 6. Sen. Mary Landrieu (D) is in trouble, for sure, and Rep. Bill Cassidy (R) looks well positioned to dethrone her — but a lot can change from Nov. 4 to Dec. 6. The appropriate, cautious designation for the Bayou State is still Toss-up, but we’re adding a little twist this week: Because we do not expect either Landrieu or Cassidy to get at least a majority of the vote in the all-party primary Louisiana holds on Election Day, we’re calling the race “Toss-up/Leans Runoff.”

Cassidy would likely start the runoff as a favorite, but, again, the circumstances under which that standalone election would be fought are unknown: Would that seat make the majority for either side, or will it just be icing on the cake for one party or the other? And will it even represent the end of the national Senate campaign? Perhaps not. Hold that thought.

The sixth GOP seat might also be provided by Iowa or Colorado. Two weeks ago we moved the Hawkeye State contest from Toss-up to leaning toward GOP nominee Joni Ernst. She led Rep. Bruce Braley (D) by six points in Des Moines Register and Quinnipiac University surveys and had shown other signs of progress.

Just last weekend, yet another DMR poll found her ahead by only a statistically insignificant one point, and Quinnipiac found her up just two on Wednesday morning. We doubt she dropped so far or that Braley moved up so quickly, so the first DMR poll and the Quinnipiac poll were probably just too favorable to her. That said, Ernst is still slightly leading the polling averages after USA Today/Suffolk University found her up four points Wednesday afternoon. The rating here remains a very fragile Leans Republican.

We’re gun-shy on Colorado because public polls have badly undershot Hispanic turnout and thus Democratic performance in both 2010 and 2012. Most notably, in the state’s 2010 Senate race, now-Sen Michael Bennet (D) led exactly one of the contest’s final 18 polls, but wound up defeating Ken Buck (R) on Election Day. Rep. Cory Gardner (R) has been moving up on Sen. Mark Udall (D), assisted by an unexpected endorsement from the Denver Post, which backed President Obama twice, Bennet in 2010, and Udall himself six years ago. For what it’s worth, Gardner’s lead in the polling averages is nearly three points in HuffPost Pollster and two in RealClearPolitics, and he probably needs a slightly larger cushion to be regarded as the likely winner. After all, in 2010 Buck was ahead by 1.4 in the former and three in the latter on Election Day, but still lost by 1.7 points.

Democrats have quite a number of bright spots, headed by North Carolina. While a late GOP surge could wipe out her slight lead in the polling averages, Sen. Kay Hagan (D) has fairly consistently blocked state House Speaker Thom Tillis (R) from gaining ground in the highly competitive Tar Heel State. Hagan will have to keep up the pressure lest the party tilt of 2014 swamp her ship on Nov. 4. The National Republican Senatorial Committee just dropped $6.5 million into the contest, hoping that Tillis can erase Hagan’s small but persistent lead. (It’s worth nothing, though, that an investment this late in the game buys fewer ads than one made months in advance because of pricier advertising rates.) This state remains a tenuous Leans Democratic hold.

Other Democratic states that might have been more competitive, such as Michigan, Minnesota, and Oregon, have never developed for the GOP. Sen. Jeanne Shaheen (D) is still holding off former Massachusetts Sen. Scott Brown (R) in New Hampshire: She is not completely out of danger in the volatile Granite State but the current advantage is hers. In Virginia, Sen. Mark Warner (D) has been hit with a nasty story involving the possible promise of a judicial post for a relative of a state senator to keep the official from resigning, and we wouldn’t be surprised to see this contest decided by single digits. Virginia, after all, is a highly competitive state. Yet Warner is well established and a victory by Republican Ed Gillespie would be a massive upset.

What of the endangered GOP seats?

Sen. Mitch McConnell (R) still maintains his polling edge over Alison Lundergan Grimes (D) in Kentucky, and the Democratic Senatorial Campaign Committee is no longer advertising there, which is a perhaps telling sign about the current state of the race.

Sen. Pat Roberts (R) of Kansas may be making a comeback after having been left for dead on the battlefield. His independent (and Democratic supported) challenger Greg Orman is learning how difficult it is to best even a weak, damaged Republican in ruby red Jayhawk territory. We’re moving this contest from Leans Independent back to Toss-up in light of some more favorable numbers for Roberts in recent days.

Money the DSCC might have spent in Kentucky now appears to be going to Georgia, which is good news for Michelle Nunn (D) in her challenging battle against David Perdue (R). Perdue has led most public polls by about three to four points — though Nunn led by three in a new SurveyUSA poll Wednesday — but Perdue needs to get to 50% to avoid a runoff. Right now, he’s stuck between 46-47% in the poll averages. Perdue also apparently has been hurt by comments from several years ago about outsourcing jobs.

Democrats and Republicans still seem to be holding out hope that they can get their respective candidate over 50% on Election Day (Perdue still has the better shot). But we’re not so sure either will make it. So we’re giving this race the same designation we have in Louisiana: Toss-up/Leans Runoff.

By the way, the runoff would be on Jan. 6, 2015, a month after the possible Louisiana overtime and three days after the technical start of the 114th Congress. Pardon us, but Georgia’s runoff law is just this side of insane.

Our projection remains a five-to-eight seat Republican gain in the Senate, and with less than three weeks to go we would much rather be holding the cards Republicans have been dealt versus the ones dealt to the Democrats as both sides play for a Senate majority. Despite the likelihood of two runoffs, it’s not impossible to imagine the GOP having a good enough night that they get to 51 seats without Georgia or Louisiana: That would mean holding all of their present seats (minus Georgia, at least on Election Day) and capturing (in order of least to most GOP difficulty) Montana, West Virginia, South Dakota, Arkansas, Alaska, Iowa, and Colorado. That would represent a great night for the GOP, and they have at least a decent chance in all seven races (North Carolina could be the eighth).

That said, a month ago the Republican position looked much stronger in Georgia and Kansas (states they already hold), as well as South Dakota (a seat that until recently looked like an easy pick-up). Republicans could and probably should win all three, even if it takes them until January to do so in the Peach State. But 2014 has been a crazy enough election that one of these seats could possibly slip through the GOP’s fingers.

Table 1: Crystal Ball Senate ratings changes

Map 1: Crystal Ball Senate ratings

A contracting House map?

The big news over the past week in the House has been the game of “triage” being played by the Democratic Congressional Campaign Committee. Basically, this is when a committee cancels ad reservations in certain districts late in the campaign and moves that money to other races in order to play defense.

The Democrats have recently canceled ad spending in some Republican-held districts they hoped to win earlier in the cycle. Most notably, they cut $2.8 million from VA-10, an open Northern Virginia seat being vacated by retiring Rep. Frank Wolf (R). They moved some of that money to CA-7, the Sacramento-area seat of Rep. Ami Bera (D), who is locked in a close race with former Rep. Doug Ose (R) and who has been the target of a few million bucks’ worth of recent ad spending from outside conservative groups.

Without going into painstaking detail on all of these shifting ad moves, the takeaway for us is that the House map isn’t really expanding to a great degree: New ad reservations by both parties appear to almost exclusively be going to contests that that we have long seen as highly competitive, races that we list as Toss-ups or as leaning to one party or the other in our Crystal Ball House ratings.

In fact, instead of expanding, the map is slightly contracting as Democrats focus on protecting incumbents instead of pursuing increasingly untenable offensive opportunities in Republican-held seats. That gives us further confidence that the Republicans will in fact add more seats to their majority, but probably not significantly more. Our outlook remains a six-to-nine seat Republican net gain. An eight-seat addition would get the GOP to 242 seats, which would match their 2010 performance; a 13-seat gain, something we do not expect at this time but which would not be impossible, would push the GOP to 247 seats, or their biggest majority since the one they won in the 1928 election.

House ratings changes

Now we just noted that, for the most part, the overall House map isn’t changing much: The races we thought would be the most competitive generally remain so, particularly the ones currently held by Democrats. In the Republican-held districts, a number of Democratic challenges have flamed out over the past several months, which isn’t a surprise in a Republican-leaning year.

Moving from Likely Republican to Safe Republican this week are seven such seats: Reps. Jeff Denham (CA-10), Steve King (IA-4), John Kline (MN-2), Steve Pearce (NM-2), and Joe Heck (NV-3), as well as open seats PA-6 and WI-6. A combination of underwhelming Democratic candidates (CA-10, NV-3, and PA-6 in particular) and a too-tough environment help explain why these seats should stay Republican. It’s worth noting that liberal talk show host Bill Maher could’ve picked many better targets for his “Flip a District” campaign than Kline.

Democrats also appear resigned to GOP wins in three upstate New York districts: Reps. Chris Gibson (NY-19) and Tom Reed (NY-23), as well as the open NY-21, which would be a Republican takeover. All three shift from Leans Republican to Likely Republican.

While there hasn’t been a major erosion in Democratic prospects across the country — this is not 2010 — a couple of late-breaking districts are looking worrisome for the Democrats. Joining the Toss-up column this week is the open seat in ME-2, where state Sen. Emily Cain (D) faces off against ex-state Treasurer Bruce Poliquin (R). Both sides are engaged here, and both sides believe the race is close. This is a district where President Obama got 53% of the vote, but Rep. Mike Michaud (D, ME-2), who is vacating the seat to run for governor, is a Blue Dog (moderate) Democrat who is probably a better fit for the district than Cain, who has a more liberal profile. Additionally, Poliquin probably has better name ID than Cain. That said, it’d still be a mild surprise if this district swung to the Republicans.

Also, the Democratic Congressional Campaign Committee recently announced a $600,000 ad buy in IA-1, an open seat being vacated by Rep. Bruce Braley (D), who is running for Senate. Obama got 56% here in 2012, but this is 2014, and state Rep. Pat Murphy (D) needs help against Rod Blum (R), a software company owner. Braley himself only won by two points in 2010 in this district, which changed slightly in post-2010 redistricting.

Finally, it would be a total shock if either Reps. Lynn Jenkins (R, KS-2) or Kevin Yoder (R, KS-3) lost. But we’re hearing that underfunded Democratic challengers to both are actually within striking distance, and the uncertainty in Kansas’ Senate and gubernatorial races have both House incumbents on alert. Jenkins, long-time House watchers will recall, was one of the few Republican bright spots in 2008: She beat freshman Rep. Nancy Boyda (D) after Boyda had pulled a massive upset in the Topeka-based seat two years earlier. If anyone outside of Kansas knows about Yoder, it’s probably because of his skinny-dipping in the Sea of Galilee during a 2011 congressional fact-finding trip to Israel. Yoder’s antics inspired one of this cycle’s funniest ads, from challenger Kelly Kultala (D). It seems preposterous to think Democrats could win House seats in Kansas in a year like 2014, but plenty of preposterous things seem possible in the Sunflower State this year: Both races come on to the board as Likely Republican.

Overall, we now list 232 races at least leaning to the Republicans, 189 at least leaning to the Democrats, and 14 Toss-ups. We list just 31 races in the most competitive Toss-up or Leans categories, and most of those — 21 of 31 — are currently held by Democrats.

Of the 14 Toss-ups, three are held by Republicans, and 11 by Democrats. However, it’s not at all clear that the GOP has an obvious lead in any of those 14 races. And some of the Republicans’ best targets are not necessarily moving their way. For instance, Reps. Ron Barber (D, AZ-2) and Rick Nolan (D, MN-8) have been in deep trouble at various points in the cycle, but now it looks like both have better-than-even odds to win (though we are not leaning either race to the Democrats at this point).

The same is true of some of the other Toss-ups, although Republicans will still probably end up winning a fair number of these races when it’s all said and done. That it is difficult to figure out which ones those might be, though, is as good of indication as any that GOP House gains might be modest.

Table 2: Crystal Ball House ratings changes