Tuesday, April 17, 2012

Wisc poll says only GOPs and Indys will vote in Senate recalls. They're wrong

Saw that PPP released another poll this week on the Wisconsin recall races, and unlike 6 weeks ago, the numbers weren't all that encouraging. This poll had Scott Walker up 5 on Tom Barrett, 7 on Kathleen Falk, and double digits on Kathleen Vinehout and Doug La Follette. The Senate numbers showed all 4 Dem candidates behind, albeit with one race at 2 points and with both candidates under 50%. This could be seen as even more worrysome because PPP nailed the Wisconsin Senate recall elections last Summer.

So what happened? Did a whole lot of people just change their mind and decide "it's working?" Uhhh, no. Let's let PPP explain it themselves.
The biggest change is probably that this was our first time using a likely voter screen, We found with almost all of the recall elections last summer that the electorate just wasn't as Democratic as 2008, even if people were at least more excited about the recalls than they were about the 2010 elections. Gov. Scott Walker gains a few points as a result of that.

In other words, they consciously changed who they were polling, and it shows in the demographic numbers at the bottom of the page.

Demographics of April PPP poll
Dem 31
GOP 32
"Independent" 37

Liberal 18
Moderate 43
Conservative 38

Obama voter 50
McCain 41
Didn't vote 9

Let's go back to the 2010 CBS exit poll in Wisconsin, and remember that voting population in a GOP-slanted year- 21 liberal, 43 moderate, 36 conservative, and 37 Dem, 36 GOP, 28 Indy. Then remember that Obama carried Wisconsin 56-42 in 2008. So PPP's screen indicates that many of the Obama voters in 2008 won't show up here, and the voters taking their place will be more likely to be GOP or independent (and remember, a sizable amount of "Independents" are Tea Baggers, who are nothing more than Republicans in disguise).

However, when you look at PPP's success in 2011, the theory that a lot of Obama voters leave the scene isn't true. Let's use Craig Gilbert's excellent article using Wisconsin Senate districts and the 2008 and 2010 vote, then compare to PPP demographics in their accurate polls in the 6 GOP senate elections that took place last August.

voting by Senate districts, 2008, 2010, PPP poll 2011
District 2- Obama +6, Walker +16, 2011- McCain +3, Walker +10
District 8- Obama +4, Walker +9, 2011- Obama +0, Walker +3
District 10- Obama +2, Walker +18, 2011- McCain +2, Walker +12
District 14- Obama +5, Walker +16, 2011- McCain +2, Walker +7
District 18- Obama +4, Walker +16, 2011- McCain +3, Walker +8
District 32- Obama +23, Walker +1, 2011- Obama +14, Barrett +2

Looking at the data, looks like the dropoff is generally the same for both Walker and Obama voters (the La Crosse-area District 32 is the exception, as Obama's dropoff is 9 vs. Walker's 3). So let's run the same test with PPP's 2012 polls on the 4 GOP Senate recalls. The one slight difference is that the poll asks who the person would favor in a recall guv election, Walker or Barrett.

voting by Senate districts, 2008, 2010, PPP poll 2012
District 13- McCain +3, Walker +26, 2012- McCain +4, Walker +19
District 21- Obama +12, Walker +8, 2012- Obama +0, Walker +7
District 23- Obama +12, Walker +13, 2012- Obama +4, Walker +15
District 29- Obama +9, Walker +16, 2012- McCain +2, Walker +23

All of a sudden Scott Walker gets more support in District 23 (Eau Claire) and Distirct 29 (Wausau) than he did when he was less known in the 2010 elections? In districts that are heavily unionized and with a huge UW campus in District 23? And in districts where the recall threshhold was easily reached? BULLSHIT. Between that and the double-digit drop-offs for Obama in District 21 (Racine) and District 29, you know that poll sample is slanted badly for the GOP, except for the Fitzgerald-Compas race in District 13, which if anything favors Lori. Not surprsingly, that race is 54-40, which is a bit big but within striking distance, and an 12 point shift that would be a huge help for Walker's Dem opponent.

So let's make this change be reflective of what really happened in the 2011 recall elections, and the results might look more like this. Then I'll divide the differences by 2 to show what the shift in the overall numbers might be.

District 13- McCain +10, Walker +19 (shift from poll GOP +3)
District 21- Obama +4, Walker +0 (shift from poll Dem +7.5)
District 23- Obama +4, Walker +7 (shift from poll Dem +4)
District 29- Obama +2, Walker +9 (shift from poll Walker +9)

So let's now use the adjusted Obama and Walker figures, and we can try to create a hybrid that might give us a better idea.

District 13- Fitz +17 (GOP)
District 21- Lehman +5.5 (Dem)
District 23- Moulton +6 (GOP)
District 29- Petrowski +5

Looks a whole lot closer now, doesn't it? And after all of the GOP and Walker moves against women, including the limiting of damages from the Equal Pay Law, does anyone not think this will shift further Dem as the election gets closer? Of course it will.

The last thing I want to bring up on this poll is another comment from PPP's Tom Jensen, whose Senate polls have samples that are consistently 75% GOP and INdependent outside of District 21.
One thing we saw in all these state senate polls is a much more GOP-friendly electorate than 2008. That suggests to me that Democratic voters are suffering from recall fatigue— this has been going on for so long and we are now so far removed from the protests in early 2011 that the passion Democratic voters have for coming out and removing these folks from office just isn't what it was a year ago. It's possible those folks could be reenergized by the campaign over the next couple months but right now Democratic interest in these elections isn't what it needs to be.

Sounds like someone's not out on the ground. Because if you don't think we're not going to be blasting full-force by June 5 and our determination to blow Scott Walker and the GOP Legislature out of power has flagged, you aren't talking to anyone that's actually here in Wisconsin. After gathering over 900,000 signatures and being this close to making history, we'll be damned if we'll fizzle out, not turn out to vote, sit back and let this thuggish gang continue wrecking the state we love. Hell no.


  1. Well, your analysis does make me feel a little better about that survey, so thanks.

    A Daily Kos reader who lives up in District 29 had a rather disheartening report that the Seidel campaign was thrown for a loop by Galloway resigning and hasn't really been able to pivot into a different kind of campaign. Apparently Seidel's people are nowhere to be seen, while Petrowski supporters are everywhere. Also, he reports that Petrowski is inexplicably popular up there.

    I wouldn't be surprised if there aren't similar problems in the other districts. I hope the Dems all get their acts together after the primary (yeah, tall order).

  2. Thanks for reading, Sue. I think you need to tie the Senate recalls with the Gov/ Lt. Gov recall. Just bring up that "Jerry Petrowski voted with Scott Walker", and have Seidel and the Dems say "We will stop the Walker agenda."

    Do NOT turn these races into individual situations, because these GOP Senators sure didn't vote as individuals, and they were recalled because they didn't act as individuals. So why treat them as such now.

  3. Want a follow up? Here you go. Dem pollster says PPP is way off in underestimating Dem turnout in these races, and that Lehman is up in District 21, and that Moulton/ Dexter and Seidel/ Petrowski are virtual tossups.