USA Mid-term Elections--Off to the Races

[onejayhawk]

I did not cache that czech.

Were you referring to Sabato's new "Leaning Runoff" designation?



Marginally related, This is pulled from the RCP Presidential Approval poll. What's with Rassmussen?

Poll | Date | Sample | Approve | Disapprove | Spread
FOX News | 10/12 - 10/14 | 831 LV | 39 | 53 | -14
ABC News/Wash Post | 10/9 - 10/12 | 629 LV | 39 | 57 | -18
Rasmussen Reports | 10/13 - 10/15 | 1500 LV | 48 | 50 | -2
FOX News | 9/28 - 9/30 | 845 LV | 39 | 53 | -14
Associated Press/GfK | 9/25 - 9/29 | 958 LV | 42 | 58 | -16

J

 

[bhsup]

I don't really think this will affect anything, but it's a dumb gaffe.

Kansas Democrats fire spokesman over online insult

Topeka — The Kansas Democratic Party fired its communications director Thursday over an online posting that mocked three towns in the state's southeast corner, and two of the party's top officials promised to visit the region to apologize.

The fired party spokesman, Dakota Loomis, issued a statement asking people not to hold "my own immaturity and stupidity" against Democrats. The flap over Loomis' posting comes as Democrat Paul Davis remains in a tight race with Republican Gov. Sam Brownback.

Loomis posted his comment last week on a site dedicated to University of Kansas basketball, describing the towns of Cherryvale, Columbus and Galena as leading contenders for the "most craphole small towns" in the state. The comment had been deleted when the Pittsburg Morning Sun reported on it Tuesday.

Now, on a side note, this IS Kansas we're talking about here, so by default all the towns are crap hole towns. Should Loomis have been fired for speaking the truth?? Reporting to you live from South Iowa, this is bhsup signing off.

 

[onejayhawk]

I'm a Mainer, and I can't see King caucusing with the GOP unless they take a clear majority in the Midterms. I don't think many in the GOP today would want him caucusing with them, to be clear. So many of them hate Susan Collins to begin with.

That's pretty much everyone's opinion. The New York Times giving 3-1 odds that the Republicans will have that clear majority.

My question is how much will King (and possibly Orman) matter if they do expand the Republican majority?

J

Moderator Action: edited out accidental reference to ad bot - ori

 

[bhsup]

This has to be the funniest article title all year.

Obama makes rare campaign trail appearance, people leave early

(Reuters) - President Barack Obama made a rare appearance on the campaign trail on Sunday with a rally to support the Democratic candidate for governor in Maryland, but early departures of crowd members while he spoke underscored his continuing unpopularity.

 

[onejayhawk]

Cute. There have been a lot of stories on the subject of the Presidency and the midterm. Here is the President in his own words.

I am not on the ballot this fall. Michelle’s pretty happy about that. But make no mistake: these policies are on the ballot. Every single one of them.​

http://www.washingtonpost.com/blogs...-really-wish-president-obama-didnt-say-today/

This was a good week for the Democrats. Georgia looks to be headed for a run off. South Dakota is now in WtHK land (Who the H Knows). Iowas twitched back in their favor. There are now several paths to save at least a purple Senate. Baby blue is probably out, because of independents choosing to go with the winners, but that is a thin margin.

The thing to remember is that a late break for the Democrats would save 51 seats. A late break would let the Republicans reach 55, plus King from Maine. The Expected value is expected to fall in the middle, say 52 plus King, or 51 plus King and Orman.

Many people watch the Presidents approval ratings, currently at -11.8 in the RCP average. The one Nate Silver touts is Congressional generic ballot, currently favoring the GOP by 3.4, after favoring Democrats by more than a point in August. For comparison, in 2006 the Democrats led by 11.5, winning by 7.9. In 2010, the Republicans led by 9.4, winning by 6.8. If that spread holds up, the Republicans will reach 51 or 52 seats, plus independents. That said, in 2002 the Republicans led on the generic ballot by 1.7, but won by 4.6. I discount that to 9/11 hangover, but it is still in the books.

On the general subject of Presidential approval numbers, I post this upthread.

Poll | Date | Sample | Approve | Disapprove | Spread
FOX News | 10/12 - 10/14 | 831 LV | 39 | 53 | -14
ABC News/Wash Post | 10/9 - 10/12 | 629 LV | 39 | 57 | -18
Rasmussen Reports | 10/13 - 10/15 | 1500 LV | 48 | 50 | -2
FOX News | 9/28 - 9/30 | 845 LV | 39 | 53 | -14
Associated Press/GfK | 9/25 - 9/29 | 958 LV | 42 | 58 | -16

The answer is in the way Rasssmussen does its polls--Strongly Disapprove|Disapprove|Approve| Strongly Approve. The results noted above are 40|10|27|21. RCP quotes that as 50 Disapprove vs 48 Approve = 2 Disapprove. Rassmussen quotes it as 40 vs 21 = -19. There are things to dislike about both methods, but the second is more consistent with other pollsters.

J

 

[onejayhawk]

Two weeks out there is a surprising lack of activity. One minor change, RCP moves Arkansas out of the tossup column to leans Republican. That's it. Both sides continue to pour money into key states. Clearly losing candidates provide soundbites. Same old song, same old dance, new week.

Wednesday should bring a round of updates.

J

 

[Cutlass]

Folks on the radio are calling the Connecticut governor's race the most contentious in the country. It's pretty ugly, and a lot of outside money is being spent. A recent poll has Foley up enough to be called a clear favorite.

 

[onejayhawk]

There is an amazing lack of activity for this close to an election. Generally, it has been more of the same, with time running out.

The NYT gave some hard to understand illustrations of Democrat chances. How about 538's? Nate Silver understands how we think.

A 60 percent chance of an outcome occurring means there’s a 40 percent chance of it failing to occur. As 60-40 underdogs, Democrats’ chances of keeping the Senate would be about as good as Ted Williams’s chances of getting a base hit in 1941.

Over the past week or two, the FiveThirtyEight forecast has drifted slightly more toward Republicans. As of Wednesday night, the GOP’s chances of a Senate takeover were up to 66 percent, its highest figure on the year. Sixty-six percent might seem a lot different than 60 percent; it tends to read as “2-to-1 favorites” rather than “just slightly better than a coin flip.” But it isn’t much of a change, really; Democrats still have a 34 percent chance of prevailing. The difference between a 40 percent chance and a 34 percent chance is one additional “hit” for every 17 attempts. Essentially, Democrats have fallen from Williams’s chances of getting a hit in 1941 to Tony Gwynn’s in 1989.
...
Without Georgia as an option, Democratic chances of keeping the Senate would be down to about 25 percent. Twenty-five percent chances come in fairly often, too — they come in 25 percent of the time! — but Democrats would be downgraded from Gwynn to Alfredo Griffin.​

http://fivethirtyeight.com/datalab/the-democrats-path-of-last-resort-is-georgia/

The gist of his article is that Democrats' hopes are largely in run offs, GA and LA. I would add an independent, KS. Historically for small elections, Republicans turn out their vote better than Democrats. Even if Orman wins in Kansas, he is not assured to vote with the Democrats. Still I like the analogy--moving from Ted Williams best season to Alfredo Griffin's average. With two outs and a man at 2B, Alfredo Griffin could get it done, but Teddy Ballgame would look a lot better.

Sabato also has an article along the same lines. http://www.centerforpolitics.org/cr...-with-a-good-chance-of-a-republican-majority/

J

 

[onejayhawk]

Just over a week to go and things are still settling. Still, the picture is emerging for a modest Republican victory in Congress. In the state gubernatorial races, it looks very even gains of one or two either way. The most watched was the Senate Races because there was a good chance of a party flip. That seems to be coming. Professor Dumbledor may not need to change the banners, but the party behind the gavel will very likely be Republican.

To use the NY Times very middle of the pack projections, the Republicans have 47, Democrats 45, with seven in play. In one of those states, Kansas, there is no Democrat on the ballot. Digging further, both parties have two leaners. Like horseshoes, leaners count in polling. That makes the score 50-47, with Kansas, Iowa and Georgia left. All three are unusual.

Georgia is the only solid chance the Democrats have for a flip. The Republicans have four in the bank bag, ready for deposit. This is an open seat and looks likely to go to a run off. The Democrat is Michelle Nunn, from an important political family in Georgia. David Perdue led form most of the last four moths, but this has always been close. The presence of third party candidate Amanda Swafford points this at a run off, which should favor Perdue.

In Kansas, no Democrat is on the ballot. Greg Orman, an independent is running against incumbent Pat Roberts. This was supposed to be an easy win for Roberts. Indeed, the Democrat withdrew to give Orman a clear field. With the first poll in September, Orman was ahead 8%. Roberts has come back to poll evenly. Roberts has the Republican GotV machine, which is likely worth 2% in the actual vote. The polls call this even, but it looks like a fairly easy hold to me.

The Third state is Iowa: Bruce Braley (D)against Joni Ernst (R). Of all the Senate races, this is the most politically polarized. Ernst is an officer in the National Guard and very conservative. Braley is a liberal's Liberal. The winner here will be party bedrock. What makes Iowa unusual is the quality of the polling. Most people don't know it, but statistics was largely developed for agricultural applications. Iowa is cutting edge and the Des Moines Register poll is the best of the best. They have this as Ernst by 1%, with Braley closing. GotV could be critical in this one.

In my opinion, the Democrats must sweep these races, or win two and pull a rabbit out of the hat in Colorado, Arkansas or Louisiana. Both are unlikely. In contrast, if the Republicans win one and hold their leaners, they also get sitting independent Angus King (Maine). My prediction is for a Republican majority of 54 - 46. We may be have two years of vetos coming.

http://www.realclearpolitics.com/epolls/2014/senate/2014_elections_senate_map.html
http://www.nytimes.com/newsgraphics/2014/senate-model/comparisons.html

In the House, look for the GOP to pick up a handful of seats, say 7-11. In State races, the Democrats net two. That would be a clear win, but no wave.

J

 

[Antilogic]

Were you referring to Sabato's new "Leaning Runoff" designation?

That does fix one of my major issues with Sabato's forecasting.

It's looking like the GOP is holding slim leads in a lot of Senate races at the moment, excluding Georgia where Michelle Nunn (D) has pulled ahead (although I switched up Colorado and North Carolina in my predictions a few months ago, I'm pretty happy that I called this as the late sleeper). The governor's map is an utter mess right now, we might be looking at a record number of incumbent losses and turnovers.

 

[onejayhawk]

It's looking like the GOP is holding slim leads in a lot of Senate races at the moment, excluding Georgia where Michelle Nunn (D) has pulled ahead (although I switched up Colorado and North Carolina in my predictions a few months ago, I'm pretty happy that I called this as the late sleeper). The governor's map is an utter mess right now, we might be looking at a record number of incumbent losses and turnovers.

You are right about the Governor's races being a mess. For example, Sabato favors four flips, a run off in Georgia and six toss ups. Tie races run from the most Republican, Kansas, to the most Democrat, Massachusetts. RCP's no toss up map has eight flips, counting Alaska.

One interesting question is whether it will matter significantly in the turnout? If so, will the turnout matter in the Congressional and Senate races? Another question is whether either party has any momentum? Polling is a lagging indicator.

J

 

[onejayhawk]

Princeton (Wang) did an article on the state races. The headlines says most of it, "In State Races, As Much Suspense As 2006 and 2010 Combined"

One passage talks about the unusual number of incumbents in jeapardy:

Nine of the eleven governors in close races are incumbents. Usually, they would be safe bets for re-election: from 2010 to 2013, the re-election rate of incumbent governors has been 88%. With 28 governors running for reelection this year, we might have expected three or four to be defeated. This year, two are already as good as headed out the door. Not included in this chart is Democrat Neil Abercrombie of Hawaii, who was defeated in the primary election. Also not shown is Republican Tom Corbett of Pennsylvania, who currently lags by nine percentage points in his race and is nearly certain to lose. So the rate of involuntary retirements is close to the historical record, over a week before Election Day and before even a single general election has been completed.

Since senators began to be elected by popular vote in 1914, their re-election rate has been 89%. Twenty-eight sitting Senators will be on the ballot next week. Potential losers include six close contests that feature an incumbent in the race: Alaska, Colorado, Kansas, Kentucky, New Hampshire, and North Carolina. Four of these incumbents are Democrats. In addition, two Democratic Senators lag by four points or more, Mark Pryor of Arkansas and Mary Landrieu of Louisiana.​

While a Senate flip has been more likely than not for some time, it is interesting to note an unusual number of tight races, 7-10 depending on criteria. Also interesting are the way a couple of state races interact with Senate races. For example, in Kansas Republican Governor Sam Brownback is a close race. Senator Pat Roberts is unexpectedly in a tight race. Without the GotV push of the Demacrats in the gubernatorial race, independent Senate candidate Greg Orman would likely have little chance against the Republican ground game.

One area we have not mentioned much are the national fundamentals. According to RCP the Generic ballot is +3.0% for the GOP and the Presidents approval is at -12.7%. There have been no major breaks in this election cycle, but the slow creep favors the Republicans.
http://www.realclearpolitics.com/epolls/other/generic_congressional_vote-2170.html
http://www.realclearpolitics.com/epolls/other/president_obama_job_approval-1044.html

 

[Antilogic]

You are right about the Governor's races being a mess. For example, Sabato favors four flips, a run off in Georgia and six toss ups. Tie races run from the most Republican, Kansas, to the most Democrat, Massachusetts. RCP's no toss up map has eight flips, counting Alaska.

One interesting question is whether it will matter significantly in the turnout? If so, will the turnout matter in the Congressional and Senate races? Another question is whether either party has any momentum? Polling is a lagging indicator.

J

Princeton (Wang) did an article on the state races. The headlines says most of it, "In State Races, As Much Suspense As 2006 and 2010 Combined"

One passage talks about the unusual number of incumbents in jeapardy:

Nine of the eleven governors in close races are incumbents. Usually, they would be safe bets for re-election: from 2010 to 2013, the re-election rate of incumbent governors has been 88%. With 28 governors running for reelection this year, we might have expected three or four to be defeated. This year, two are already as good as headed out the door. Not included in this chart is Democrat Neil Abercrombie of Hawaii, who was defeated in the primary election. Also not shown is Republican Tom Corbett of Pennsylvania, who currently lags by nine percentage points in his race and is nearly certain to lose. So the rate of involuntary retirements is close to the historical record, over a week before Election Day and before even a single general election has been completed.

Since senators began to be elected by popular vote in 1914, their re-election rate has been 89%. Twenty-eight sitting Senators will be on the ballot next week. Potential losers include six close contests that feature an incumbent in the race: Alaska, Colorado, Kansas, Kentucky, New Hampshire, and North Carolina. Four of these incumbents are Democrats. In addition, two Democratic Senators lag by four points or more, Mark Pryor of Arkansas and Mary Landrieu of Louisiana.​

While a Senate flip has been more likely than not for some time, it is interesting to note an unusual number of tight races, 7-10 depending on criteria. Also interesting are the way a couple of state races interact with Senate races. For example, in Kansas Republican Governor Sam Brownback is a close race. Senator Pat Roberts is unexpectedly in a tight race. Without the GotV push of the Demacrats in the gubernatorial race, independent Senate candidate Greg Orman would likely have little chance against the Republican ground game.

J

I don't know any studies showing people are more likely to turnout for close gubernatorial races that would affect federal races or vice-versa, I figure there's a lot of overlap though.

I generally don't buy the momentum arguments--I've seen one too many campaigns claim they are behind but have momentum going into election day, only to see them lose by about the same margin the polls said they were going to lose by. Good turnout machines can generally swing the polls 1-2% on election day, maybe a bit more in midterms because there are more voters staying home so motivating them will have a greater impact. Polling error here, due to the lower number of polls, is probably still in the 1-2% range for the aggregate averages. If the race is outside that margin, it's likely to fall to the leading candidate barring some massive (and unlikely) error in the polling.

However, it looks like a lot of races are still in that 2-4% margin.



EDIT: Didn't catch your edit. I think those national indicators are more predictive of House races, the statewide races sometimes take on the aspects of mini-presidential campaigns where some candidates can fly above underlying factors.

 

[onejayhawk]

Momentum is something that can be shown in polling data. If a candidate's lead goes from 12% to 9% to 6% in successive weeks, his opponent has momentum. In other words it is trend over time. Polls lag behind voter sentiment so momentum can point to a decisive shift.

That said, it is usually shows up in voter turn out, which gets buried in GotV numbers. Because of this, some statisticians call it a myth.

J

 

[Antilogic]

Momentum is something that can be shown in polling data. If a candidate's lead goes from 12% to 9% to 6% in successive weeks, his opponent has momentum. In other words it is trend over time. Polls lag behind voter sentiment so momentum can point to a decisive shift.

That said, it is usually shows up in voter turn out, which gets buried in GotV numbers. Because of this, some statisticians call it a myth.

J

Yes it can be, but campaigns often claim they have momentum when it doesn't show up in the polls. See the Romney campaign in the last few weeks of the 2012 election for a recent example.



Apparently, there is some controversy in Ohio over the Plain Dealer editorial board taking down a video of a panel with Kasich, Fitzgerald, and the Green candidate whose name is escaping me at the moment. It appears that Kasich delivered an incredibly disrespectful performance, but the paper removed any trace of the video and proceeded to endorse him afterwards. Too bad Fitzgerald messed up so much early on, this could have been decisive in a closer race.

 

[Cutlass]

Spoiler :

 

[Antilogic]

That is all kinds of special.

 

[onejayhawk]

Best pic of Franken I have seen. As an ad, that is mild compred to what both sides are doing in North Carolina and Georgia.

J

 

[onejayhawk]

Does anyone have any ideas why the Washington Post has been, by far, the most down on Democrat chances in the Senate. Today they have 93% chance of a Republican majority. NYT is 68%, KOS is 66%, 538 is 64%, etc. 2 to 1 odds are a lot different 6 to 1.

Take, for example, this article. While the article is new, the WP's position as the naysayer is not.

There were lots (and lots) of reasons given for the difficulties facing Democrats. The Senate map. The historic trends of second term, midterm elections — aka the "six-year itch." Voter apathy. But the one factor that virtually every person I talked to cited as the biggest reason for the party's current predicament was President Obama. "This off-year election has become almost entirely a referendum on the president," said one Democratic consultant involved in a number of closely-fought congressional races. "It's not just anger at [the Affordable Care Act]. He has become, in my opinion wrongly, the symbol of dysfunction in Washington."​

http://www.washingtonpost.com/blogs...ment-is-terrible-for-democrats-say-democrats/

This election has more close races than any year since 2002, yet the apathy level is high. I don't get that.

Since we were talking about momentum--
http://fivethirtyeight.com/datalab/senate-update-the-momentum-mirage/

J

 

[Antilogic]

Shoot, lost a post due to some server login problem.

The short version in a sports analogy: you would make different predictions as to which team would win a football game if they were up a field goal in the first quarter or at the 2 minute warning in the fourth.

The Washington Post is really aggressively tuned to the most recent polling data and does not dampen its forecast to allow for uncertainty in the coming week--for them, it's always the end of the fourth quarter. Thus, they have been giving incredibly high probabilities this entire cycle, sometimes switching from 90% incumbent one day to 80% challenger the next on the basis of a single new poll being released. It's interesting to use if you want to know what would have happened had the election been held that morning.

Other models, like 538, the NYT Upshot, and the PEC, have either separate same-day predictions or dampen the percentages to account for uncertainty between now and election day.



(I'm definitely closer to the 538 interpretation of momentum here. Some skepticism of that term is healthy.)