2014 battle for control of the US Senate

[FuzzyCat]

Ga. unemployment rate just went up to 7.8!

 

[konradv]

Ga. unemployment rate just went up to 7.8!

Vote out those damn Republicans running the state.

 

[JakeStarkey]

Even with the high unemployment, those folks will not work in the fields for a pay check.

Farm owners, workers worry about immigration law's impact on crops

http://www.ajc.com/news/news/local/farm-owners-workers-worry-about-immigration-laws-i/nQt7D/

 

[Shanty]

Even with the high unemployment, those folks will not work in the fields for a pay check.

Farm owners, workers worry about immigration law's impact on crops

Farm owners, workers worry about immigration law's impact on crop | www.ajc.com

If they pay a living wage, Americans will work.

 

[JakeStarkey]

No, they won't, that's what the stats show in GA, Shart.

Americans will not work for $12 an hour in the fields anymore.

Immigrants will.

 

[tgards79]

Real Clear Politics does a pretty good compilation of polling (aggregate).

Here is their Senate website:

RealClearPolitics - 2014 Election Maps - Battle for the Senate

(the graphic is interactive - clickable, and pretty much everything is hyperlinked)


This is how RCP sees the battle for the Senate right now:

View attachment 31070​

It sees nine seats in play. The GOP needs 4 of those nine seats. It needs 6 seats in order to win the Senate, but 2 of them are already clearly strongly R right now, and pretty much ceded to the GOP: Montana and South Dakota. Both are show on the graphic as "Likely R", and I concur.


Let's take a look at the safe D seats:

The Northeast: Delaware, Massachusetts, Rhode Island
The Midwest: Illinois
The Southwest: New Mexico

5 seats

No one is expecting these seats to flip, even in the case of a massive GOP wave in the Fall.

Let's take a look as the safe R seats:


The South: Alabama, North Carolina, Oklahoma, Oklahoma (special) South Carolina, Tennessee, Texas
The breadbasket: Kansas, Nebraska
Big Sky: Idaho, Wyoming
The Northeast: Maine

12 seats.

No one is expecting these seats to flip, even in the case that a GOP wave does not materialize in the Fall.

Already, the GOP has a massive SAFE seat advantage over the Dems of more than 2:1


Let's take a look at the likely D seats:

The West: Hawaii, Oregon
The Midwest: Minnesota
The Northeast: New Jersey
The South: Virginia


These are the five states where the GOP is not really investing. Mark Warner is likely to win the biggest for the DEMS in this cycle, making him a future presidential candidate as well. Another important story here is the story of the power of the incumbency. Al Franken (D) barely won his Senate seat in 2008, but right now, his average over challenger Mike McFadden (R) is +10.4, which is a landslide aggregate margin. Of these states, the one state that could end up being a surprise on the senatorial level could be Hawaii.


And let's take a look at the likely GOP seats:

The South: Mississippi
The Big Sky/Breadbasket: Montana, South Dakota (O)

Here are already two pretty much guaranteed pick-ups for the Republicans in the Fall (MT, SD).

Now the next is where RCP and I don't completely agree about the level, but we do agree about the direction.

Under the leaning D states:

The Northeast: New Hampshire

The only problem I have with that is that Jean Shahean (D), according to RCP polling averages, is leading Republican Scott Brown by +10.4, which is exactly the same landslide margin that was shown for Al Franken (D) in Minnesota, so why one state should be considered likely D, but the other state is listed as leaning D is a mystery to me. Of course, the NH primary is first on September 9th, so right now the assumption is that Scott Brown, the former Republican Senator from Massachusetts, will become the Republican Senatorial nominee. Perhaps this is why RCP is classifying the two races in two different ways. Wait and see.

And under the leaning R states:


The South: West Virginia (O)

Again, I wonder why RCP is classifying this as leaning R, because right now, Republican Shelly Capito is leading Democrat Natalie Tennent by +9.3, which is a near-landslide margin.


Maybe RCP is just trying to be overly careful, but it sure seems to me that in both cases, those states are more "likely" for either D or R than "leaning", which means that most likely, the GOP already has 3 pick-ups in the bag: MT, SD and WV. Which means it only needs 3 of the statistical tossup states in order to outright have 51 Senate seats and therefore, the majority:


That leaves us with nine statistical tossups:


The South: Arkansas, Georgia (O), Kentucky, Louisiana, North Carolina
The Midwest: Iowa (O) , Michigan (O )
The Southeast / Mountain States: Colorado
The Pacific: Alaska


This is where the strategema looks extremely grim for the Democrats. Of those nine statistical tossups, five of those races are with Democratic incumbents who are all locked in tight races. There is only one race where a Republican incumbent is locked in a tight race (Kentucky - Mitch McConnell), and of the open races, two of the three are currently Democratic seats.

Here are the averages:

Alaska: AK has not had it's primary yet, but the values range from Begich (D) +0.4 over Treadwell (R) to Begich +12.0 over Miller. However, it looks as if Miller has no chance of winning the GOP primary (he beat Murkowski in 2010 and then lost to her as a write-in candidate in the Fall of that year).

Arkansas: Cotton (R) +3, and incumbent Mark Pryor has not won in a poll since April.

Colorado: Udall (D) +1.5

Georgia: Perdue (R) +3.2

Iowa: Ernst (R) +0.8

Kentucky: McConnell (R) +2.5

Louisiana: Cassidy (R) +1

Michigan: Peters (D) +4

North Carolina: Tillis +1.3


Please remember that the aggregate values I am quoting are from today, 11 August 2014, and could already change again in the next days.

Now, 8 of those 9 margins (aggregates) are well within the MoE and the Michigan margin is just outside the standard MoE, but right now, the GOP is a nose ahead in 6 of those 9 races. And remember, the GOP only needs to win 3, assuming that MT, SD and WV all go as pretty much everyone expects them to go.


Almost eight months ago, I put out this thread:


Congressional Elections compared to Presidential Terms 1855-present US Message Board - Political Discussion Forum


Quote at the end:

So, what can we learn from this information?

1.) We have had divided government a lot more than most people realize.

2.) The statistical probability that the President's party suffers major losses in a mid-term election, especially in a 2nd term mid-term, is extremely high.

3.) Divided government is not necessarily a bad thing. See: Eisenhower, Clinton. Likewise, unified government is not necessarily good. See: Hoover, Carter.

So, before the pundits go blabbing their mouths off about the 2014 elections, know that history is on the side of the GOP in this election.

And indeed, what we are seeing right now is exactly in line with electoral history.

And then there is Angus King (I - ME), who, should the GOP only win 5 seats for some reason, could play kingmaker and decide to caucus with the GOP. So, in reality, the GOP only needs to win 5 seats, but 6-8 are very likely.

Now, there are two seats that the Democrats really COULD win:

Georgia - and - Kentucky.

But even that is dicey.

And for this reason, the mention of Angus King. Should the GOP win 7 seats, but lose 2, it could still control the Senate, with Angus King (I) switching sides.

When the last primaries are over with and the polling for the key races comes in, then I will be following the numbers quite closely, but the aggregates, from pollsters from all over the spectrum, are pointing to a very, very good outcome for the GOP in November.

What's the absolute top-line?

Well, I will take RCP's take, which currently shows 45 DEM seats in the basket, and were Angus King to switch sides, then the Senate could move from 54 (D) - 45 (R) 1 (I) to 54 (R) - 45 (D) - 1 (I). It could go from D+9 to R+9, which would then be a partisan shift of R+18.

Please notice that I included the geography when listing the states. This is because the South is playing an enormous role in this: there are four southern states where the Dems could easily lose seats: WV, AR, NC, LA.

My gut tells me that the GOP is going to do better than +6 in the fall, maybe +7 or +8, but could definitely lose one seat. This is also not uncommon in electoral history. Even in historic wave mid-terms, the "other side" has often picked up at least one seat. So, that's also not a big surprise.

The next step in all of this is to see where the GOP places it's money in advertising and the sending in heavy hitters from other areas to help.

I will be updating this thread again in late August.

Oh, and [MENTION=21821]Samson[/MENTION], what was that again about "partisan hack"?

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

 

[Statistikhengst]

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

That is EXTREMELY optomistic for the Democrats EXTREMELY. And not in line with current polling data at all.

Though I am a Democrat, when it comes to polling numbers, I am brutally neutral. The numbers have been and are still pointing toward a GOP take-over of the Senate in 2014, which would also be right in line with electoral history.

 

[Statistikhengst]

Well, the Labor Day weekend has come and gone and as is a big tradition in US-American politics, mid-term election season is underway and we now go into 5th gear.

In exactly 9 weeks from today, voters across the country will go to the polls to elect local, state and national representatives. We know that all 435 seats in the US HOR will be officially up for election (although some will be completely non-competitive) and roughly 1/3 of all Senate seats will be up for grabs.

Also, 36 governors' seats are up for election.

We should have about 240,000,000 US-Americans who would be eligible to vote and maybe, if we are lucky, about 170,000,000 of them will register. Voter turnout in mid-term elections has historically been much lower than for presidential elections. Wait and see. In one month, I will be putting out a massive thread with all of the voter registration statistics throughout the Union.

I also want to once again point to the hard facts about mid-terms, especially mid-terms that occur within the second term of a 2-term presidency:

Congressional Elections compared to Presidential Terms 1855-present US Message Board - Political Discussion Forum

Grab yourself a cup of coffee and take time to really read that thread and let the numbers sink in. They tell an important story that is incontrovertible.

In mid-terms elections, going all the way back to 1854, the first year that the then-fledgling GOP was on the national ballot for such, the opposition party has traditionally made gains, usually, substantial gains, in congress. Way back in January, I wrote that the GOP had the better cards in it's hand, and now, 8 months later, that prognosis has not changed. The GOP still has the better cards in it's hand.

Most believe that mid-terms are a referendum on the sitting POTUS and his administration. I wouldn't dispute that argument. Absent a possibility to "vote the bum out" or vote vigorously to keep him, many voters express their displeasure with the POTUS by voting in the mid-term for the opposition party.

The GOP already has control of the HOR and I fully expect that they will ever so slightly expand their lead. I am suspecting that they might pick up a net 11 seats, maybe less, but this is just a shot in the dark as many HOR races will never be polled and sometimes, the whackiest of surprises can happen. But if the generic aggregate is our guide, about 11 would be right.

However, it's the Senate that all eyes appear to be fixed upon and Republicans know that the magic number is right around SIX, depending on whether or not the DEMS pick up one or two seats in the Senate themselves. All of the reasoning for this has already been laid out in the OP, no need to rehash.

But as of this point in time, we go into the "hot" period of the election cycle.

Already, some things are worth noting:

RCP had West Virginia as "leans GOP" when I wrote the OP - and now, WV been moved to "likely GOP", just as I surmised would happen when I wrote the OP.

At the same time, the Senate seat in Ruby-Red Kansas, which was listed as "likely GOP", has now been moved to "lean GOP". Now, were this a presidential election with an immensely popular Democratic candidate on the ballot for Prez, I could surmize that the DEMS could pull-out a surprise in Kansas, but in a mid-term with so much acrimony across the nation, I am pretty darned sure that this seat stays red.

This means that the GOP really does have 3 of the six seats it needs to get over the finish line in November: MT, SD and WV. It only needs three more, assuming that the DEMS cannot pick-up a seat or two. Wait and see.

The one other factor worth noting is the generic polling. At this time 4 years ago, the Democratic party started to take a massive dive in the aggregate generic polling. That is not happening this time, at least not yet.

Here is a chart going back to 2002. I think it's pretty self-explanatory:

That graphic was put out at KOS, which conservatives will not like, but the math dudes at KOS crunch the numbers just the same as the math dudes at RCP. In fact, the KOS math geeks used RCP's numbers.

Thus far, in generic polling, the Democratic party is not taking the dive it took four years ago, but that could still happen. Wait and see. Even if the end generic shows an absolute tie, it would still be advantage GOP, as many statisticians believe that the DEMS would need about a +7 in AGGREGATE polling just to overcome the effects of Republican Gerrymandering since 2011, and that is not going to happen. Now, I'm not the biggest fan of generic polling, but the aggregate usually does provide some useable information. Just remember: Gallup was off on it's final 2010 generic by 9.3 points. It predicted GOP +15 in 2010. The GOP won on election night 2014 by +5.7. That's a pretty dismal performance on the part of Gallup.

All in all, it looks pretty much like the mid-terms are headed for the results I generally predicted in January.

1.) The GOP retains the HOR.

2.) The GOP captures the Senate, but not with a filibuster-proof majority. This means that the HOR can pass as many Articles of Impeachment or Obamacare revocations it wants, and those things will simply die in the Senate.

3.) The State gubernatorial races will be pretty evenly split, with a couple of pretty deep losses on each side looming, for instance, in Pennsylvania, Kansas and in Illinois. However, there are more (R) gubernatorial seats in the toss-up column than (D) races. The Democrats could make some surprising gains here.

Starting next week, one week after Labor Day, I will be posting the 9 critical battleground state (Senate) aggregates and looking pretty deeply into a number of polls. I am also expecting the "war of the pollsters" to begin shortly. That's always fun to watch. Already, Rasmussen has upped the number of polls.

One final thing: in every mid-term in my memory (since 1974), there have been unexpected surprises. Some candidate may jump the shark or get caught having sex with someone other than his/her wife/husband. Plagarism scandals are become quite popular these days. A nice drug scandal could pop up. And of course, unsavory hot-mic comments can be quite damaging (see: 47%). So, somewhere along the way, a race that was probably not even on anyone's radar could suddenly pop-up.

 

[tgards79]

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

That is EXTREMELY optomistic for the Democrats EXTREMELY. And not in line with current polling data at all.

Though I am a Democrat, when it comes to polling numbers, I am brutally neutral. The numbers have been and are still pointing toward a GOP take-over of the Senate in 2014, which would also be right in line with electoral history.

I did it straight from the polling. I'd like you to tell me where you think it is wrong.

 

[Statistikhengst]

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

That is EXTREMELY optomistic for the Democrats EXTREMELY. And not in line with current polling data at all.

Though I am a Democrat, when it comes to polling numbers, I am brutally neutral. The numbers have been and are still pointing toward a GOP take-over of the Senate in 2014, which would also be right in line with electoral history.

I did it straight from the polling. I'd like you to tell me where you think it is wrong.

I go through all of the aggregates and not one one them, not even from KOS, shows D51-R49 right now.

Much more accurate is

D45 - R46 - in contention 9.

 

[tgards79]

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

That is EXTREMELY optomistic for the Democrats EXTREMELY. And not in line with current polling data at all.

Though I am a Democrat, when it comes to polling numbers, I am brutally neutral. The numbers have been and are still pointing toward a GOP take-over of the Senate in 2014, which would also be right in line with electoral history.

I did it straight from the polling. I'd like you to tell me where you think it is wrong.

I go through all of the aggregates and not one one them, not even from KOS, shows D51-R49 right now.

Much more accurate is

D45 - R46 - in contention 9.

A couple of things to note: the services often use an average of many polls even if they are more than two months old. I only go back to July 1 or, if there are five polls in the last two weeks, only them (more important going forward). Plus services like to add "special sauce," which is essentially their own take on "intangibles." In 2012 I was correct on 33 out of 35 Senate (same as Nate) and in 2008 I was 35 for 35. Plus 45/46 with 9 toss-ups is correct. I called the toss-ups.

 

[JakeStarkey]

My analysis right now is GOP 52 to Dem 48.

 

[Statistikhengst]

Or check this out:
Born To Run The Numbers Election 2014 Senate 2014 Election Update Dems 51 GOP 49...Plenty of Toss-ups...Could Senate Control Come Down to Louisiana In December

That is EXTREMELY optomistic for the Democrats EXTREMELY. And not in line with current polling data at all.

Though I am a Democrat, when it comes to polling numbers, I am brutally neutral. The numbers have been and are still pointing toward a GOP take-over of the Senate in 2014, which would also be right in line with electoral history.

I did it straight from the polling. I'd like you to tell me where you think it is wrong.

I go through all of the aggregates and not one one them, not even from KOS, shows D51-R49 right now.

Much more accurate is

D45 - R46 - in contention 9.

A couple of things to note: the services often use an average of many polls even if they are more than two months old. I only go back to July 1 or, if there are five polls in the last two weeks, only them (more important going forward). Plus services like to add "special sauce," which is essentially their own take on "intangibles." In 2012 I was correct on 33 out of 35 Senate (same as Nate) and in 2008 I was 35 for 35. Plus 45/46 with 9 toss-ups is correct. I called the toss-ups.

Well, good!

I do lots of electoral statistics, so I am sure we are going to bump into each other now and then.

 

[tgards79]

My analysis right now is GOP 52 to Dem 48.

Hey Jake, where can I see your analysis?

 

[daws101]

a potential game changer Campaign manager for Mitch McConnell resigns over Ron Paul bribery allegations

 

[Antares]

Yes

My analysis right now is GOP 52 to Dem 48.

Hey Jake, where can I see your analysis?

Jake, I'd like to see it too?

 

[HenryBHough]

Something frequently overlooked:

Conservative leaning voters tend to distrust telephone pollsters. They (more frequently than liberals or "moderates", whatever in hell those are) just hang up on poll-calls. Or fill the pollster so full of shit that their eyes turn brown.

In-person polling could be expected to be more fruitful but who does THAT anymore?

As to internet polling? Yeah. Right.

 

[Statistikhengst]

Something frequently overlooked:

Conservative leaning voters tend to distrust telephone pollsters. They (more frequently than liberals or "moderates", whatever in hell those are) just hang up on poll-calls. Or fill the pollster so full of shit that their eyes turn brown.

In-person polling could be expected to be more fruitful but who does THAT anymore?

As to internet polling? Yeah. Right.

Which explains why the last round of polling had a CONSERVATIVE mathematical bias.

NOT.

 

[HenryBHough]

Love those brown eyes, Statist!

 

[Statistikhengst]

Your blindness bothers you, what?

You can always push that "informative" button and you will feel better right away!