2014 battle for control of the US Senate

[Dot Com]

HBH is good at space-filler, posts My gawd but Repub voters have a lot of hate

As to the OP, keep up the good work of posting the updates.

 

[HenryBHough]

And there she goes, calling once more on strangers....

 

[Dot Com]

^ place-filler post

 

[rightwinger]

I cannot count the number of telephone pollsters I have filled so full of shit that their eyes turned brown. I'm sure both sides are doing the same thing so trust those polls as much as your low IQ allows!

I call bullshit

 

[JakeStarkey]

I cannot count the number of telephone pollsters I have filled so full of shit that their eyes turned brown. I'm sure both sides are doing the same thing so trust those polls as much as your low IQ allows!

I call bullshit

HBH being surveyed by phone.

 

[Statistikhengst]

Since lots of polls are conducted on the weekend, I suspect that a bevy of end-polling results will be coming out today and tomorrow, so I may wait until very late on Tuesday to do the weekly analysis. Thereafter, if time permits (and that is a big IF right now, stuff going on in my life), I may do an analysis once a day.

One thing, however, is absolutely sure: mathematically, we are NOT seeing evidence of a GOP wave anywhere. The generic aggregate margin for the GOP is decidedly lower than it was at this point in time in 2010, a number of races (IA, AR, AK, CO) that should have opened way up for the GOP - were there a real wave - are just as close as they were two weeks ago, if not closer, esp. in the case of Iowa. And don't forget, the RCP aggregate in 2010 was almost 4 points to the Right of reality, due mostly in part to very, very bad polling from both Gallup (a now discredited organization) and FOX News. This means that the real possibility exists that a +3.7 for the GOP on paper may actually mean a tie in the generic aggregate. That is the beauty of building a historical database, we learn a lot about trends over time. Really, it's not rocket science.

In Georgia, the possibility of the GOP actually losing a seat, at least on November 4, is now very, very large, but hope abounds for the GOP that this goes into overtime and in a pure two-way runoff, that Perdue can yet prevail.

In Kansas, the possibility of the GOP losing a seat to an Independent is very, very large. So, where the GOP needed 6 to get to 51, it may very well need 8. However, if Angus King (I-ME) were to switch sides, then the GOP would only need 7. That is a doable task, but again, MATHEMATICALLY, there is no sign of a large GOP wave, not even a sign of a small wave.

The GOP is narrowly ahead in a number of traditionally crimson red states where Democratic incumbents or open seat contenders are struggling. That does not constitute a wave. Were this a wave, then, for instance, Tom Cotton (R) would be at least +10 over incumbent Pryor (D-inc) in Arkansas, a state that Mitt Romney (R) carried by +24 points.

And I will remind again that there is a distinct possibility that polling in Colorado is off, just as it was off in 2008, 2010 and 2012 - to the RIGHT, to note, because of the gross miscalculation of the Latino vote. That's not the partisan in me speaking, that is historical fact speaking, something I've addressed a number of times on this thread.

 

[Statistikhengst]

On a side note, I turned in my absentee ballot last week and did my civic duty. That is always a good feeling.

I want to encourage everyone in USMB to go vote. I don't care who you vote for, that is your business, but go vote. Do your civic duty.

 

[pepperpot]

Since lots of polls are conducted on the weekend, I suspect that a bevy of end-polling results will be coming out today and tomorrow, so I may wait until very late on Tuesday to do the weekly analysis. Thereafter, if time permits (and that is a big IF right now, stuff going on in my life), I may do an analysis once a day.

One thing, however, is absolutely sure: mathematically, we are NOT seeing evidence of a GOP wave anywhere. The generic aggregate margin for the GOP is decidedly lower than it was at this point in time in 2010, a number of races (IA, AR, AK, CO) that should have opened way up for the GOP - were there a real wave - are just as close as they were two weeks ago, if not closer, esp. in the case of Iowa. And don't forget, the RCP aggregate in 2010 was almost 4 points to the Right of reality, due mostly in part to very, very bad polling from both Gallup (a now discredited organization) and FOX News. This means that the real possibility exists that a +3.7 for the GOP on paper may actually mean a tie in the generic aggregate. That is the beauty of building a historical database, we learn a lot about trends over time. Really, it's not rocket science.

In Georgia, the possibility of the GOP actually losing a seat, at least on November 4, is now very, very large, but hope abounds for the GOP that this goes into overtime and in a pure two-way runoff, that Perdue can yet prevail.

In Kansas, the possibility of the GOP losing a seat to an Independent is very, very large. So, where the GOP needed 6 to get to 51, it may very well need 8. However, if Angus King (I-ME) were to switch sides, then the GOP would only need 7. That is a doable task, but again, MATHEMATICALLY, there is no sign of a large GOP wave, not even a sign of a small wave.

The GOP is narrowly ahead in a number of traditionally crimson red states where Democratic incumbents or open seat contenders are struggling. That does not constitute a wave. Were this a wave, then, for instance, Tom Cotton (R) would be at least +10 over incumbent Pryor (D-inc) in Arkansas, a state that Mitt Romney (R) carried by +24 points.

And I will remind again that there is a distinct possibility that polling in Colorado is off, just as it was off in 2008, 2010 and 2012 - to the RIGHT, to note, because of the gross miscalculation of the Latino vote. That's not the partisan in me speaking, that is historical fact speaking, something I've addressed a number of times on this thread.

Been watching a lot of the televised debates lately vs the candidates and have noticed there is a lack of what seems to be preparation by the democrats. Most of my family are dems and they just sit and yell at the screen, its amusing. I mean ebola and ISIS are the big topics it seems, why don't they have prepared answers? I noticed at the Shaheen vs Brown debate and the Hagan vs Tillis debates whenever the Koch brothers came up, the audience booed. I think its becoming a losing tactic and they need to go more aggressively after the republicans. They seem to be stuck using the same two or three themes that worked in 2012, but its a brave new world now.

 

[Statistikhengst]

Since lots of polls are conducted on the weekend, I suspect that a bevy of end-polling results will be coming out today and tomorrow, so I may wait until very late on Tuesday to do the weekly analysis. Thereafter, if time permits (and that is a big IF right now, stuff going on in my life), I may do an analysis once a day.

One thing, however, is absolutely sure: mathematically, we are NOT seeing evidence of a GOP wave anywhere. The generic aggregate margin for the GOP is decidedly lower than it was at this point in time in 2010, a number of races (IA, AR, AK, CO) that should have opened way up for the GOP - were there a real wave - are just as close as they were two weeks ago, if not closer, esp. in the case of Iowa. And don't forget, the RCP aggregate in 2010 was almost 4 points to the Right of reality, due mostly in part to very, very bad polling from both Gallup (a now discredited organization) and FOX News. This means that the real possibility exists that a +3.7 for the GOP on paper may actually mean a tie in the generic aggregate. That is the beauty of building a historical database, we learn a lot about trends over time. Really, it's not rocket science.

In Georgia, the possibility of the GOP actually losing a seat, at least on November 4, is now very, very large, but hope abounds for the GOP that this goes into overtime and in a pure two-way runoff, that Perdue can yet prevail.

In Kansas, the possibility of the GOP losing a seat to an Independent is very, very large. So, where the GOP needed 6 to get to 51, it may very well need 8. However, if Angus King (I-ME) were to switch sides, then the GOP would only need 7. That is a doable task, but again, MATHEMATICALLY, there is no sign of a large GOP wave, not even a sign of a small wave.

The GOP is narrowly ahead in a number of traditionally crimson red states where Democratic incumbents or open seat contenders are struggling. That does not constitute a wave. Were this a wave, then, for instance, Tom Cotton (R) would be at least +10 over incumbent Pryor (D-inc) in Arkansas, a state that Mitt Romney (R) carried by +24 points.

And I will remind again that there is a distinct possibility that polling in Colorado is off, just as it was off in 2008, 2010 and 2012 - to the RIGHT, to note, because of the gross miscalculation of the Latino vote. That's not the partisan in me speaking, that is historical fact speaking, something I've addressed a number of times on this thread.

Been watching a lot of the televised debates lately vs the candidates and have noticed there is a lack of what seems to be preparation by the democrats. Most of my family are dems and they just sit and yell at the screen, its amusing. I mean ebola and ISIS are the big topics it seems, why don't they have prepared answers? I noticed at the Shaheen vs Brown debate and the Hagan vs Tillis debates whenever the Koch brothers came up, the audience booed. I think its becoming a losing tactic and they need to go more aggressively after the republicans. They seem to be stuck using the same two or three themes that worked in 2012, but its a brave new world now.

Most of the debates I cannot see from where I live, but I am not so sure Democrats are unprepared in terms of information. They are probably just not expecting any sane person to use a health crisis as a battering ram in an election, but as in love and war, in elections, (almost) anything is fair game.

 

[rightwinger]

Since lots of polls are conducted on the weekend, I suspect that a bevy of end-polling results will be coming out today and tomorrow, so I may wait until very late on Tuesday to do the weekly analysis. Thereafter, if time permits (and that is a big IF right now, stuff going on in my life), I may do an analysis once a day.

One thing, however, is absolutely sure: mathematically, we are NOT seeing evidence of a GOP wave anywhere. The generic aggregate margin for the GOP is decidedly lower than it was at this point in time in 2010, a number of races (IA, AR, AK, CO) that should have opened way up for the GOP - were there a real wave - are just as close as they were two weeks ago, if not closer, esp. in the case of Iowa. And don't forget, the RCP aggregate in 2010 was almost 4 points to the Right of reality, due mostly in part to very, very bad polling from both Gallup (a now discredited organization) and FOX News. This means that the real possibility exists that a +3.7 for the GOP on paper may actually mean a tie in the generic aggregate. That is the beauty of building a historical database, we learn a lot about trends over time. Really, it's not rocket science.

In Georgia, the possibility of the GOP actually losing a seat, at least on November 4, is now very, very large, but hope abounds for the GOP that this goes into overtime and in a pure two-way runoff, that Perdue can yet prevail.

In Kansas, the possibility of the GOP losing a seat to an Independent is very, very large. So, where the GOP needed 6 to get to 51, it may very well need 8. However, if Angus King (I-ME) were to switch sides, then the GOP would only need 7. That is a doable task, but again, MATHEMATICALLY, there is no sign of a large GOP wave, not even a sign of a small wave.

The GOP is narrowly ahead in a number of traditionally crimson red states where Democratic incumbents or open seat contenders are struggling. That does not constitute a wave. Were this a wave, then, for instance, Tom Cotton (R) would be at least +10 over incumbent Pryor (D-inc) in Arkansas, a state that Mitt Romney (R) carried by +24 points.

And I will remind again that there is a distinct possibility that polling in Colorado is off, just as it was off in 2008, 2010 and 2012 - to the RIGHT, to note, because of the gross miscalculation of the Latino vote. That's not the partisan in me speaking, that is historical fact speaking, something I've addressed a number of times on this thread.

Been watching a lot of the televised debates lately vs the candidates and have noticed there is a lack of what seems to be preparation by the democrats. Most of my family are dems and they just sit and yell at the screen, its amusing. I mean ebola and ISIS are the big topics it seems, why don't they have prepared answers? I noticed at the Shaheen vs Brown debate and the Hagan vs Tillis debates whenever the Koch brothers came up, the audience booed. I think its becoming a losing tactic and they need to go more aggressively after the republicans. They seem to be stuck using the same two or three themes that worked in 2012, but its a brave new world now.

Most of the debates I cannot see from where I live, but I am not so sure Democrats are unprepared in terms of information. They are probably just not expecting any sane person to use a health crisis as a battering ram in an election, but as in love and war, in elections, (almost) anything is fair game.

The biggest challenge for Dems in this election cycle has been Republicans have kept their mouths shut. They have been schooled in not talking about legitimate rape, birth control and immigration

 

[pepperpot]

Since lots of polls are conducted on the weekend, I suspect that a bevy of end-polling results will be coming out today and tomorrow, so I may wait until very late on Tuesday to do the weekly analysis. Thereafter, if time permits (and that is a big IF right now, stuff going on in my life), I may do an analysis once a day.

One thing, however, is absolutely sure: mathematically, we are NOT seeing evidence of a GOP wave anywhere. The generic aggregate margin for the GOP is decidedly lower than it was at this point in time in 2010, a number of races (IA, AR, AK, CO) that should have opened way up for the GOP - were there a real wave - are just as close as they were two weeks ago, if not closer, esp. in the case of Iowa. And don't forget, the RCP aggregate in 2010 was almost 4 points to the Right of reality, due mostly in part to very, very bad polling from both Gallup (a now discredited organization) and FOX News. This means that the real possibility exists that a +3.7 for the GOP on paper may actually mean a tie in the generic aggregate. That is the beauty of building a historical database, we learn a lot about trends over time. Really, it's not rocket science.

In Georgia, the possibility of the GOP actually losing a seat, at least on November 4, is now very, very large, but hope abounds for the GOP that this goes into overtime and in a pure two-way runoff, that Perdue can yet prevail.

In Kansas, the possibility of the GOP losing a seat to an Independent is very, very large. So, where the GOP needed 6 to get to 51, it may very well need 8. However, if Angus King (I-ME) were to switch sides, then the GOP would only need 7. That is a doable task, but again, MATHEMATICALLY, there is no sign of a large GOP wave, not even a sign of a small wave.

The GOP is narrowly ahead in a number of traditionally crimson red states where Democratic incumbents or open seat contenders are struggling. That does not constitute a wave. Were this a wave, then, for instance, Tom Cotton (R) would be at least +10 over incumbent Pryor (D-inc) in Arkansas, a state that Mitt Romney (R) carried by +24 points.

And I will remind again that there is a distinct possibility that polling in Colorado is off, just as it was off in 2008, 2010 and 2012 - to the RIGHT, to note, because of the gross miscalculation of the Latino vote. That's not the partisan in me speaking, that is historical fact speaking, something I've addressed a number of times on this thread.

Been watching a lot of the televised debates lately vs the candidates and have noticed there is a lack of what seems to be preparation by the democrats. Most of my family are dems and they just sit and yell at the screen, its amusing. I mean ebola and ISIS are the big topics it seems, why don't they have prepared answers? I noticed at the Shaheen vs Brown debate and the Hagan vs Tillis debates whenever the Koch brothers came up, the audience booed. I think its becoming a losing tactic and they need to go more aggressively after the republicans. They seem to be stuck using the same two or three themes that worked in 2012, but its a brave new world now.

Most of the debates I cannot see from where I live, but I am not so sure Democrats are unprepared in terms of information. They are probably just not expecting any sane person to use a health crisis as a battering ram in an election, but as in love and war, in elections, (almost) anything is fair game.

The biggest challenge for Dems in this election cycle has been Republicans have kept their mouths shut. They have been schooled in not talking about legitimate rape, birth control and immigration

There is an article about just that this morning. It basically says the reason for that is the GOP has picked more women to run.

How the war on women is changing - CNN.com

 

[Statistikhengst]

So, in 6 days, we Americans will go to the polls and elect a new US-Congress. Actually, many of us have already voted. I sent in my absentee ballot 10 days ago.

Regardless of your political orientation, which is no one's business but your own, I encourage EVERYONE who is a US citizen to go vote. It is a right enumerated in the US-Constitution and I personally consider it to be a civic duty.

The time frame between 2 weeks before this election and 1 week before this election has seen a number of races tighten-up more rather than open-up. This already flys in the face of conventional wisdom and electoral polling statistics, which means that in many ways, the 2014 mid-terms will be a set of elections for the record-books, I suspect.

The real shift right now continues seen in Georgia, where Democrat Michelle Nunn is picking up steam and it really does look like a tied-up horserace. There are, for my way of thinking, four real horseraces: Georgia (Perdue vs. Nunn), Kansas (Roberts vs. Orman), North Carolina (Tillis vs. Hagan) and Iowa (Ernst vs. Braley)

I wrote two weeks ago that Gary Peters (D-MI) was going to leave the battlegrounds. His aggregate stood at +9 last week which was well inside the comfort zone. This week, it now stands at +11, so the decision to remove him from the battlegrounds was a correct decision. The RNC has also pulled money out of MI, which means they have given up on it.

In Kentucky, the DNC had pulled funds out of Kentucky and it looked like Mitch McConnell will be re-elected, but then newer polling convinced the DNC to switch course, probably all for naught.

Also, one week ago, I noted that usually around the -3 to -2 week period, pols, pollsters, pundits and the like often call it the break-away period, where one candidate starts to surge into a major lead. We are not seeing that, either. This year is somehow just a little different that other mid-terms.

Here was the posting from eight days ago:

2014 battle for control of the US Senate Page 9 US Message Board - Political Discussion Forum


And here are the aggregates as of today, October 29, 2014:

DEMOCRATS LEADING:
​

Aggregate, NH:

August 11, 2014: Shaheen +10.4
September 16, 2014: Shaheen +3.5
September 22, 2014: Shaheen +5.0
October 2, 2014: Shaheen +4.0
October 7, 2014: Shaheen +4.6
October 15, 2014: Shaheen +3.5
October 21, 2014: Shaheen +2.6
October 29, 2014: Shaheen +2.2

The needle has moved 0.4 in Scott Brown's (R) direction, but Shaheen has lead in 5 of the last 6 polls, and they are all well within the 2-week standard time-frame for making realistic aggregate values. That being said, Shaheen has lost a lot of ground since August. This race could easily tip at the last minute, but for now, we are seeing a small but resilient lead for the incumbent Democrat. A disadvantage for Shaheen is that New Hampshire is one of 14 states in our Union that does NOT allow early voting, so she has had no way to "bank" some votes before election day. That being said, the last round of polling for New Hampshire, in 2012, was off to Right by about 3.4 points (Obama vs. Romney): see the bottom of this posting.

REPUBLICANS leading:
​

Aggregate, AK:

August 11, 2014: -no aggregate was possible-
September 16, 2014: Sullivan +1.3
September 22, 2014: Sullivan +1.3
October 2, 2014: Sullivan +4.7
October 7, 2014: Sullivan +4.7
October 15, 2014: Sullivan +4.4
October 21, 2014: Sullivan +4.4
October 29, 2014: Sullivan +2.2

There has been new polling in Alaska (finally) and the needle has moved +2.2 points toward Begich (D-inc), a considerable shift and pretty much a mirror-image of the New Hampshire race. An aggregate +2.2 is well with the MoE, so were this figure to hold, these races could easily be coin-flips. In both cases, the underdog has made gains.

​

Aggregate, CO:

August 11, 2014: Udall +3.7
September 16, 2014: Udall +1.5
September 22, 2014: Udall +0.6
October 2, 2014: Gardner +1.5
October 7, 2014: Gardner +0.6
October 15, 2014: Gardner +1.4
October 21, 2014: Gardner +3.0
October 29, 2014: Gardner +3.3

The needle has moved +0.3 points in Gardner's direction. In the postings from October 15th and from October 7th, I reminded how very off polling was in Colorado and in Nevada in 2010, also in 2012. Reason: gross miscalculation of the Latino vote. But on paper, Gardner has the lead. +3.3 is still within the MoE.

Aggregate, KY:

August 11, 2014: McConnell +2.5
September 16, 2014: McConnell +5.2
September 22, 2014: McConnell +5.2
October 2, 2014: McConnell +5.3
October 7, 2014: McConnell +4.2
October 15, 2014: McConnell +3.0
October 21, 2014: McConnell +4.7
October 29, 2014: McConnell +4.4

The needle has moved +0.3 toward Grimes (D)- which is statistical noise, but this race has been at +3 to +5 for McConnell now for 6 weeks straight. He has a somewhat slim but very stable lead over his Democratic challenger. Now only that, McConnell has won in tight races before, he is a survivor. And he is winning in a ruby-red state that Mitt Romney carried by almost +23 point two years ago.

Aggregate, AR:

September 16, 2014: Cotton +2.5
September 22, 2014: Cotton +2.5
October 2, 2014: Cotton +3.6
October 7, 2014: Cotton +3.7
October 15, 2014: Cotton +4.4
October 21, 2014: Cotton +3.6
October 29, 2014: Cotton +5.0

The needle has jumped +1.4 toward Republican challenger Tom Cotton and +5 is definitely outside the MoE. If these numbers hold, then Cotton will easily defeat Pryor (D-inc) in a ruby-red state that Mitt Romney carried by almost 24 points two years ago.

Please notice the numeric relation between the current aggregates in both Arkansas and Kentucky vis-a-vis the 2012 electoral statistic: in both states, the Republican candidates for US-Senate are running ca. 19 points behind Romney's landslide margins from 2012. Both states are in the South and both states are part of the "Bible Belt".

Double aggregate, LA:

Jungle Primary, October 2, 2014: Landrieu +1.2
Jungle Primary, October 7, 2014: Landrieu +2.7
Jungle Primary, October 15, 2014: Landrieu +2.7
Jungle Primary, October 21, 2014: Landrieu +2.8
Jungle Primary, October 29, 2014: Landrieu +4.2

Two-"man" race, possible runoff, October 2, 2014: Cassidy +4.6
Two-"man" race, possible runoff, October 7, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 15, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 21, 2014: Cassidy +5.3
Two-"man" race, possible runoff, October 29, 2014: Cassidy +4.5

There has been an interesting change in this double aggregate. In the Jungle primary; the needle just moved +1.4 toward Landrieu (D-inc), definitely outside of the MoE. And the needle moved +0.8 toward her in the polling for a possible runoff. Only, third party candidate Maness is really holding well at an aggregate of 9 points. This means that Landrieu needs at least a +6 aggregate to have a chance of getting over 50% in the Jungle Primary This will very likely go into overtime. I did the mathematical extrapolation for this on October 7th.



New addition to the battlegrounds, which will leave the battlegrounds after today:

Aggregate:

October 21, 2014 (RCP): Rounds +9.8
actual aggregate: Rounds +3.5
October 29, 2014 (RCP): Rounds +12.0

The newer polling shows that Rounds is very likely to win in SD, after all. SD goes off the battleground list after today. The possibility of him losing in a three man race, with the opposition so split against him, is null.

THE REAL DOGFIGHTS - DOWN TO THE WIRE:​

Aggregate, IA:

August 11, 2014: Ernst +0.8
September 16, 2014: Braley +1.4
September 22: Braley +0.1
October 2, 2014: Ernst +2.8
October 7, 2014: Ernst +1.9
October 15, 2014: Ernst +1.2
October 21, 2014: Ernst +2.5
October 29, 2014: Ernst +1.7

The needle has moved 0.8 points toward Braley (D), Ernst has lost ground since last week, with one poll showing an absolute tie one poll showing Braley ahead by a nose. If you remove all polls that are more than two weeks old (the two bottom polls), then the aggregate is: Ernst +1.6, practically the same. Iowa is living up to it's tradition of being a real battleground/tossup state: in large national waves for President, Iowa goes easily with the national winner (Clinton '92 and '96, Obama '08). In squeaker elections, it goes by a hair for the winner in the national popular vote (Gore '00, Bush '04) and in a moderate election win, it goes with a moderate value (Obama '12) for the winner. In presidential politics, you have to go back 26 years to find Iowa bucking the national trend.

In statewide politics, we have had less of a chance to see how "battlegroundy" (not a word, but should be ) this state can be, due to long-term incumbents at the Senatorial level. A +1.7 for any candidate going into the final days of an election cycle can only mean that this may just as well be coin toss. And I suspect that directly before election day, this race will get tighter. Polls in Iowa close at 10 PM, EDT and I strongly suspect that this race will not be called until the next day. The disadvantage for Braley is that Iowa does early voting and no-excuse absentee voting and this means that lots of voters had a chance to already cast ballots when Ernst (R) was at a higher-water-mark.

Aggregate, NC:

August 11, 2014: Tillis +1.3
September 16, 2014: Hagan +3.7
September 22,2014: Hagan +5.0
October 2, 2014: Hagan +4.2
October 7, 2014: Hagan +3.7
October 15, 2014: Hagan +1.5
October 21, 2014: Hagan +1.2
October 29, 2014: Hagan +1.2

The needle has not budged since October 21 and there have now been three tied polls. Within five weeks time, Hagan's aggregate has shrunk from +5 (right at the outer edge of the true battleground zone) to a mere +1.2. The one factor in Hagan's favor is that she has maintained an aggregate lead now for 7 weeeks and North Carolina, like Iowa, does early voting and no-excuse absentee ballots. The polls in North Carolina close at 7:30 PM EDT on election day and I strongly suspect that this race will not be called until either well into the next morning or day.

Aggregate, GA:

August 11, 2014: Perdue +3.2
September 16, 2014: Perdue +3.0
September 22, 2014: Perdue +3.3
October 2, 2014: Perdue +3.4
October 7, 2014: Perdue +3.1
October 15, 2014: Perdue +2.7
October 21, 2014: Perdue +0.6
October 29, 2014: Perdue +0.5

The needle has moved 0.1 points toward Michelle Nunn (D), which is nothing more than statistical noise. This race has stood still over over week and every single one of those polls are new polls.

This makes this race one of the two closest aggregates at this time and the largest overall shift over last two weeks. The problem is, with a third party candidate likely to score at least 2 points or more, a +.5 to +1 win for either major candidate means that no one gets over 50%, which means that Georgia, like Louisiana, is very likely to go into overtime. The polls close in all of Georgia at 7 PM, EDT and I am pretty sure that this race will not be called for many hours, if not for a day. Georgia, like North Carolina and Iowa, has early voting and no excuse absentee ballots. And Georgia also has a problem with some 40,000 ballot applications that for some mysterious reason have still not been processed, which is in and of itself downright negligent, if not criminal. Those ballot applications, presumably all from legal US Citizens and citzens of the state of Georgia, come from heavily African-American voting areas, so legal challenges to voters who were outright suppressed by heavy-handed tactics on the part of some election officials in GA could muddy up this race even more.

Aggregate, KS:
September 22, 2014: Orman +1.2
October 2, 2014: Orman +5.3
October 7, 2014: Orman +5.3
October 15, 2014: mathematical tie
October 21, 2014: mathematical tie
October 29, 2014: Orman +0.9

The needle has move 0.9 toward Orman (I), putting him close to where he was 5 weeks ago, but nowhere near his high-water mark from October 7th.

One week ago, I noted this:

Monmouth has released a poll of KS:

http://www.monmouth.edu/assets/0/32...1087/145f0962-9f03-4ef3-a9e7-78534f38d6d2.pdf

Orman 46 / Roberts 46.

That poll, from a respected national pollster, has still not been included in the aggregate, which I find quite suspect. If you include the poll into the aggregate, then Orman's lead shrinks ever so slightly to +0.7. This is a pretty sure sign that the GOP base is coming back to Roberts (R). This question is: is that enough?

For KS, contrary to conventional wisdom, is NOT a GOP-majority state in terms of voter registration. It is a high GOP minority state with a substantial plurality against the D's, currently: R (44.11%), I (30.84%), D (24.33%), margin: R +19.78%.

Source.

This can only mean that for Orman to be doing this well, he is sweeping both the D and I vote and taking probably about 11% of the R vote at this time. I will do a better extrapolation of this later. Kansas does both early voting and no-excuse absentee voting, just like North Carolina, Iowa and Georgia. The polls for most all of Kansas close at 8 PM EDT, but because of at least one county that closes at 9 PM EDT, the state will not even be releasing votes until 9 PM EDT.

I cannot imagine any calls from any credible news sources for North Carolina, Iowa, Georgia or Kansas until well into the night, or even the next day.

In New Hampshire, however, the call may come earlier than people think. In 2012, the polling aggregate in NH for President Obama was: +2.50. In reality, Obama won by +5.58% and New Hampshire was called for the President at 10:03 PM EDT on election night on CNN and it was called even earlier for him on: CBS, ABC, NBC, MSNBC and FOX.

FACIT: we have seen races tighten up or stay exquisitely tight in: Alaska, New Hampshire, Iowa, North Carolina, Georgia and Kansas.

We have seen the races open or close very slightly in: Kentucky and Colorado.

We have seen the races open in a statistically important way in: Arkansas and Lousiana.

That being said, in the generic aggregate, the GOP lead SHRUNK since last week:

Generic aggregate, October 2, 2014: GOP +2.9
Generic aggregate, October 7, 2014: GOP +2.1
Generic aggregate, October 15, 2014: GOP +2.4
Generic aggregate, October 21, 2014: GOP +3.4
Generic aggregate, October 29, 2014: GOP +2.0

Shift: +1.4 toward the Democratic Party.
+2.0 is well within the MoE.

For comparision purposes, the generic aggregate from 2010, with the polls in just about the same time frame, was GOP +8.2.

So, the GOP is 6 full aggregate points BEHIND where it was in 2010 and that end aggregate was off by more than 3 points to the Right. So, the generic may be actually tied-up right now.

The generic ballot still does not show a GOP wave forming, but the critical point is that the GOP is winning where it needs to win, which is exactly in line with 160 years of US mid-term election history:

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

This is all going to be a matter of voter interest and GOTV, plain and simple. And I bet that the control of the Senate will go into overtime in both Georgia and Louisiana. And to make matters worse, according to Georgia's calendar:

2014 elections and voter registration calendar

federal runoffs for Georgia are not in December, they are in JANUARY. The runoff date for the Senatorial, if I have read this correctly, is Tuesday, January 5th, 2015, after the next session of Congress has already begun. That is just crazy. The runoff in Louisiana is on December 2nd, 2014.

Those are the current stats.

-Stat

 

[Interpol]

Yup, it's all GOTV at this point for control of the Senate. Although it looks to be a down year for the Democrats, in the overall it's hard to ignore that they're more competitive in some places that have historically been easier for Republicans to win. It looks like the best McConnell can do is about 53-47. No one can say he's got the "Big Mo" in that race. In fact, it appears he is one of the candidates out there who isn't finishing strong.

Joni Ernst isn't finishing strong either. Jean Shaheen and Kay Hagan on the Democratic side are in the same boat as Ernst in that voters are a little more reluctant to say they support them when polled because they're incumbents, not the most popular thing to be this year.

Perdue isn't finishing strong either while Michelle Nunn seems to be starting to turn the tide in Georgia. That race and the one in Kansas appear to be dead heats.

Mark Begich has come out of nowhere in recent days to lead in a couple or three polls in Alaska.

Pryor needs a big turnout, but it looks like Arkansas is slipping away from the Democrats. Same too with Udall in Colorado and Landrieu in Louisiana, even if she wins the first vote.

Even though RCP now shows 45-45 with 10 undecided states, I'm going to say 49-45 for the GOP with 6 way-too-close-to-call states. GOP has to go 2-4 for control while Democrats have to go 5-1.

Harder for the Dems to keep control, but things look a lot tighter than a few weeks ago. No longer a GOP wave as much as it is a shift.

 

[Statistikhengst]

Yup, it's all GOTV at this point for control of the Senate. Although it looks to be a down year for the Democrats, in the overall it's hard to ignore that they're more competitive in some places that have historically been easier for Republicans to win. It looks like the best McConnell can do is about 53-47. No one can say he's got the "Big Mo" in that race. In fact, it appears he is one of the candidates out there who isn't finishing strong.

Joni Ernst isn't finishing strong either. Jean Shaheen and Kay Hagan on the Democratic side are in the same boat as Ernst in that voters are a little more reluctant to say they support them when polled because they're incumbents, not the most popular thing to be this year.

Perdue isn't finishing strong either while Michelle Nunn seems to be starting to turn the tide in Georgia. That race and the one in Kansas appear to be dead heats.

Mark Begich has come out of nowhere in recent days to lead in a couple or three polls in Alaska.

Pryor needs a big turnout, but it looks like Arkansas is slipping away from the Democrats. Same too with Udall in Colorado and Landrieu in Louisiana, even if she wins the first vote.

Even though RCP now shows 45-45 with 10 undecided states, I'm going to say 49-45 for the GOP with 6 way-too-close-to-call states. GOP has to go 2-4 for control while Democrats have to go 5-1.

Harder for the Dems to keep control, but things look a lot tighter than a few weeks ago. No longer a GOP wave as much as it is a shift.

Indeed. 3 key Democratic retirements are making some of this possible, + the GOP is winning in some states that default Republican in prez cycles. I would not place too much value on the polling for Colorado.

Gesendet von meinem GT-I9515 mit Tapatalk

 

[Statistikhengst]

So, in 3 days and a couple of hours, we Americans will start to go to the polls and elect a new US-Congress. In about 4 days, most of the results, but not all, will be in.

This is a quick in-between analysis, because I am quite sure there will be flurry of last minute polls both conducted and published this weekend.

On Monday, I will open up a 2014 results thread with links to all sorts of sources.


There are, for my way of thinking, four real horseraces: Georgia (Perdue vs. Nunn), Kansas (Roberts vs. Orman), North Carolina (Tillis vs. Hagan) and Iowa (Ernst vs. Braley)

I wrote two weeks ago that Gary Peters (D-MI) was going to leave the battlegrounds. His aggregate stood at +9 last week which was well inside the comfort zone. Four days ago, it stood at +11, it is now at +12.2 so the decision to remove him from the battlegrounds was a correct decision. The RNC has also pulled money out of MI, which means they have given up on it. Likewise, in South Dakota, Mike Rounds is now up to +12.2, so logically, SD is no longer in the analysis.

DEMOCRATS LEADING:
​

Aggregate, NH:

August 11, 2014: Shaheen +10.4
September 16, 2014: Shaheen +3.5
September 22, 2014: Shaheen +5.0
October 2, 2014: Shaheen +4.0
October 7, 2014: Shaheen +4.6
October 15, 2014: Shaheen +3.5
October 21, 2014: Shaheen +2.6
October 29, 2014: Shaheen +2.2
November 1, 2014: Shaheen +3.4

The needle has moved 1.2 back toward incumbent Shaheen, who has lead in 6 of the last 8 polls, and they are all well within the 2-week standard time-frame for making realistic aggregate values.

REPUBLICANS leading:
​

Aggregate, AK:

August 11, 2014: -no aggregate was possible-
September 16, 2014: Sullivan +1.3
September 22, 2014: Sullivan +1.3
October 2, 2014: Sullivan +4.7
October 7, 2014: Sullivan +4.7
October 15, 2014: Sullivan +4.4
October 21, 2014: Sullivan +4.4
October 29, 2014: Sullivan +2.2
November 1, 2014: Sullivan +2.6

There has been new polling in Alaska (finally) and the needle has moved +0.4 points toward Sullivan. It's still close, but Sullivan has a lean but consistent lead.

Aggregate, CO:

August 11, 2014: Udall +3.7
September 16, 2014: Udall +1.5
September 22, 2014: Udall +0.6
October 2, 2014: Gardner +1.5
October 7, 2014: Gardner +0.6
October 15, 2014: Gardner +1.4
October 21, 2014: Gardner +3.0
October 29, 2014: Gardner +3.3
November 1, 2014: Udall +3.6

The needle has moved again +0.3 points in Gardner's direction. In the postings from October 15th and from October 7th, I reminded how very off polling was in Colorado and in Nevada in 2010, also in 2012. Reason: gross miscalculation of the Latino vote. But on paper, Gardner has the lead. +3.6 is still just within the MoE.

Aggregate, KY:

August 11, 2014: McConnell +2.5
September 16, 2014: McConnell +5.2
September 22, 2014: McConnell +5.2
October 2, 2014: McConnell +5.3
October 7, 2014: McConnell +4.2
October 15, 2014: McConnell +3.0
October 21, 2014: McConnell +4.7
October 29, 2014: McConnell +4.4
November 1, 2014: McConnell +5.5

The needle has moved 1.1 points toward incumbent McConnell. He is winning in a ruby-red state that Mitt Romney carried by almost +23 point two years ago.

Aggregate, AR:

September 16, 2014: Cotton +2.5
September 22, 2014: Cotton +2.5
October 2, 2014: Cotton +3.6
October 7, 2014: Cotton +3.7
October 15, 2014: Cotton +4.4
October 21, 2014: Cotton +3.6
October 29, 2014: Cotton +5.0
November 1, 2014: Cotton +7.0

The needle has jumped +2.0 toward Republican challenger Tom Cotton and +7 is definitely outside the MoE, and now outside the battlegrounds. If these numbers hold, then Cotton will easily defeat Pryor (D-inc) in a ruby-red state that Mitt Romney carried by almost 24 points two years ago.

Double aggregate, LA:

Jungle Primary, October 2, 2014: Landrieu +1.2
Jungle Primary, October 7, 2014: Landrieu +2.7
Jungle Primary, October 15, 2014: Landrieu +2.7
Jungle Primary, October 21, 2014: Landrieu +2.8
Jungle Primary, October 29, 2014: Landrieu +4.2
Jungle Primary, November 1, 2014: Landrieu +4.2

Two-"man" race, possible runoff, October 2, 2014: Cassidy +4.6
Two-"man" race, possible runoff, October 7, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 15, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 21, 2014: Cassidy +5.3
Two-"man" race, possible runoff, October 29, 2014: Cassidy +4.5
Two-"man" race, possible runoff, October 29, 2014: Cassidy +4.5
Two-"man" race, possible runoff, November 1, 2014: Cassidy +4.5

No change since Wednesday. A runoff is absolutely going to happen in this state.

THE REAL DOGFIGHTS - DOWN TO THE WIRE:​

Aggregate, IA:

August 11, 2014: Ernst +0.8
September 16, 2014: Braley +1.4
September 22: Braley +0.1
October 2, 2014: Ernst +2.8
October 7, 2014: Ernst +1.9
October 15, 2014: Ernst +1.2
October 21, 2014: Ernst +2.5
October 29, 2014: Ernst +1.7
November 1, 2014: Ernst +1.2

The needle has moved 0.5 points toward Braley (D). Flip a coin.

Aggregate, NC:

August 11, 2014: Tillis +1.3
September 16, 2014: Hagan +3.7
September 22,2014: Hagan +5.0
October 2, 2014: Hagan +4.2
October 7, 2014: Hagan +3.7
October 15, 2014: Hagan +1.5
October 21, 2014: Hagan +1.2
October 29, 2014: Hagan +1.2
November 1, 2014: Hagan +1.6


After two weeks of the needle not budging at all, it has now moved +0.4 toward Hagan (D-inc). Within six weeks time, Hagan's aggregate has shrunk from +5 (right at the outer edge of the true battleground zone) to a mere +1.6.Dogfight. Flip a coin.

Aggregate, GA:

August 11, 2014: Perdue +3.2
September 16, 2014: Perdue +3.0
September 22, 2014: Perdue +3.3
October 2, 2014: Perdue +3.4
October 7, 2014: Perdue +3.1
October 15, 2014: Perdue +2.7
October 21, 2014: Perdue +0.6
October 29, 2014: Perdue +0.5
October 29, 2014: Perdue +0.5

The needle has moved agin 0.1 points toward Michelle Nunn (D), which is nothing more than statistical noise. This race has stood still over over week and every single one of those polls are new polls.

This makes this race one of the two closest aggregates at this time and the largest overall shift over last two weeks. The problem is, with a third party candidate likely to score at least 2 points or more, a +.5 to +1 win for either major candidate means that no one gets over 50%, which means that Georgia, like Louisiana, is very likely to go into overtime. The polls close in all of Georgia at 7 PM, EDT and I am pretty sure that this race will not be called for many hours, if not for a day.

Flip a coin.

Aggregate, KS:
September 22, 2014: Orman +1.2
October 2, 2014: Orman +5.3
October 7, 2014: Orman +5.3
October 15, 2014: mathematical tie
October 21, 2014: mathematical tie
October 29, 2014: Orman +0.9
November 1, 2014: Orman +1.0

The needle has moved 0.1 toward Orman (I) - statistical noise - putting him close to where he was 5 weeks ago, but nowhere near his high-water mark from October 7th. Flip a coin.

FACIT: in the last three days, we have seen races tighten up or stay exquisitely tight in: Alaska, Iowa, North Carolina, Georgia and Kansas.

We have seen the races open or close very slightly in: Colorado.

We have seen the races open in a statistically important way in: New Hampshire, Arkansas and Kentucky.

In the Generic ballot:

Generic aggregate, October 2, 2014: GOP +2.9
Generic aggregate, October 7, 2014: GOP +2.1
Generic aggregate, October 15, 2014: GOP +2.4
Generic aggregate, October 21, 2014: GOP +3.4
Generic aggregate, October 29, 2014: GOP +2.0
Generic aggregate, November 1, 2014: GOP +2.5

Shift: +0.5 toward the GOP.
+2.5 is well within the MoE.


Compared to the generic from 4 years ago:

Right before election day 2010, the generic aggregate stood at: GOP +9.4

Now, election day 2010 was on November 2nd, not on November 4th, and RCP's final calculation is only for the last week of polls. If we extend that to the final two weeks but eliminate all polls conducted at an end day of 10/30/2010 (3 days before the election, which would be today in this cycle, and those polls have not yet been published in this cycle), then the aggregate would be GOP +8.3.

Right now, the generic aggregate shows the GOP 5.8 points behind 4 years ago.


So, the changes over the last three days only confirm movement we expected to see in two of the deepest of red states: Kentucky and Arkansas. It also confirms that the race is still a pure dogfight in Georgia, Kansas, Iowa and North Carolina.

The GOP is still winning, but still, no sign of a wave.

BTW, YouGov is predicting on the HOR level that the GOP will pick up between 6-12 seats. I have been predicting all year long about 11 seats.

Final analysis on Tuesday afternoon, and some individual polls may be analysed in between.

 

[Statistikhengst]

So, tomorrow, Americans will go to the polls to elect a new congress. Actually, assuming a 40% voter turnout from 184 million registered voters, that means that about 73.6 million ballots will be cast, and of those ballots, 17.1 (or about 23%) million or so have ALREADY been cast, per early voting:

Before comparing the aggregates to the last days and weeks, I want to provide what is called a "funnel-view", based on the aggregates from RealClearPolitics (a Right-leaning sight, but like HuffPo and TPM, does a very good job of collecting the numbers most of the time):

Way outside of the battlegrounds:
SD*: Rounds (R) +12.2
MI*: Peters (D) +12.2
(those two were in battleground-consideration for a while, now no longer)

Outside of the battlegrounds, likely to be non-competitive:
AR*: Cotton (R) +7.1
KY: McConnell (R-inc) +6.5
LA*: Landrieu (D-inc) +5.5 (LA "Jungle-Primary)

Inside the battlegrounds, outside the MoE:
CO*: Gardner (R) +3.9

Inside the battlegrounds, within the MoE, directly outside the tossops (+2):
AK*: Sullivan (R) +2.4

Tossups (less than +2):
IA*: Ernst (R) +1.8
GA*: Perdue (R) +1.8
NC*: Hagan (D-inc) +1.6
NH: Shaheen (D-inc) +1.4
KS*: Orman (I) +0.7

*states with early voting. See: map

Here is a link to that map:

Absentee and Early Voting

That map is clickable so that you can see the exact parameters for states that do early voting. This is important, for in races that have tightened up on the last week in states with early voting, that may be of little help to the losing side, for the winning side may have "banked" enough votes in order to get over the finish-line. This applies to BOTH parties and also to Independent candidate Orman (KS). So, of the states that many have been tracking as battlegrounds over the last month, only New Hampshire and Kentucky do not have early voting of any kind at all.


I am also organizing the categories differently this time:

Outside of the battlegrounds.

Within the battlegrounds.

Within the MoE.

True-tossups.

Likely to go into overtime:

Outside of the BATTLEGROUNDS:​

Aggregate, AR:

September 16, 2014: Cotton +2.5
September 22, 2014: Cotton +2.5
October 2, 2014: Cotton +3.6
October 7, 2014: Cotton +3.7
October 15, 2014: Cotton +4.4
October 21, 2014: Cotton +3.6
October 29, 2014: Cotton +5.0
November 1, 2014: Cotton +7.0
November 3, 2014: Cotton +7.1

The needle has jumped +0.1 toward Cotton (R) - this is statistical noise. Cotton is going to pick-up Arkansas for the GOP and AR will likely be the first state to be called as a loss for a Democratic incumbent on election night. The polls close at 8:30 PM EDT (07:30 CDT) and I suspect that within one hour after poll closings, the race will be called.

Aggregate, KY:

August 11, 2014: McConnell +2.5
September 16, 2014: McConnell +5.2
September 22, 2014: McConnell +5.2
October 2, 2014: McConnell +5.3
October 7, 2014: McConnell +4.2
October 15, 2014: McConnell +3.0
October 21, 2014: McConnell +4.7
October 29, 2014: McConnell +4.4
November 1, 2014: McConnell +5.5
November 3, 2014: McConnell +6.5

The needle has moved 1 point toward incumbent McConnell over this weekend, and this is statisticall important. Mitch McConnell (R-inc) is going to retain his seat and what started as a very close, competitive contest has opened up to a comfortable lead for him. In fact, I suspect that his end-margin will surpass his 2008 margin. Some of the polls in Kentucky close at 6:30 PM EDT, the rest close at 7:00 PM EDT (6:00 PM CDT) on election night, and if this aggregate holds, the networks should be able to call this race by 08:30 PM at the latest. They will probably wait until enough of the Louisville vote is in to confirm that the votes out of this Democratic stronghold would not be able to overcome a McConnell lead elsewhere.

Double aggregate, LA:

Jungle Primary, October 2, 2014: Landrieu +1.2
Jungle Primary, October 7, 2014: Landrieu +2.7
Jungle Primary, October 15, 2014: Landrieu +2.7
Jungle Primary, October 21, 2014: Landrieu +2.8
Jungle Primary, October 29, 2014: Landrieu +4.2
Jungle Primary, November 1, 2014: Landrieu +4.2
Jungle Primary, November 3, 2014: Landrieu +5.5

Two-"man" race, possible runoff, October 2, 2014: Cassidy +4.6
Two-"man" race, possible runoff, October 7, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 15, 2014: Cassidy +5.6
Two-"man" race, possible runoff, October 21, 2014: Cassidy +5.3
Two-"man" race, possible runoff, October 29, 2014: Cassidy +4.5
Two-"man" race, possible runoff, October 29, 2014: Cassidy +4.5
Two-"man" race, possible runoff, November 1, 2014: Cassidy +4.5
Two-"man" race, possible runoff, November 1, 2014: Cassidy +4.6

So, an interesting thing is happening in Louisiana: Democratic incumbent Landrieu's margin is increasing, which means it is extremely likely that she wins the Jungle-Primary tomorrow night. The bad news is that 4 of the last 5 polls show third party Maness (R) with double digits (11, 13, 15, 15), so there is absolutely no way that Landrieu is going to get to 50%. I can already now call overtime for this race and then it is a matter of motivating enough voters to come out and vote a second time for the US-Senate. Maness is guaranteed to lose and the assumption is that the R vote that went for him will likely shift to Cassidy in the runoff. The polls close in all of Louisana at 9:00 PM EDT (8:00 PM local time) and I assume this race will be called within two hours. A +5.5 lead is pretty easy to determine. The stations will wait long enough to see what kind of margin Landrieu is getting out of New Orleans in order to call the race overall.

Traditionally, third party candidates' percentages tend to be lower in the actual count as compared to the poll aggregates, but in this case, Maness' percentage has climbed so dramatically, I am not sure that his will drop. He really COULD hit 15%.

Within the battlegrounds:​

Aggregate, CO:

August 11, 2014: Udall +3.7
September 16, 2014: Udall +1.5
September 22, 2014: Udall +0.6
October 2, 2014: Gardner +1.5
October 7, 2014: Gardner +0.6
October 15, 2014: Gardner +1.4
October 21, 2014: Gardner +3.0
October 29, 2014: Gardner +3.3
November 1, 2014: Gardner +3.6
Novermber 3, 2014: Gardner +3.9

The needle has moved again +0.3 points in Gardner's direction, for the third time in a row. Just as reminder that polling in the SW has been very, very off in the last 4 election cycles, we will really need to wait and see how this truly shapes up. The polls close in CO at 9:00 EDT (7:00 PM MDT) and I don't expect this race to be called for at least 3 hours, if not considerably longer. Hunker down for a long count in this race.

Within the MoE:​

Within the MoE, but not yet a tossup:

Aggregate, AK:

August 11, 2014: -no aggregate was possible-
September 16, 2014: Sullivan +1.3
September 22, 2014: Sullivan +1.3
October 2, 2014: Sullivan +4.7
October 7, 2014: Sullivan +4.7
October 15, 2014: Sullivan +4.4
October 21, 2014: Sullivan +4.4
October 29, 2014: Sullivan +2.2
November 1, 2014: Sullivan +2.6
November 3, 2014: Sullivan +2.4

There has been new polling in Alaska the needle has moved +0.2 points toward Begich, which is likely statistical noise. It's still close, but Sullivan has a lean but consistent lead. However, the Democrats are claiming that they have visited virtually every single Inuit village to campaign for votes, and those Inuits don't have telephone lines or internet access, so, wait and see. The polls close in AK at 01:00 AM EDT on November 5th and I doubt this race will be called anytime until late on Wednesday.

THE TOSSUPS (5):​

Aggregate, IA:

August 11, 2014: Ernst +0.8
September 16, 2014: Braley +1.4
September 22: Braley +0.1
October 2, 2014: Ernst +2.8
October 7, 2014: Ernst +1.9
October 15, 2014: Ernst +1.2
October 21, 2014: Ernst +2.5
October 29, 2014: Ernst +1.7
November 1, 2014: Ernst +1.2
November 3, 2014: Ernst +1.8

The needle has moved 0.6 back toward Ernst, a reversal from two days ago. Flip a coin. That being said, Ernst has now led in the aggregate consistently for one month. Neither candidate has put this race away, it is purely a matter of GOTV. Same thing for Georgia. The polls close at 10 PM EDT (9 PM CDT) in Iowa and I am sure this race will not be called for hours, if not a day or more.

Aggregate, GA:

August 11, 2014: Perdue +3.2
September 16, 2014: Perdue +3.0
September 22, 2014: Perdue +3.3
October 2, 2014: Perdue +3.4
October 7, 2014: Perdue +3.1
October 15, 2014: Perdue +2.7
October 21, 2014: Perdue +0.6
October 29, 2014: Perdue +0.5
November 1, 2014: Perdue +0.5
November 3, 2014: Perdue +1.8

The needle has moved 1.3 points toward Perdue, however, for weeks, we have seen a Leap-frog of polls for the opposing sides: a week for Perdue, a week for Nunn. With a Libertarian on the ballot, it is unlikely that either Perdue or Nunn get to 50%, so as is the case with Louisiana: overtime. The polls close in the Peach state at 7:00 PM EDT and this race will surely not be called until the wee hours of the morning.

Aggregate, NC:

August 11, 2014: Tillis +1.3
September 16, 2014: Hagan +3.7
September 22,2014: Hagan +5.0
October 2, 2014: Hagan +4.2
October 7, 2014: Hagan +3.7
October 15, 2014: Hagan +1.5
October 21, 2014: Hagan +1.2
October 29, 2014: Hagan +1.2
November 1, 2014: Hagan +1.6
November 3, 2014: Hagan +1.6

No change over the weekend. Dogfight. Flip a coin. With such a narrow aggregate, early voting probably doesn't help either side in this case. The polls close in NC at 7:30 PM EDT and I suspect this race will not be called until 3 or 4 AM on November 5th, maybe even days later.

Aggregate, NH:

August 11, 2014: Shaheen +10.4
September 16, 2014: Shaheen +3.5
September 22, 2014: Shaheen +5.0
October 2, 2014: Shaheen +4.0
October 7, 2014: Shaheen +4.6
October 15, 2014: Shaheen +3.5
October 21, 2014: Shaheen +2.6
October 29, 2014: Shaheen +2.2
November 1, 2014: Shaheen +3.4
November 3, 2014: Shaheen +1.4

The needle has moved 2.2 points to Republican challenger Brown, making this race, statistically, a sudden dogfight. That being said, Rasmussen, a statistically proven Right-leaning pollster, just put out a poll showing Shaheen up by +7 and over the 50 mark, even at 52, which Rasmussen rarely, if ever, does with Democratic incumbents (see: Obama in Ohio, Wisconsin, Iowa & Colorado in 2012, Harry Reid in NV 2010, Patty Murray in Washington 2010, etc, etc, etc). Again, that being said, reduce the time-frame for this race down to just one week (five polls instead of 7), then the margin is Shaheen +1. In 11 weeks, Shaheen has lost 10 points on her aggregate and gone from being a pretty safe D-win to a true dogfight. Flip a coin. Scott Brown (R) could conceivably pull out the upset of the night here, for New Hampshire does not do early voting, so Shaheen has had no opportunity to "bank" early votes. The polls close in NH at 8 PM EDT and I suspect this race will not be called until 3 or 4 am on November 5th.

Aggregate, KS:

September 22, 2014: Orman +1.2
October 2, 2014: Orman +5.3
October 7, 2014: Orman +5.3
October 15, 2014: mathematical tie
October 21, 2014: mathematical tie
October 29, 2014: Orman +0.9
November 1, 2014: Orman +1.0
November 3, 2014: Orman +0.7

The needle has moved 0.3 toward Roberts (R) - statistical noise, but what I do not understand is why the latest Rasmussen poll, from 10/23/2014, showing Orman up +5:

Election 2014 Kansas Senate - Rasmussen Reports trade

has already been pulled out of the statistic. RCP has polls in the end statistic for other states that are as old as 10/12, so why remove this poll? This is very, very inconsistent and frankly, suspect. When you recalculate that aggregate for a two week time-frame, with Rasmussen, which would be consistent, then the aggregate is:

Orman +1.6.

I just want to make it very clear the RCP, a Right-leaning sight, deliberately removed a poll that is not yet 14 days old (as of Wednesday, it should be removed from the statistic for a two-week time frame, but it was already removed on Sunday) because the result did not jive with what they want to see. Take a look back at Arkansas, Kentucky, Lousiana, Colorado, Alaska and Iowa. In all of those aggregates, you will see polls older than 10/23. So, essentially, I just caught RCP red-handed and have screenshots to prove it for the future.

That being said, if you limit the time frame to one week, then it is an absolute tie

That's what HuffPo shows. But RCP has decided to arbitrarily (and I believe, deliberately) pick a time frame BETWEEN one and two weeks, completely inconsistent with it's other aggregates. That is BAD methodology.

Orman's saving grace is that early voting has been underway in KS since last Tuesday. Most polls close in KS at 8:00 PM EDT (7:00 PM local time) but some on the western border to Colorado close at 9:00 PM EDT and so the state will not be releasing any numbers until 9 PM EDT. As is the case with NH, GA, NC and IA, I suspect that this race will not be counted until the wee hours of November 5th, or maybe days later.

OVERTIME:​

Already listed above, both Georgia and Louisiana are very, very likely to go into overtime. In fact, Louisiana is guaranteed to go into overtime.


----------------------------------------------

FACIT: over the weekend, we have seen races tighten up or stay exquisitely tight in: Alaska, Iowa, North Carolina, New Hampshire, Georgia and Kansas.

We have seen the races open or close very slightly in: Colorado and Arkansas.

We have seen the races open in a statistically important way in: Kentucky.

In the Generic ballot:

Generic aggregate, October 2, 2014: GOP +2.9
Generic aggregate, October 7, 2014: GOP +2.1
Generic aggregate, October 15, 2014: GOP +2.4
Generic aggregate, October 21, 2014: GOP +3.4
Generic aggregate, October 29, 2014: GOP +2.0
Generic aggregate, November 1, 2014: GOP +2.5
Generic aggregate, November 3, 2014: GOP +2.2

Shift: +0.3 toward the Democratic party.
+2.2 is well within the MoE.

Here the generic from 2010:

The current generic is 7.2 points UNDER the GOP lead from 2010, and that lead was 3 points off to the Right (the statistic here at RCP is FALSE, the GOP won the House with +6.1 and the Senate with +5.7 in 2010), so a +2.2 GOP generic can just as well mean that the Democrats may actually WIN in the NPV on Tuesday.

The GOP is still winning where it needs to, but still, no sign of a wave.

On election night, MT, SD and WV will probably be called immediately for the GOP. Those are all open seats in deep RED states where the Democratic Senator is retiring.

That brings the GOP to 48. However, the race in KS will surely be undecided for hours, so this brings the GOP to 47, actually.

The races in KY and AR should be called pretty quickly, for incumbent McConnell (R) and challenger Cotton (R), making Arkansas the first GOP pick-up of the night. That brings the GOP to 49.

As of 10 PM EDT, that should be the state of affairs on election night. The polls will have been closed for 2.5 hours in NC, 2 hours in GA and NH, 1 hour in CO and KS and they will have just closed in IA. The polls will close 3 hours later in AK. I STRONGLY suspect that few key races will be called between 10-11 PM EDT, apart from Louisiana, which will then go into overtime, anyway. This means that the GOP stays at 49 for the time being.


Five of the six races mentioned are officially true dogfights - no one with any sense in his head should be making 100% calls right now. And CO, as I have pointed out repeatedly, may end up being a surprise. Either CO or AK may bring the GOP to 50 as of 2 AM, but that's pretty much that.

The rest is a crap-shoot.

The real bad news of the weekend is for Shaheen (D) in NH. Scott Brown (R) really COULD flip this race.

The other real bad news of the wekeend is for the GOP in Kansas. On the eve of the election, an Independent is still ahead in the Senatorial and a Democrat is decidedly ahead in the Gubernatorial. I suspect that bedrock RED Kansas is about to experience a historic tectonic shift in it's politics.

Either way, when all is said and done, the statistical probability that the GOP gets to 51 seats is around 90% now. It may take days or weeks to get there, but the GOP will get there and President Obama will join the ranks of Eisehower, Nixon, Ford, Reagan, Clinton, Bush 41 and Bush 43 - of having a complete opposition Congress.

Final report tomorrow at around 5:30 PM to 6:00 PM EDT.

-Stat

 

[WelfareQueen]

It looks the GOP will control the Senate 52-48. I see the Dems holding on to NC and NH but they lose Colorado and Iowa. All the other close races go to the GOP except Kansas. There...I have no clue.

Orman has promised he'll caucus with whichever side wins (i.e. GOP or Dems). The Dems seem to think he'll automatically caucus with them....I would say if Orman is being truthful, that remains to be seen.

Latest Nate Silver prediction is GOP control Senate 75% probability.

 

[Mindful]

I'm following these events with interest, but difficult to keep up, due to travelling. I'm in Greece right now, the U.S. next week.

When is the actual voting happening? Thanks.

 

[Derideo_Te]

I'm following these events with interest, but difficult to keep up, due to travelling. I'm in Greece right now, the U.S. next week.

When is the actual voting happening? Thanks.

Right now!

 

[Statistikhengst]

So, voting is underway in this great land of ours. Here are the end-polling statistics, as provided in aggregate by RealClearPolitics (RCP).

As I did two days ago, first, the funnel-view:

Way outside of the battlegrounds:
SD: Rounds (R) +12.2
MI: Peters (D) +12.8 (+0.6 over November 2)
(those two were in battleground-consideration for a while, now no longer)


Tonight's marquee races:
KY: McConnell (R-inc) +7.2 (+0.7 over November 2)
AR: Cotton (R) +7.0 (-0.1 over November 2)
LA: Landrieu (D-inc) +5.7 (+0.2 over November 2)
CO: Gardner (R) +2.5 (-1.4 over November 2)
AK: Sullivan (R) +2.4 (unchanged over November 2)
GA: Perdue (R) +2.8 (+1.0 over November 2)
IA: Ernst (R) +2.3 (+0.5 over November 2)
NC: Hagan (D-inc) +1.2 (-0.4 over November 2)
NH: Shaheen (D-inc) +0.8 (-0.6 over November 2)
KS: Orman (I) +0.8 (+0.1 over November 2)


And here the end-polling graphics from RCP (also in funnel-view order):

No need for a huge analysis anymore, the people are voting even right now. But there was lots of information and predictions in the analysis from November 2, 2014:

2014 battle for control of the US Senate Page 10 US Message Board - Political Discussion Forum


In the Generic ballot, the end-statistic is: GOP +2.4

That is 6.9 points UNDER the GOP generic aggregate value from the 2010 mid-term elections.


Two final points and then I'm done. On to watch the returns come in soon.

One single poll can be wrong. It's always possible. But a slew of polls all showing the same result, from independent-from-each-other pollster, cannot all be wrong. For all of the races where the aggregate spread is +2.5 and above, every single poll went for the candidate that we assume is going to win.

Anything from +2.4 down to +0.8 is a crap-shoot, a coin-flip. That aggregate is so within the standard MoE, anything can happen. That being said, Ernst (R), Sullivan (R) and Perdue (R) all hold the better cards in their hands.

The big shift from Sunday is the downward shift for Gardner in CO, and I have already written many, many times over that polling in the SW is notoriously off. Were a real surprise to happen tonight, it may very well be in Colorado.

Prepare for a long count, folks.