Sunday, October 28, 2012

Late October Polling Update - Senate Races

The control of the senate will come down to the following 13 races, and Democrats start out with a pretty high baseline of 44 seats since Debbie Stabenow in Michigan and Martin Heinrich in New Mexico should roll to easy victories.

Florida (Nelson +8) (based on 11 polls) - The median has remained constant all week, and the three polls releasing senate numbers this week showed Nelson beating Mack by 4 to 8 points.  Mid-September was the last time a poll showed Mack winning, and that poll was an outlier by Gravis Marketing.


Ohio (Brown +7.5) (based on 12 polls) - All the polling attention in Ohio has been lavished on the presidential contest, mostly because Sherrod Brown pretty much has this thing locked up.  "Only" five polls were released this week with senate numbers, and none of them showed a lead for Josh Mandel; Survey USA's 1-pointer from Wednesday seems to be an outlier. 


Pennsylvania (Casey +6) (based on 9 polls) - One would expect polls' margins to be evenly scattered across a spectrum with outliers occupying both margins.  This is still not the case in Pennsylvania, where most polls have Republican Tom Smith's support at or below 42%, except that Quinnipiac, Rasmussen and Susquehanna have his support at 45% or above.  At least one group will be wrong on election day.

With the pretty sure bets in Florida, Ohio and Pennsylvania, Democrats will have 47 seats.  

Massachusetts (Warren +4.5) (based on 8 polls) - Three new polls this week put Elizabeth Warren's support at an average of 50%, and it seems like she's starting to pull away.  The median poll spread last week was only 3 points.  Democratic seat number 48.


Connecticut (Murphy +3) (based on 7 polls) - There were four new polls in Connecticut this week: the two Republican-leaning polls said Murphy was basically tied with McMahon, while the two Democratic-leaning polls said Murphy was basically up by 5.  One of those Democratic-leaning polls was a Quinnipiac poll (tilts D+1.4 in the presidential race)  that had McMahon up by 1 on October 4th.  Their latest shows a seven-point swing towards Murphy.  This should be Democratic seat number 49.


Missouri (McCaskill +3) (based on 4 polls) - Right now the spread in Missouri is the same as in Connecticut, but the demographics of the state are light-years apart, and the trends are going in opposite directions.  The new Mason-Dixon poll showing McCaskill only up 2 drove the poll spread down 3 points by itself.  A note of caution: Missouri is pretty thinly polled.  Right now this is Democratic seat number 50, which would give Democrats control assuming Barack Obama is reelected.


Wisconsin (Baldwin +2) (based on 7 polls) - Only two polls came out this week, and they both showed a race within the margins of error.  The median poll spread for Baldwin dropped half a point this week.  If Baldwin can hold on, this is the Romney-proof 51st Democratic senate seat.


Virginia (Kaine +2) (based on 8 polls) - Three polls in Virginia bothered to survey this blockbuster senate race this week.  Two of them were Rasmussen and Gravis, so the median poll spread for Kaine dropped half a point this week.  The Washington Post poll showing Kaine up by 7 is an outlier from one of the most Democratic-leaning outlets (D+2.6 in the presidential race).  If Kaine can hold on, this is the Romney-and-Angus-King-proof 52nd Democratic senate seat.


Arizona (Carmona +1) (based on 3 polls) - Some other firm needs to poll this race, because we have polls from PPP and the weird outlier from a local firm defining this contest as a Carmona win.  I am skeptical.  This is the final race showing a Democratic lead currently in the median spread, the 53rd senate seat, a Romney-proof, Angus-King-proof, Joe-Manchin-proof Democratic majority.


Montana (tied) (based on 4 polls) - Another state in need of more polls, the only non-partisan poll that came out last week was from Pharos Research Group, which showed Democrat Jon Tester up two points.  Right now Tester is only being held back by a Mason-Dixon poll from September or else he'd be the 54th Democratic seat.  


Of the 13 senate battlegrounds, Republicans are currently ahead in only 3 of them.

Nevada (Heller +1) (based on 6 polls) - Three new polls this week, and two showed a tightening of the senate race in Nevada; the median spread slipped from Heller +5 to Heller +1.  


North Dakota (Berg +3) (based on 4 polls) - Despite the addition of Pharos Research Group to the mix, this race could still use more polling.  


Indiana (Mourdock +4) (based on 3 polls) - No poll has been taken since the Republican Mourdock's theological views on rape came to light, but as his view relies more on faith than junk science like Todd Akin's, it might not matter to Indianans. 

Late October Polling Update - Presidential Contest

Mitt Romney's momentum has stopped, speaketh Lord Silver from his throne inside the Castle of the Gray Lady.  And the high lord of poll analysis was right -- it has been a pretty good polling week for Barack Obama overall.  The following graphs show the median of the active independent polls in each of the presidential battleground states.  I'll do the senate races in a separate post.

The presidential election will come down to these 11 states.  Last week I ran down the numbers from the perspective of Mitt Romney, so in the interest of fairness I will do it from Barack Obama's perspective this time.  His baseline is 201 electoral votes.

Michigan (O+5) (based on 7 active polls) - Michigan remains unchanged from last week, as only one (outlier) poll by Baydoun/Foster was done this week.  That poll showed an improbable tie in Michigan, but the pollster has a 3.2 point Republican-leaning house effect, second in magnitude only to Susquehanna in my database.


Pennsylvania (O+4.5) (based on 10 active polls) - Pennsylvania remains mostly aspirational for Mitt Romney.  Five polls released this week in Pennsylvania ranged from Obama +3 to Obama +6.  With Michigan and Pennsylvania looking like they're in the bag for Obama, his electoral vote total becomes 237.


Ohio (O+4) (based on 17 active polls) - There are varying definitions for swing state, but they all include this one.  Nate Silver's computers say that Ohio has a 50-50 chance of deciding the election, which basically means it's as important as all the other states combined.  This is great news for Obama.  

Nine polls were published about Ohio this week, and Mitt Romney wasn't leading in any of them.  The last two polls that had showed a Romney lead came from ARG (R +2.3) and Gravis Marketing (R +3.1); both firms have done more recent polls showing either a tie (Gravis) or a 2 point Obama lead (ARG).  Adding 18 more electoral votes to the total, Obama has 255, 15 to go.

Wisconsin (O+4) (based on 7 active polls) - It would seem that the presidential race in Ohio would be closer than Wisconsin or Iowa judging by all the polling done in the Buckeye State, but all three of them show similar numbers.  Three polls were released in Wisconsin this week.  The latest one released on Friday showed a tie, but it was a Rasmussen poll (R +2.6).


Iowa (O+3) (based on 7 active polls) - Gravis Marketing (R +3.1) and PPP (D +0.5) apparently switched names just for their Iowa polling this week.  PPP showed Romney up 1 point last weekend, while Gravis just released a poll showing Obama up 4, which was only their third poll out of 23 in my database to show a spread bluer than the median.  With Wisconsin and Iowa, Obama would have 271 electoral votes and a second term. But he'll win at least one more state.


Nevada (O+3) (based on 8 active polls) - No 3-point margin has ever felt safer for Obama than his lead in Nevada.  Five new polls this week all showed slender leads for Obama, and it's a lead that just makes sense given the demographics of the state and the recent election history.  With Nevada, Obama would have 277 electoral votes.


New Hampshire (O+1) (based on 7 active polls) - New Hampshire had four new polls this week: two showed a change in Obama's favor from previous polls, one showed a change in Romney's favor from a previous poll, and one was a new entry that showed a 3-point Obama win.  If Obama were to pull off a victory in this Democratic-leaning state, he would have 281 electoral votes.


Virginia (O+0.5) (based on 12 active polls) - Just this evening Virginia tilted to Obama for the first time in 9 days based not only on the 4-point Obama margin in the Washington Post poll but also on the latest poll from Gravis Marketing exterminating an older poll showing Romney up five. Virginia is the last state I show Obama winning as of right now, and it would give him 294 electoral votes.

North Carolina (R+0.5) (based on 8 active polls) - North Carolina suffers from partisan polling divergence, similar to the Pennsylvania senate race.  You can see from the chart that there are two pollsters (Rasmussen and Gravis) that show Romney's support well over 50% in the Tarheel State.  ARG showed Romney's support back in September right at 50%, and there was one Civitas poll that had Romney's support at 53% in early September (a new Civitas poll put it at 48% this week).  The one firm that does not have a Republican bias that has polled North Carolina at all this month is North Carolina-based PPP, which has this race tied.  Yet most other poll aggregators show North Carolina as Romney country, with Obama having a better shot at winning Florida than at winning North Carolina.  I don't agree, and I wish we could see some updated polling numbers from Marist and Survey USA that could back this up.  It may be moot to the electoral college, but I think Romney will win North Carolina by less than 2 points if he wins it at all, and it certainly less than the 6-point and 8-point margins Rasmussen and Gravis are showing.


Colorado (R+1) (based on 10 active polls) - Four new polls came out this week, and they were literally all over the place.  Rasmussen had Romney by 4, but PPP had Obama by 4.  The trend in three of them though favored Romney, while the fourth (PPP) only barely favored Obama (+3 to +4).  Therefore Colorado moved from a tie last week to Romney +1 this week.

Florida (R+1) (based on 15 active polls) - Five new polls came out this week, and Obama was ahead in none of them (though he did manage a tie in the Pharos Research Group poll if you round to the nearest whole percent).  One glimmer of hope for Obama is that the latest poll that came out (Rasmussen) showed a three point swing from the week before towards Obama.  If Obama wins all these 11 battleground states, he will have 347 electoral votes.

After Florida, there is a steep jump to the next nearest state (Arizona, R+8), so I feel pretty confident that the universe of potential Obama victories has a rigid ceiling at 347 electoral votes.  

I will have more on the Senate races tomorrow. 

Thursday, October 25, 2012

Do Polls Drive the Narrative, or Does the Narrative Drive the Narrative?

We're at the stage in election coverage where everything has now been said, but articles have to still keep being written by journalists.  Now there is an uptick in meta-campaign articles, not about how the media portrays horse-race campaign coverage, but about how the media portrays how the media portrays horse-race campaign coverage. 

The flagship article for this meta trend is this Alec MacGillis article from The New Republic. He argues that the media has been responsible for this illusion of the presidential race tightening because it provides a more compelling narrative, ignoring evidence that would indicate the race isn't actually that close.

Over at the Washington Post, Chris Cillizza tweeted after the debate: “Romney win in first debate still more meaningful for arc of race than Obama win in third debate.” And yes, if we declare it less meaningful for the “arc”—another word for story!—then of course it will be so. We liberal reporters do love our tautologies, even if we’re not always aware of them.

The problem with MacGillis's anti-narrative narrative is that there was actually a substantial and sustained drop in the polls that began roughly around the night of that first debate.  And I'm not talking about national tracking polls like Gallup's weird presidential tracker-of-fleeting-passions-of-likely-voters.  I mean a drop in the state polls, the only polls that matter.

This is a chart of the average margin in polls in the nine closest battleground states, as determined by my ongoing research into charting the median active credible poll in each state.  The nine battleground states are Florida, Virginia, North Carolina, New Hampshire, Ohio, Wisconsin, Iowa, Colorado and Nevada.  Positive values indicate an Obama lead, negative values indicate a Romney lead.

Obama crested around October 3rd, the night of the first debate, a time when most poll averages had Florida, Colorado and Virginia in the Obama column.  The numbers kept plummeting after the second debate, but may be leveling off now, though there has been no post-third-debate polling as of yet.

This isn't an artist's rendering of a media-driven campaign narrative.  Based on those polls, here's what the electoral college looks like:

Some have indicated that this Obama drop may have nothing at all to do with any debate or really any campaign event whatsoever, and that it merely represents conservatives finally rallying around their presidential candidate.  There are also liberal poll Unskewers who believe that the polling firms themselves, led by arch-conservative pollster Scott Rasmussen, are forging poll numbers in order to drive the numbers-based narrative pictured above. 

If this was the case, I would have expected the same trend to be present in the senate polls.  If conservatives are rallying around their presidential candidate, they should also be driving their senate to higher polling margins (after all, since the vast majority of senate polls are conducted at the same time as presidential polls, they're sampling the exact same people).  Or if there are poll-driven Republican narratives, the Rasmussens of the world should be able to drive the composition of the senate rightward too.  But this hasn't the case.

Here's the same chart as above, except with D-R senate margins.

Democrats are currently doing about the same as they were on October 3rd: kicking ass overall in the 14 closest races.  Because of the polls, the senate projections basically all look something like this:

Why the stark difference in polling "trajectory" (to use Alec MacGillis's favorite word) between the presidential contest and the senate contests ?  The most logical conclusion is a nationally televised debate that showed fickle center-right undecideds that Mitt Romney wasn't the scary/bumbling caricature they'd heard fleeting rumors about.

Wednesday, October 24, 2012

Cartogram 2012 - Reapportionment Madness!

This is the 2012 version of this thing I made in 2008. Now with even more abstract state shapes!  Why does Florida muffin-top out to Louisiana?  How did Maryland become an object in a Picasso painting? Why does New York look like a syringe? (click on the picture for full-size)

Sunday, October 21, 2012

Mid-October Polling Update

I have continued tracking and charting all of the polls in each state and deeming the winner of each state as determined by the median active credible poll.  As of today, these are the thirteen closest races that will determine the presidency.  I have presented them in descending order of likelihood of a  Mitt Romney victory.
(Jump to the senate graphs!)
Mitt Romney's baseline is 170 electoral votes.  He needs to find 100 more to win. 

Arizona (R+8) - The wild Hispanic voting bloc has long been rumored to live in this state.  Many Obama supporters claimed to have spotted it, but it remains elusive in shaping elections.

Missouri (R+6) - Obama was never expected to do well in this former bellwether state, and he isn't, though Rasmussen's recent poll showing Romney up by 11 is a bit of an outlier.


Florida (R+2) - On October 3rd, the night of the first debate, the median poll spread showed Barack Obama up by three points.  Today the median spread has him down two.  Florida experienced one of the largest post-first-debate swings (5 points).  It is interesting to me that Florida has been so competitive given the fact that it has more retirees, more Republican-leaning Latinos of Cuban descent, and more Southerners than the typical swing state.  I will be surprised if Barack Obama can win the critical I-4 corridor.

North Carolina (R+0.5) - The median poll shows North Carolina still in play. This is because the median uses some polls from September, during Obama's pre-debate surge.  Polls in October show an increasing Romney lead.

With these states and all the other ones that are not competitive, Mitt Romney has 235 electoral votes, only 35 away from the presidency. 

Virginia (tied) - Yesterday the median showed a one-point margin for Mitt Romney, though a PPP poll this morning moved the median to a tie.  On October 2, the spread was Obama +5. Since then, Romney has surged and Obama has slipped.  Should Romney improve any more nationally, Virginia will be one of the first states to tilt to Mitt Romney, adding 13 more electoral votes.
Colorado (tied) - Colorado has been very close for a very long time and continues to be close. Nine more electoral votes for Romney would give him 257 out of a needed 270.

New Hampshire (O+1) - Still blue for now.  In 2008 Obama earned a 9-point victory in New Hampshire.  In 2004, New Hampshire was the only state Bush won in 2000 to go for Kerry.  Basically New Hampshire has moved from being a slightly red state to being bluer than the nation as a whole.  Not this time though.  Obama's post-debate slump has narrowed his margin in the Granite State from 6 points to 1 point in the median poll spread.  Should this state, one of many in which Romney owns property, continue to tilt towards its summer resident, Romney will have four more electoral votes (261).

Iowa (O+2.5) - Iowa was originally thought to be one of Obama's strongest of the "swing state firewall" states, but Obama's margin has shrunk as it has in most of the other swing states, down from 5.5 to 2.5.  Winning this state full of Mitt Romney's best demographic (white people) would give Romney 267 electoral votes, three away from winning.

Nevada (O+2.5) - Nevada hasn't shown nearly the falloff in Obama margins that many other states have.  Right now Nevada is the tipping point state, but I think a more likely Romney path to victory goes through Wisconsin or Ohio.

Wisconsin (O+3) - The reason I think this state more likely to go Romney's way than Nevada is because of the trend, though like Ohio, Wisconsin has consistently shown Obama on top.

Ohio (O+3.5) - Like in 2004, Ohio will be the most watched state, even if it doesn't end up being the closest.  A Romney win in Ohio would mean he wouldn't have to win Iowa, Wisconsin, Nevada or even Colorado (as long as he gets Virginia and New Hampshire).  But if his performance on election night isn't strong enough to take Ohio, it definitely won't be strong enough to take Michigan or Pennsylvania.

Michigan (O+4) - There hasn't been a serious poll that has shown Romney with the lead in the state his father used to govern. Still, the median spread is only 4 points.

Pennsylvania (O+5.5) - There also hasn't been a serious poll showing Romney with the lead in Pennsylvania, but there was a Susquehanna College poll that showed him on top, which says more about Susquehanna College than it does about Mitt Romney's chances in Pennsylvania.

The rest of the states will support Barack Obama easily.


The senate is different.  Since the night of the first debate, Obama's margin has waned an average of 2.7 points in each of the thirteen states listed above.  But the margin for Democratic senate candidates has mostly remained unchanged, averaging +0.3 in the direction towards Democrats during the same time.  Here are the 14 senate races that will determine the majority in 2013, again ranked according to Republican support in descending order.

North Dakota (R+9) - Before a new poll came out today showing Rick Berg with a 10 point lead over Heidi Heitkamp, Democrats could have been excused for thinking this was going to be a close race.

Nevada (R+5) - Nevada is basically Opposite State.  There will be a lot more cross-ticket voters voting for Romney / Democratic senator than there will be voting for Obama / Republican senator in most other swing states (Ohio, Florida, Virginia, Missouri, Indiana, Arizona maybe).  Nevada will feature more Obama / Heller voters than Romney / Berkley voters. 

Montana (R+2) - Jon Tester was supposed to be the most easily beatable Democrat out of this year's crop of incumbents, and this is still the case two weeks before the election, but it certainly doesn't mean that this race will be an easy win for Denny Rehberg.

Indiana (R+1.5) - There have basically been two non-partisan non-internet polls in this race, and they say opposite things. But assuming the latest Rasmussen poll showing Mourdock with a 5 point lead is true, this race will give Republicans their 47th seat in the U.S. Senate. 

The median poll spread has Democrats winning the rest of the seats, though I don't especially trust it in the case of Arizona.

Arizona (D+1) - In a Senate race featuring the word "pretty", this is an ugly race featuring ugly candidates.  After Richard Carmona's gaffe coincided with Jeff Flake's personal attack ad so perfectly, I'd expect Carmona's numbers to plummet.  If you give this one to Republicans, that puts their total at 48.  But it will be hard for them to find 2 or 3 more seats.

Virginia (D+2.5) - One of the few races to show a trend similar to Obama's October, Tim Kaine is still polling a couple points higher than Barack Obama, which should keep this seat Democratic.

Missouri (D+3) - There has been just one poll taken since October 3rd in this close / insane senate race.  This makes no sense to me.  That one poll was performed by the usually Republican-leaning Rasmussen Reports, and it showed an 8-point Claire McCaskill lead. 

Wisconsin (D+3) - The persistent lead by Tammy Baldwin continues (except for in one Marquette poll).

Connecticut (D+3) - Chris Murphy continues to lead Linda McMahon, though polling methodology seems to be all over the place as evidenced by the scatter in the thin lines below.

Massachusetts (D+3) - The trends are all positive for Elizabeth Warren right now.  If Romney pulls off an upset on November 6th, expect a lot of inconsolable Democrats to say "at least Elizabeth Warren won in Massachusetts!" as they cry into their soy lattes.   

If Republicans are to take control of the Senate, they will have to win at least two and probably three out of the five preceding races (Virginia, Massachusetts, Connecticut, Wisconsin, Missouri).  Because they ain't winning any of the next four.

Ohio (D+8) - Sherrod Brown's got this thing on lock.  There has been no consistent movement in the polls since September and no single credible poll showing Josh Mandel with any sort of lead.

Florida (D+8.5) - Same story in Florida.  Barring a massive October surprise, Bill Nelson's lead is insurmountable two weeks out.

Pennsylvania (D+9) - Really weird polling in Pennsylvania.  In every other state, the polls are either in agreement with each other or scattered randomly over a large range.  In Pennsylvania there are two distinct groups of pollsters: those who show Tom Smith's support above 45% (maybe could win!) and those who show it below 39% (no chance).  There are no middle-ground pollsters.  Since there are just three pro-Smith pollsters (Rasmussen, Susquehanna, Quinnipiac) and five anti-Smith pollsters (PPP, Siena, Franklin & Marshall, Philadelphia Inquirer and Muhlenberg College sort of), the median poll lies in the anti-Smith camp. 

Michigan (D+10.5) - The Michigan senate seat has ceased to be on even crazy people's radar.