I am obsessed. I am excited. I am befuddled. I am confused. I am resigned. And I have something to say.
Explanation below the fold.
This cycle has seen more U.S. House polls than any in history. And I devour each and every one.
I'm completely obsessed with this election, which is nothing new for me, except that the fact we're now expected to take the House amplifies it for me.
I am excited because a slew of polls has come out demonstrating that we're ahead in this race, that race, the other race, including races that we never expected to win.
I am befuddled because some of these polls just seem too good to believe. The folks at Constituent Dynamics are on drugs with too many of their poll results. No way is Massa up 12 on Kuhn, Hall up 9 on Kelly, or Maffei up 8 on Walsh in the NY races. Out of 54 races polled on the Majority Watch list, I peg 8 in our favor as not even passing the sniff test.
I am confused because quite a few polls, both positive and negative for us, make no sense at all. Could Dem incumbent Julia Carson in IN-07 really have blown a 20-point lead in one month to fall 3 points behind a Gooper with no state or national party backing? Could Jack Davis fall from 15 points up to 3 points down in a single week against Gooper incumbent and NRCC Chair Tom Reynolds in NY-26, all based on a Gooper ad attacking Davis' "protectionism"? And this after Davis supposedly had turned a 2-point deficit into a big lead in the preceding 2 weeks? Could Kirsten Gillibrand in NY-20 go from down 1 to up 13 to down 14 to down 2 all in the course of 3 days? And how could Zogby's phone poll, not the interactive shit he uses for the Wall Street Journal, have Braley losing by a whopping 47-34 to Whalen in IA-01 when all other polls show Braley comfortably ahead and Whalen barely reaching 40 if at all? (And please note the list in the IA-01 poll link errantly leaves out the Des Moines Register mid-September "Iowa Poll" that showed Braley up 44-37)
I am resigned to two things. First is the lesson taught by "Mystery Pollster" Marc Blumenthal: House races are very hard to poll, so much so that the "various challenges have made many media outlets and public pollsters wary of surveys in House races." A race that's hard to poll is highly likely to produce some inaccurate polls. It's more than just sampling error, it's "differences in pollster methodology" that produce vastly different results between 2 polls for the same race even if both are taken roughly around the same time. Second is that maybe analysts like Charlie Cook, National Journal, and Congressional Quarterly are the way to go for evaluating races. They presumably have some inside access to party committees, individual campaigns, and maybe even state party people, and they don't rate races merely by a smattering of inconsistent or unbelievable polls.
But alas, I ultimately am too obsessed not to let my heart flutter with every excitement or deflate with every disappointment as a House poll comes in from here and from there.