If I had the time, I'd do it myself, but--with a high school water polo tournament to attend in the morning and a high school football game to attend in the afternoon--I just don't. I suppose what I'm hoping to find is a Ricochetian whose curiosity coincides with mine, but who either doesn't have kids yet or has his child-rearing years behind him: in short, someone with a free couple of hours on Saturday morning.
Here's what's up: Compare the House rankings on Rasmussen, the Cook Political Report, and Stu Rothenberg, I've heard several knowledgeable people comment over the last couple of days, and what you'll find is a lot less overlap in the "tossup" races than you might suppose. Each seems to be looking at different polls--or weighting the same polls in different ways. And if these pollsters--the best of the best--differ, then the rest of us may safely conclude, right along with Paul Rahe, that in these final days of the campaign the whole country is flying by the seat of its pants.
Does any Ricochetian have the time and interest actually to put together a spreadsheet, comparing the "tossup" races on those three sites?
- Comment (14)
- · Quote
- · UnfollowFollow (2)
Comments :
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Sorry, got kids, but at least can pass along a link to Google's election map interface, which features input from Cook and Rothenberg but alas not Rasmussen: http://bit.ly/9t5qhQ I think many of these prognostications are likely inaccurate and out-of-date, but at least there are five of them!
Sep '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
It is possible your post about the bookstore says more about the probable results of the election than the polls.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Only have time to offer this fact, which is significant: In 2008 Obama won in only 242 congressional districts while winning the popular vote by 7.3%, yet in 2004 Bush won in 255 districts while winning the popular vote by 2.4%.
A big problem for Dems is that their support is very concentrated; therefore they will likely win some districts by whopping margins but are vulnerable to losing many seats by small margins.
For Senate races, this quirk is not so relevant, but its significance for House races is huge--difficult to measure, but huge--and potentially even more huge with Obama's support being so AfAm-heavy, which leads to an ever-more concentrated support geographically.
Truly, this election has no easy historical equivalent with which to compare it. Here's a safe prediction: We'll win somewhere between 35 and 100 seats. Should be interesting.
11 days.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Based on what Jim Geraghty's mentor "Obi Wan" is saying, plus Pat Caddell and others, I'm getting the idea that a true black swan event--the mega-wave we're all expecting--is impossible to predict by definition. Obi Wan seems to say we won't know for sure until we see the polls just a few days before the election. So we're stuck in a situation where the best evidence for a mega-wave is the lack of evidence.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
One problem with polls is they exclude certain personality types and lifestyles. I know plenty of voters who when asked to participate in a poll quickly hang up or throw away the letter.
Jul '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Nothing short of a spirit-crushing defeat for the other side leading to long cycles of suicidal despair is what I wish for. Hot, cold or just right -- I'll accept revenge at any temperature it comes to me. Does this make me a bad person?
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
My intuition (hey, why do only women have an intuition! Not true or fair!) tells me that the bad shape of the economy; the socialistic nature of what has happened in the last two years; and, yes "rumors" of the wave itself will pull many voters off the sidelines and into the voting booth; citizens who are scared; and citizens who, in most, if not all other elections, didn't bother voting because no one viable represented their views.
Combine the above with the demise of traditional telephones, and I don't think the pollsters can poll with a MOE of less than 10-15%.
What interests me is how polling of the primaries and referenda earlier this year panned out; yes, primaries aren't polled as extensively as the general, but was the MO referendum on Obamacare predicted to be so lopsided? What about the victories of various TeaParty-ish candidates over the incumbent GOP office holder?
Oct '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Bad; very bad; rotten to the core in fact!. Welcome to the club!
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Good Berean
Bad; very bad; rotten to the core in fact!. Welcome to the club! · Oct 23 at 10:13am
That makes three of us.
Oct '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Good Berean
Bad; very bad; rotten to the core in fact!. Welcome to the club! · Oct 23 at 10:13am
Really, revenge is best served cold. But in this case even tepid would be fine.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
ParisParamus: Combine the above with the demise of traditional telephones, and I don't think the pollsters can poll with a MOE of less than 10-15%.
Oct 23 at 8:30am
Powerline had an interesting piece discussing the polls from the NJ, VA, and MA races to the actual results, and a lot of them were not only close, but dead on.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Every election cycle I make hundreds of phone calls.
That experience has led me to conclude that no matter how really, really, really smart the pollsters are; and how really, really, really scientificy they think their methods are, they really have no clue.
Like today, probably 80% of the numbers you call the people don't even pick up, the phone either just rings, or you get a machine, or it's a wrong number. Then 10% of the remaining hang up on you as soon as they realize you're someone they don't know.
Same thing goes for going door-to-door, etc.
So how the pollsters can tease anything meaningful out of that kind of sample, I have no idea.
May '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
Plus you're a talking frog, which probably scares some of them off.
(Seriously, that's great you make calls. I always intend to do such things but, alas, never quite get around to it.)
Oct '10
Re: More Evidence That the Polls Represent Art, not Science; or, Anybody Out There Looking for Something to Do on Saturday Morning?
I know lots of people who no longer have a "home" phone and only use their cell. Pollsters do not reach these people, a significant portion of the busy younger crowd. The people who answer polls are an odd bunch to my mind. I have rarely told anyone how I voted, and certainly not a stranger on the phone. But then I live in a socialist paradise (Canada) and so am perhaps more paranoid. We once had a Member or Parliament refuse to help one of his constituents because he knew the gentleman hadn't voted for him. Never found out how he knew.