Sabato's Crystal Ball

Senate 2014: Roberts Slips in Kansas — And What’s Up in South Dakota?

Plus a slight upgrade for GOP House prospects

Kyle Kondik, Managing Editor, Sabato's Crystal Ball October 9th, 2014

facebooktwittergoogle_plusredditpinterestmail

Come January, Americans may witness something that, up to now, only 6% of the country’s population has ever seen: a senator from Kansas who is not a member of the Republican Party.

That’s just one reason why the Sunflower State’s Senate race is the most interesting in the country with a month to go.

Kansas last elected a Democratic senator in 1932, George McGill, who was defeated for reelection in 1938 (President Roosevelt’s “sixth-year itch” midterm, when Democrats lost 71 seats in the House and six in the Senate). So only Americans who are 75 and over — a little more than 1/20th of the nation’s population — have been alive to see a non-Republican senator in one of the nation’s consistently bedrock Republican states.

Kansas won’t be electing a Democrat to the Senate this year, because there is no Democrat on the ballot. Rather, independent businessman Greg Orman is challenging — and leading — unpopular incumbent Sen. Pat Roberts (R).

Roberts won a weak victory over his controversial primary opponent, physician Milton Wolf, in early August. Instead of pushing aggressively ahead in his general election fight, it’s now clear that Roberts expected to coast to victory against Orman and then-Democratic candidate Chad Taylor in a three-way race. But Taylor withdrew from the contest after his party’s prodding, leaving Democrats — and anti-Roberts independents and Republicans — with Orman as their main alternative to the incumbent (Libertarian Randall Batson is also on the ballot).

The polling here, both public and private, is starting to pile up, and it is now clear to us that Orman is indeed leading by somewhere in the vicinity of five points, which is about his lead in the polling averages. At this point, we have to look at him as at least a small favorite with 26 days to go until Election Day: So we’re moving Kansas from Toss-up to Leans Independent. Combined with Gov. Sam Brownback’s (R) precarious position in the governor’s race, this may well be a historic Year of the Great Revolt in Kansas, as we suggested last week.

If he won, Orman would be the third member of the Senate elected as an independent: Sens. Angus King of Maine and Bernie Sanders of Vermont are the others. During the 2012 cycle, our final rating in King’s race was “Likely Democratic/Independent” because we believed King would ultimately caucus with the Democrats. He did, although it’s theoretically possible he may switch to the Republicans at some point, particularly if the GOP wins the Senate this year.

With Orman, the rating is strictly Leans Independent, because we honestly do not know with which major party he would caucus. He could determine the majority if, at the opening of the 114th Congress, there are 50 Republicans, 49 Democrats, and Orman. This specific outcome is unlikely but certainly possible. Given the heat he’s taking from the GOP, perhaps Orman would pick the Democrats. Or maybe he’d be looking way ahead to reelection in 2020, realizing he’d have a better chance as an Independent Republican. But who knows? The point is that there is 100% certainty that a reelected Roberts would vote for a Republican majority leader, while Orman is a giant question mark. As a smart operative told us this week, Kansas voters seem to have decided to fire Roberts, but they do not yet know if they want to hire Orman. This is one campaign that truly matters, and the outcome is thoroughly unpredictable.

This is the third straight week that we have made a fairly aggressive ratings change in the contests for the Senate. Two weeks ago, we moved Sen. Kay Hagan (D-NC) from Toss-up to Leans Democratic because of her consistent polling lead. Last week, we moved the open Iowa Senate race from Toss-up to Leans Republican because we perceived a small edge for state Sen. Joni Ernst (R). And this week we’re moving Kansas to Leans Independent because of Orman’s clear lead and Roberts’ directionless campaign (though that may be changing). Many others still view these races as Toss-ups, and perhaps they are right.

That said, the Tar Heel State race has hardly moved at all in two weeks, so we still feel confident about our move there. The hour is growing late for Thom Tillis (R) to make his move against Hagan. We are less certain in Iowa, a race that very well may legitimately be a tie but where we probably would still rather be Ernst than her opponent, Rep. Bruce Braley (D). And while Orman is leading right now, his support might be soft, making him susceptible to Republican advertising painting him as a crypto-Democrat and President Obama backer. Kansas, after all, is a very Republican state, and GOP advertising will point out, subliminally, that Orman and Obama both start with “O” and have the same number of letters…

Our big picture outlook remains a Republican gain of five to eight seats in the Senate, and we believe that (with one exception) the playing field is narrowing, not growing. Republicans remain positioned to win the majority, but they are not certain to do so, and any majority they do grab is likely to be fairly narrow.

The exception alluded to above is South Dakota. Could the Mount Rushmore State, which analysts have long counted as in the bag for Republicans, be turning into another migraine for the GOP? It has long been thought that ex-Gov. Mike Rounds (R) had a punched ticket to the Senate, but he has steadily drifted down in polling averages, and incredibly he is now below 40% (38.7% in RealClearPolitics, 37.2% in HuffPost Pollster). Rounds is being bedeviled by his gubernatorial administration’s involvement in the state’s EB-5 visa scandal. As governor, Rounds was a strong proponent of the program, which gives visas to immigrants who invest at least $500,000 in areas that are rural or suffer from high unemployment. However, a torrent of negative press is dragging down Rounds, with reports of mismanagement of the program by his appointees, a bankruptcy filing by the program’s largest investment, and even the high-profile suicide of a former Rounds’ appointee who faced criminal indictment for allegedly pocketing money from a state grant.

Beyond the EB-5 visa affair, some Tea Party elements remain alienated from Rounds because of his refusal to sign the Americans for Tax Reform pledge to not raise taxes and because of some budget decisions back when he was governor. Rounds got about 55% of the vote in his primary, far outpacing any of his weak competitors but not exactly consolidating his base. What may save Rounds is his split opposition. Independent former Sen. Larry Pressler, who served as a Republican in the Senate from 1979 until his defeat in 1996, has moved up rapidly, though he still is in third place behind Rick Weiland (D) in polling averages. (A fourth candidate, independent conservative Gordon Howie, is cutting into Rounds’ support.)

While Pressler hasn’t said who he will caucus with if he were to win, it’s clear that a closely divided Senate with Orman and Pressler (and maybe King, too?) as the key swing votes would be a thing of magnificent chaos.

The Democratic Senatorial Campaign Committee — which had seemingly written off this race — entered the race with guns blazing on Wednesday: Bloomberg reported that the DSCC will put $1 million in South Dakota in the final weeks of the campaign, mostly on television advertising to attack Rounds with the hope that Weiland or Pressler — who endorsed Obama in 2008 and 2012 — will prevail and then caucus with the Democrats. It also remains possible that one or the other will stop campaigning and endorse the other, which would really put Rounds in a bind.

The intervention by the Democrats here means the race is getting serious, and polls confirm Rounds’ weakness. We’re moving this race from Likely Republican to Leans Republican.

Table 1: Crystal Ball Senate ratings change

Map 1: Crystal Ball Senate ratings

A slight uptick for House Republicans

Alan Abramowitz, Crystal Ball senior columnist and Emory University professor, uses a U.S. House model to try to predict the net change of seats in each election. The model uses a set of simple factors, as described in the Crystal Ball earlier this cycle:

The midterm forecasting model predicts the change in Republican House seats based on three factors — the number of seats held by Republicans in the current House, the margin by which the Republican candidate won or lost the previous presidential election and the Republican margin in the generic ballot in early September.

Given that one knows the presidential margin and the number of current GOP seats at the start of the cycle, the only variable that changes during the election cycle is the House generic ballot, a national poll that asks voters whether they will support a Republican or a Democratic candidate in their local House race.

As of Wednesday afternoon, Republicans held a 0.5 percentage point generic ballot edge in the HuffPost Pollster average, and a 2.1 point lead in the RealClearPolitics average. That translates to Republicans adding about six to nine seats to their majority.

Coincidentally, we took a fresh look at our House ratings this week, and we arrived at precisely the same evaluation using our district-by-district analysis. So we’re making a slight change to our House outlook, moving it from a projected Republican gain of five to eight seats to a projected gain of six to nine.

This tiny shift is emblematic of this year’s largely standstill battle for the House.

We have a few House ratings change to make this week.

Table 2: Crystal Ball House ratings changes

Mostly, these changes just reflect some GOP seats moving off the competitive board as Democrats focus on playing defense. For instance, there seems little chance that two Ohio Republicans, Reps. Bill Johnson (R, OH-6) and David Joyce (R, OH-14), will fall to Democrats this year, so we’re moving both from Likely Republican to Safe Republican. Ohio now lacks a single truly competitive House race this year, and the governor’s race has gone dark after the implosion of Ed FitzGerald (D), the hapless challenger to Gov. John Kasich (R).

Moving from Leans Republican to Likely Republican are Reps. Dan Benishek (R, MI-1) and Tim Walberg (R, MI-7). Walberg is a bit conservative for this swingy district and has underperformed in the past — he won the seat in 2006, lost it in 2008, and then regained it in 2010 — but it’s just hard to see him losing in a year like this. Democrats appear to agree: The Democratic Congressional Campaign Committee recently cut its ad buy in the Detroit market, which covers about a third of the population in this district. Benishek, a 2010 victor who surprisingly held on to his Upper Peninsula seat in 2012, also appears to be moving into the clear: The DCCC on Wednesday cut its ad buy for the district. House Majority PAC, which supports Democratic House efforts, also recently cut ad buys in both districts.

Meanwhile, we’ve heard whispers that Rep. Bill Keating (D, MA-9) is sweating bullets over his reelection bid against attorney John Chapman (R). Keating should still be OK in this district where President Obama won with 55% of the vote, although it’s worth noting that Republican Senate candidates won it in 2012 and 2013. In the late stages of the national House campaign, some surprising upsets can occur, and so we’re keeping an eye on this seat, which shifts from Safe Democratic to Likely Democratic. (The Boston Globe took an expansive look at the race earlier this week.)

Looking at the ratings as a whole, we have 232 seats that are at least leaning to the Republicans, which is just two short of the 234 with which they started the cycle. On the Democratic side, we have 190 seats at least leaning to them, 11 short of the 201 with which they started the cycle.

That leaves 13 Toss-ups. It seems reasonable at this point, based on where we see those 13 races falling, that Republicans should be able to win about two-thirds of them, or nine of 13. That would put them at 241 seats, a seven-seat addition that would be almost identical to the 242 seats they held after the 2010 wave. However, there’s almost a month to go, and some House races we don’t see as Toss-ups right now may actually end up flipping parties. That’s probably good for Republicans because the anti-White House environment, which is typical for a midterm, suggests that late movement will probably be movement to the GOP.

However, we still see Republican gains being limited to single digits — or very low double digits — barring a late surge.