NC Senate Remains Razor-Tight; Control of Senate Likely At Stake

Published by:

At the moment, I would say the two states that are most determining who controls the next Senate body are Alaska and North Carolina. I’ve had both states as close Toss-Ups all year. In both instances, I’ve had Begich(D) and Hagan(D) often ahead by the tiniest of margins, but neither has yet to nearly approach the magical 50% number. Thus, I remain quite cautious about Democrat leads in either state as both could easily come down to a handful of votes. While Louisiana, Arkansas, or Colorado could host some surprises for either side, the races in Alaska and North Carolina still seem to be the decisive front line. For now, I still have Dems clinging to 50 seats, though I could easily see that number drop to 48 if they don’t pull out both NC and Alaska.

Today, for the third straight model, Hagan(D) is clinging to a lead between 0.5% and 1.0%. Thus, the race is close, but it also appears pretty stable. That may imply a lack of movement down the home stretch. Still, because both candidates are quite unpopular, there is a lot of play left here. Today, Hagan hits 46.0% in the model to Tillis’s 45.1%. While I think the Libertarian candidate may pull a few percent, I highly doubt he’ll pull 8-9%, meaning that the remaining voters have to go somewhere. Even as the election nears, there’s a large swath of non-committal voters, especially in a race that has already spent millions in defining ads. I have the race tied in one model; Hagan’s biggest lead in any line is 1.6%, hence the continually tight overall model. While I think it’s fair to say Hagan probably has the slight edge here, this is a race that could still easily go either way and may truly come down to turnout. We’ll have to see how September goes once Hagan unloads the rest of her huge cash war chest, but, so far, no one is budging in this race.

NORTH CAROLINA SENATE MODEL: Hagan(D) 46.0% – Tillis(R) 45.1%

Prediction: Toss-Up

Mary Burke(D) As Close As Ever in Wisconsin Governor Race

Published by:

Mary Burke has cut Scott Walker’s already tiny lead in half in today’s Wisconsin Governor model. In the few weeks since I last modeled that race, Scott Walker’s lead has declined from 2.3% to 1.1% today. As I always say, the most important thing in modeling is looking at the trends. While the change has been incremental, Mary Burke has been slowly closing on Scott Walker all summer in the model, and, as of today, it has yet to stop. In three of the four model lines, I have Burke now down by 1.0% on the nose. The fascinating thing in this race is that, with over two months left to go before the election, both sides are strongly dug in. Scott Walker is sitting at about 48% in three of the model lines while Burke is at about 47% in three of the model lines. Overall, the “election day” model moves Walker(R) up to 49.2% with Burke(D) at 48.1%, well within margins of error, especially in a Democrat leaning state. Two things appear to be happening: 1) Scott Walker remains a polarizing figure, and 2) Mary Burke has so far done a good job at becoming a viable alternative to him. She’s consolidated the agitated Democrat-base in that state. Now the question is whether she can hold onto this momentum as the final campaign kicks into full gear and pick up those few undecided voters that are left in this race. With numbers like this, barring some gaffes or scandals, this race is one of those where saying it comes down to Get Out the Vote Efforts is a fact as opposed to a cliche.

WISCONSIN GOVERNOR MODEL: Walker(R) 49.2% – Burke(D) 48.1%


Georgia Senate Model “Breaks;” Race Moves Back to Leans GOP

Published by:

Well, that happened fast. I just wrote yesterday that the Georgia Senate model was trending toward Perdue(R), even if you weren’t seeing it in the top line number yet. As I often write, the model figures tend to be conservative, adapting slowly to dramatic changes in a race. When I say “conservative,” I don’t mean politically or ideologically, I mean statistically. In other words, in the Todd Akin race last cycle, for example, it took a few data points to show how dramatic the change in that race was. So, if you lean to the impatient side, that’s the downside of modeling. The upside is, the models have proven pretty accurate in the end and also don’t overreact to momentary blips in a race. And so, officially today, the Georgia Senate model has “broken.” Again, for those who don’t follow the blog regularly, I use the term model “break” when the central measures of the data points make a pretty big jump in one model run. When that happens, it tends to indicate a new reality of the race.

So, why has the Georgia race broken today? Ever since the contentious GOP primary run-off, David Perdue has been riding a wave of support that would indicate a new dynamic to that race. The negative Kingston ads have stopped, and it’s become a Democrat vs. Republican contest as opposed to a Democrat and Republican vs. Republican contest. As such, I have Perdue moving from less than a 1% margin yesterday to a 3.1% margin today. A 2.1% shift in one model, especially with this much data, is thus what I call a model “break.” I have Nunn now behind in all four model lines. She’s still as close as 2% in two of the model lines but down as much as 6% in another. So, I’m currently looking at a 2-6% Perdue range, coming in at that 3.1% margin overall. Thus, I’m moving the race back to “Leans GOP” today. That reduces the number of my Toss-Up races to five and pushes the GOP total up to 49 with the Democrats still stuck at 46 with leans implemented.

Where does this race go from here? Nunn is certainly still in it, though her window appears to be quite narrow. For the first time today I have a candidate in this race hitting 50.0% in the model. Nunn has some things in her favor: 1) she’s been mostly on the sidelines during the GOP primary process, 2) her campaign will have plenty of money to make its case post-Labor Day, 3) she’s still getting really positive things to happen, such as a recent endorsement by Zell Miller.  So, just because of my change in the rating today, this race is still very much worth watching.

In other news, for the second model run in a row, I have Gov. Deal(R) holding onto a narrow 2.0% lead in the Governor’s race over Jason Carter. That race remains a Toss-Up.

GEORGIA SENATE MODEL: Perdue(R) 50.0% to Nunn(D) 46.9%

RACE RATING: Moves from “Toss-Up” to “Leans GOP”

Georgia Gov and Senate Holding as Toss-Up Races

Published by:

A few weeks closer to the election and it still looks like Georgia may have two knock-down, drag-out fights on its hands. That may be a good thing for Democrats as two competitive races may help turn out voters in an off-year election.

Two weeks ago, Gov. Deal(R) had a tiny 1.5% lead over Jason Carter. The model today shows a similar race as Deal is up 2.0%, 49.2% to Carter’s 47.2%. In the Senate race, David Perdue(R) was up 0.9% last model and is now up 0.8%, virtually unchanged. A few notes though on where these races may be heading-

There is still a lack of data here, post-Senate GOP primary run-off. The data that has come out post-primary has generally shown a move toward Perdue. It’s too early to tell (still) whether that race is going to continue to move to his direction or whether we’re still living in a land of data showing what is essentially a post-primary bounce. Remember, Perdue has gotten plenty of free media during July as well as swamping the airways in paid media in his effort to defeat Jack Kingston. Of note here is Michelle Nunn’s rather large war chest. Excluding outside money, Nunn is going to have plenty of money to make her case down the stretch. Once that money kicks in full swing post-Labor Day, we’ll see where this race heads. Out of the four model lines, I still have Nunn ahead in one and Perdue ahead in three. Perdue’s leads are as high as 2% while Nunn is up only 0.5% in her most positive line.

The data in the Governor’s race is moving the other way. While the Senate race data has slowly been tilting GOP, Jason Carter appears to be holding his own or even advancing in some of the data. So, why the jostle back down to 2.0% deficit? Because, as always, do NOT over-read model to model shifts. Instead, look at the overall trends. A month or two ago, this race leaned GOP with Deal margins in the mid-single digits. Now, this race is playing in the 1-3% Deal range. If you continue to see margins like this, it’s a full indication that Carter really is hanging in there. I do still have Deal ahead by as much 4% in one model line. Carter’s best model line has him at an exact tie.

Stay tuned to these races- they’re pretty fascinating.


Iowa Senate Continues To Be One of Closest in Country

Published by:

Out of the Toss-Up races, Iowa and Colorado are the two that really ‘should’ be leaning toward Democrats by this point. But, two weeks closer to the election since my last model update in Iowa and that race remains quite close. The good news for Democrats? I have Braley(D) up in all four model prongs. The bad news? All prongs of the model show his lead between one and two percent. Those are the makings of a close race.

So, what’s going on here? Iowa is a slightly blue state, but it’s not ‘that’ blue. Throw in an off year election when turnout is likely to be low, an unpopular President, and a Democratic candidate who has made several gaffes, and Braley has not yet been able to pull away. Ernst’s strength, meanwhile, is also her weakness. Democrats view Ernst’s far-right comments and positions to be baffling, asserting that there is no way a candidate like that could win in a purple to slightly blue state. Yet, those positions and comments are very popular with her base, especially in the conservative Western parts of the state. You have to remember that Congressman Steve King represents 1/4th of this state. Thus, what you’re seeing is that Ernst has shored up her base while Braley has some work to do convincing moderates who are leery of Ernst.

The result is, in today’s model, Joni Ernst is at 47.0% while Braley is at 48.4%. If Braley is able to break into the slightly favorable fundamentals for him in this race, that number may be about Ernst’s plateau, but that’s putting in a lot of assumptions. Braley has far more resources down the stretch, but that excludes outside money. Additionally, Braley has always had more resources, many of which he’s already spent and he has still not yet pulled away. So, we’re stuck in a close race where both candidates have their work cut out for them down the stretch.

Iowa Senate Rating: Toss-Up

Iowa Senate Model: Braley(D) +1.4%; 48.4% to 47.0% 

Ohio Governor: The Race That Never Was

Published by:

As the dog days of August are upon us, Ed Fitzgerald(D) has never made it a race in Ohio. A year or two ago, Gov. Kasich(R) appeared to be one of the most vulnerable governors in the country. He had swept into office during the GOP wave of 2010, and, like Governors Scott in Florida, Snyder in Michigan, and Walker in Wisconsin, his administration worked with state legislative bodies to bring sweeping conservative change to their states that proved initially quite unpopular. While all four governors have bounced back from their lows early in their terms, Walker, Scott, and Snyder all have races on their hands. Kasich? Not so much.

Most of it is the opponent. Democrats never really had a top tier candidate this cycle. While Scott is running against a former governor in Charlie Crist and Snyder against a former Congressman in Mark Schauer, Ed Fitzgerald is a County Executive. Doesn’t have quite the same ring to it, does it? Certainly, depending on the candidate, this may be an asset, especially in a year when many people are tired of establishment and long serving politicians on both sides. But, if nothing else, this generally means a big disadvantage in terms of money. Former Governors, Congresspersons, and high end state assembly members generally have strong ties and access to big pockets of money. Thus, it takes a special kind of Howard Dean-type candidate to figure out how to fund raise as the “little guy” in the race. Fitzgerald has never been able to become that little guy.

Back in June, John Kasich was shown to have $9.3m in the bank while Fitzgerald had just $1.9m. While Ohio is a swing state split pretty close down the middle, that only works if you have two top candidates fighting it out. When one candidate is dominating the other on the airways and other arenas, most 50/50 states are no longer 50/50 states.

A few months ago, Fitzgerald was hanging in there, down mid-single digits in the modeling. Perhaps if he were on even financial footing or had more charisma to reach out to voters in other ways, he could have closed that gap. Instead, today, Kashich hits a 7.0% margin in the model, his largest lead to date. While that is not currently the blow out it could be based on such a financial advantage, this race is looking more and more like the race that never was.

Ohio Governor Rating: Likely GOP

Model: Kasich(R) +7.0%; 50.9% to 43.9%

North Carolina Senate Moves From “Toss-Up” to “Leans Dem”

Published by:

I’ve decided to make a ratings change today in the North Carolina Senate race. Here are the dynamics:

Today, for the first time in this race, I have a candidate ahead by more than 3%, generally a pretty good zone to be in when I have enough data to play with. Hagan is now up 47.2% to Tillis’s 44.0%. The trends have definitely been in her favor as many recent polls from pollsters on both sides (PPP and Civitas) have shown her up by as much as 7%. Now, like the model, I highly doubt her lead is so large. The reason for this is mostly that many of those polls are showing the Libertarian candidate receiving between 8-11%. That’s highly doubtful come election day. The candidate is really just a “none of the above” vote right now, is not well funded, and may likely not even make it into the debates. When this happens, these candidates often do far worse on election day, to the benefit of the candidate that is more similar to their views. In this case, many of these are Tillis leaning voters who are not a fan of him but are far less a fan of Hagan.

That said, we’ll see how long this rating change lasts. In addition to watching where this 3rd party vote goes, Hagan traditionally does best in the summer when the North Carolina General Assembly is still in session. The governing body is quite unpopular in the state and Thom Tillis leads the Assembly on the House end. Thus, he gets a barrage of negative local affiliate coverage this time of year. These two factors make me believe this race may still close back a bit, but, even so, today’s data points are still showing Hagan likely to be ahead by a few points. The thing that makes me most leery is a model prong showing her up by a mere 1%, but I still have her up in all four model prongs, one as large as a 6% lead. Never test fate, but in my 2012 models, no candidate that was ahead in all model prongs ended up losing come election day, no matter how narrow those prongs were. Thus, overall, this is the reason for the change today.

North Carolina Senate Model: Hagan(D) 47.2%; Tillis(R) 44.0%

Kentucky Senate Races Remains Quite Stable; Grimes(D) Holding

Published by:

It has been a month since I last updated the Kentucky Senate model. The result? A race that looks incredibly stable. A month ago, I had Grimes(D) narrowly ahead by 0.4%. Today, I still have her up by a similar 0.2%. Unlike, say, that Georgia Governor race that has been dramatically shifting over the last month, this race hasn’t moved much at all as both sides are dug in. At the moment, I only have about 4% of voters as being truly uncommitted. Like Georgia, the fact that we’re even talking about this race 99 days until the election is great news for Democrats. If Grimes ends up wining here, that map to 51 is incredibly challenging for the GOP. The best news for Grimes is that she’s already in the mid to upper 40s in three of my model prongs, ranging from 46-48% in those lines. McConnell, meanwhile, remains stuck in the 45-46% range in those lines. His saving grace? The election day forecast prong is pushing many of those undecideds heavily into his camp, predicting they will “come home” to the conservative candidate in the conservative state. That said, the closer we get to election day, and the longer Grimes is holding her own or even ahead in those other three model lines, the better looking she’ll be. This one still looks like a stellar race…

Kentucky Senate Model: Grimes(D) +0.2%

Kentucky Senate Rating: Toss-Up

South Carolina Gov Moves from “Leans GOP” to “Safe GOP”

Published by:

Nikki Haley(R) now appears to be cruising toward re-election. She had a closer contest than predicted in 2010, but her re-election bid seems to be going well. While a few pollsters still have her up in the 3-4% range, many others have her up in the mid-teens. What is apparent is that there are no indications she’s on track to lose this race. There has not been a lot of data in this race until this last month, but now that there is, Haley is rather clearly hitting 50% or more in all model prongs. Overall, I have her up 10.2% with less than 100 days to go. Her prong range is from 50-53% while her Democrat opponent Vincent Sheheen, is stuck in the 40-42% range. Thus, even if Sheheen were to close well and make it more competitive, it still seems doubtful he’ll do much better than the slight surprise of a 4% or so margin that he achieved in 2010.

Carter(D) Moves Ahead in GA Gov Race for First Time

Published by:

For the first time, Jason Carter has taken a lead in the Georgia Governor model. Carter had been chipping away at incumbent Governor Nathan Deal for the last few weeks. Ten days ago, Carter was behind by 1.5%, his closest margin yet. Today, the race has taken a pretty big shift in its central measures, moving to a Carter +0.5% lead. When races with as much data as this one move 2.0% in one model run, it likely indicates a changing dynamic to the race. In this case, it solidifies what the model has been showing for the last few weeks: this race has officially narrowed as Carter’s numbers now appear to be very real.

Perhaps most interesting, the last two pollsters in the field, Rasmussen and Landmark, now both show Carter doing a better job against Deal than Nunn against Perdue. For now, I still have Nunn up 1.0%, but Carter’s momentum has continued while Nunn’s has stalled or is perhaps receding. Give it a few more weeks on the Senate race.

The good news for Deal is that the dynamics of the state still lean in his favor. Additionally, there are a whole bunch of undecided and weak lean voters out there, meaning this race remains well up in the air. I currently have Carter up 43.6% to Deal’s 43.1%. In the overall model prongs, Carter is mostly in the 43-45% range; Deal in the 43-44% range. Hence, both candidates are long away from the upper 40s, neither getting above 45% in the model yet. This one is sure worth watching, but Carter has appears well placed to compete against Deal now that we’re within 100 days of the election.

Georgia Governor Rating: Toss-Up

Georgia Margin: Carter(D) +0.5%