Sunday, October 17, 2010

Sucker Bet

The anarcho-libertarians at the McHenryCountyBlog are fuming over right-leaning Daily Herald's endorsement of Melissa Bean. The blog, an industry leader in cognitive dissonance, is claiming an uncredited role in a recent Herald story ...
The resurrection of Joe Walsh's campaign
which is all about the resurrection of Eight Congressional District candidate Joe Walsh's dead-as-a-dead-parrot campaign from the dead.

According to the McHenryCountyBlog, Herald reporter Kimberly Pohl was
tasked to try to explain why Walsh is still in the race. It's as if she had read my article explaining the New York Times blogger Nate Silver having lowered Bean's odds of winning from 97.8% to 80.6%. This comes over two weeks after a very large (1,381–the largest I have ever seen in a congressional race–We Ask America telephone poll showed a 41% to 41% tie
Daily Herald Tries to Cover Rear End on Joe Walsh
... a virtual dead-heat between Walsh and Bean. The We Ask America poll in question surprised your LakeCountyEye, who is marking the point spread 55/45 in Bean's favor.

Readers of this blog will recall that We Ask America at a minimum pretends to be a real polling organization ...
The Numbers Racket
They have a website. And they publish poll results. Some published numbers from their recent Tenth Congressional District poll ...

We’re baaack
Eagle-eyed operatives may wonder how one goes about pulling off a neat trick like polling 132.38% of Democrats in the 10th C/D. No doubt something to do with how compound interest is amortized over the lifetime of the loan poll. Alas, your LakeCountyEye is not a mathematician (nor a payday loan provider). However Nate Silver (aforementioned by the McHenryCountyBlog) is ...
many of the polls are either partisan-affiliated, or were "robopolls" that used automated scripts rather than live interviewers, or both. (The former are highlighted in red in the table; the latter are given the designation "I.V.R.," for interactive voice response). Polls with an explicit partisan affiliation are on average about 6 points friendlier to their candidate than those conducted by independent groups. Robopolls have not shown any persistent bias in the past — but this year, they have been 2 to 4 points more favorable to Republicans than traditional surveys, and the differences have tended to be larger in polls of House races as opposed to conducted in Senate or gubernatorial campaigns. So this is a group of polls that you'd expect to be pretty Republican-friendly.
Canaries in the Coal Mine? Or Cuckoo Polls?
The We Ask America poll showing Bean and Walsh in a dead-heat, was both a robopoll and partisan commissioned. Adjusting Silver's numbers into the equation, your LakeCountyEye marks the point spread at about ... 55/45 in Bean's favor. Note to operatives: when in McHenry County do not pass up an opportunity to cover that bet against Melissa Bean. Kerching!


Anonymous said...

Please read the string on capitolfaxblog:

on We Ask America. Their 10th district poll (even after correcting that typo) does not add up.

No matter what percents you put in for Dem, GOP, and Independents, you can't get a top line of 50-39 with those breakdowns.

Barney Baxter said...

hi Anon,
I dusted off my linear algebra textbook. The respondents to the We Ask America poll in the 10th C/D work out to roughly ...

09.0% Democrats
09.5% Republicans
81.5% Independents



Cal Skinner said...

Did you ever write about We Ask America's poll being right-on?

Barney Baxter said...

hi Cal,

I think we did an election post-mortem at ...

Whitewash Joe Walsh