Skip to main content


If you haven't been following U of F professor Michael McDonald's twitter an early vote website you should be, it's great work.  Democrats have whittled down republican's registration edge from 8.0 points yesterday to 7.3 points this morning.  McDonald claims that it is likely the final number will be below the 6.8 point registration edge republicans enjoyed in 2010, which is very promising.  With all the late mail in ballots and provisionals, CO may take a while to call.
Michael McDonald Twitter

Discuss

I'm pretty sure this information has been posted before, but let's look at the final RCP Senate averages and compare them to the last two cycles to see how we stack up (Presidential swing states and Senate seats deemed tossups).  First off, in no race (26 total) over those two cycles has the republican outperformed his or her final RCP average, not once.  This bodes well for democrats in NC and NH (and Orman in KS) who lead narrowly on the RCP average.  In CO, IA, and AK the democrats trail by 2.5, 2.3, and 2.4 points respectively.  This compares rather favorably to RCP's final average history as the median error in those 26 races favored the republican by 3.6 points.  In fact, over 80% of these 26 races featured an error that favored the republican by 2.4 points or more.  Looking at just the senate results over the last two cycles (13 races), the median republican lean was 4.1 points, with only one race favoring the republican by less than 2.5 points.  This doesn't mean we are probably going to win CO, IA, AK, or KS, but it does mean that doing so wouldn't be unexpected or defy recent history in any way.

Continue Reading

If you haven't seen already, PPP's last Iowa poll shows Ernst leading Braley by 3, a bit of a difference from their October 16th poll that had Braley leading by 1.  So where did this 4 point shift come from?  Well, in their October 16th poll, Ernst led Braley with white voters by only 2 points, which would probably be good enough for Braley to win even though whites are 90+% of the electorate in Iowa.  In their latest poll, however, Ernst ups her lead with whites to a monstrous... 2 points...  What the heck?  How can Ernst have the same lead with whites, but improve a whopping 4 points overall?!?  Well, in the October 16th poll Braley leads with non-whites by 19 points, while in the latest poll Ernst leads with non-whites by an absurd 10 points. I guess Latinos and African Americans and Asians really love them some Joni Ernst.  As Wayne Campbell might say, yeah, and monkeys could fly out of my butt!  This just highlights pollsters inability to accurately poll non-white voters.  In Iowa, the electorate is probably 2-3% African American, 2-3% Latino, and 1-2% other.  There is no way in heck Braley is winning non-white voters by fewer than 20 points, and it's probably much more.  To make a long story short, if Ernst only wins white voters by 2 points, as is shown in both of these polls, she will lose easily.

Discuss

Just a short diary, after showing leads for Gardner and Ernst of 7 and 4 points respectively in their previous CO and IA senate polls, Quinnipiac now shows a lead of just 2 for Gardner and a tie in IA.  I guess they figured out that their previous polls were garbage and are now trying to avoid a total Titanic situation.  This aint over yet, they sleepin on us.

Discuss

If you haven't seen it yet, the ever reliable Rasmussen has Ernst leading by a paltry 1 point in their (most likely) final Iowa poll, down from a 3 point lead in their last one.  Why is that so important.  Well, I looked at the final polls that Rasmussen has conducted for Senate, President, and Governor for the 2006, 2008, 2010, and 2012 cycles, and found that they have shown a tie race or the republican leading by 1-3 points in a total of 18 different races.  How many of those races did the republican actually end up winning?  The answer is one.  One time.  That was when they had Missouri tied in 2008, a race that McCain only barely won. In other words, whenever their final poll has a lead for the republican of 3 points or less, the republican has won only 5.5% of the time.  Over 4 complete cycles of polling.  This doesn't mean Ernst only has a 5.5% chance of winning, who knows they could be much better this cycle, but a win for Ernst would be rarefied air for Rasmussen polling.  So lets look at some of their other predictions.

Continue Reading

SurveyUSA's just released GA Senate poll has Purdue up 3 points on Nunn, an apparent reversal from their last 2 polls which has Nunn up by 3 points and 2 points.  Nunn's position with white voters, however, has remained remarkably consistent.  In their two previous polls, Nunn garnered 28 and 29 percent of the white vote, with about 4 percent undecided.  This would put Nunn in positioned to win outright on election night, as she needs 29-30% of the white vote to win.  In their latest poll, despite trailing Purdue overall by 3 points, Nunn still takes 28% of the white vote with 4% undecided, unchanged from the previous polls.  The difference overall comes with the African American vote, which Nunn receives only 80% compared with 15% for Perdue, much less than the previous polls.  This is obviously a ridiculous number, as Nunn is almost assuredly going to get 90+% of the African American vote.  So to reiterate, SUSA has now found Nunn getting between 28% an 29% of the white vote in 3 consecutive polls, putting her right where she needs to be to hit 50% on election night.

SUSA GA Senate Poll

Continue Reading

Suffolk University polling as Gardner leading Udall by 7 points in their latest CO senate poll!  This is great news for Udall as Suffolk is the same pollster who refused to poll NC, VA, and FL in October 2012 because Romney was so far ahead there that it wasn't worth their time.  Now they have Gardner sewing up the senate race in CO, so Udall is a shoe in for sure.  Now all we need is somebody call up Dick Morris to see what he thinks about the Colorado senate race.

In all seriousness, I can't believe USA Today is actually paying these twits for their polls when they have been so horrifically wrong in the past.  It seems like the worse you are at polling, the more money people are willing to pay you for it.

Discuss

If some thought SUSA's previous Georgia poll showing Nunn leading Perdue 48-45 was an outlier, their latest poll showing Nunn retaining her lead, this time 46-44, should dispel that notion.  The numbers inside the poll also show real strength for Nunn.

Continue Reading

SurveyUSA's latest CO senate poll is a real head scratcher.  After leading by 4 in their September poll, Udall now trails Gardner by 2.  Looking into the poll's crosstabs, however, shows the sheer ridiculousness of it.  In September, Udall led on the strength of 45-42 lead among white voters, with only a very small 48-46 point lead among Hispanics.  In the October poll white voters' voting intentions hardly budged, with Udall still leading 45-44, so what accounts for the 6 point overall change?  Looks like Hispanics are all jumping on the tea party bandwagon, with Gardner leading by an absolutely absurd 14(!) point margin, 49-35.  This is just ridiculous.  To put this into some context, Bennett lost white voters by 7 points in 2010 and still won the election.  If Udall ends up winning or even close to tying among white voters, he will sail to a very easy victory.  I can't believe they had the gall to publish such absurd results.  Stuff like this makes me irrationally angry.

SurveyUSA CO poll

Discuss

Mon Jul 14, 2014 at 04:21 PM PDT

Dem Pollsters are the Best Pollsters

by schobel94

After reading a post that broke down the polling for the AR senate race along partisan lines, I thought I would use HuffPo's create your own poll aggregator to see if there is a similar trend among the other 2014 senate races.  Unfortunately there seems to be a dearth of polling in most of the races, especially of the nonpartisan kind (sometimes only one or two), so it was difficult to get any meaningful results.  They do, however, have 2012 polls with the create your own feature, so I decided to separate the the (D), (R), and unaffiliated polls to see how they stacked up versus reality.

Continue Reading

Things looked pretty good for Mark Pryor a few months ago when several non-partisan polls showed him leading Tom Cotton by solid margins, a few even by double digits.  Since then we have been hit by an onslaught of right wing affiliated pollsters that show Cotton ahead about 5-7 points.  It gets a bit disheartening when you look at the HuffPo poll charts to see Cotton with a pretty hefty lead, albeit on the back of these republican pollsters:

2014 Pryor v Cotton

So today when yet another republican poll came out showing Cotton with a 4 point lead, I just had to look inside to see what was up.

Continue Reading

This is just a simple question, I haven't been able to parse this out.  He seemed pretty confident in his short presser that the house will pass it.  Will it get any democratic support?  Are there any republicans on record against it?  Just curious, but I suppose we will know by tomorrow.

Discuss
You can add a private note to this diary when hotlisting it:
Are you sure you want to remove this diary from your hotlist?
Are you sure you want to remove your recommendation? You can only recommend a diary once, so you will not be able to re-recommend it afterwards.

RSS

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site