Monday, November 5, 2012

A Landslide Without Coattails? -

I could go on and on. But if you want to read an account of Romney’s prospects more sober and cautious than my own, take a look at what Michael Barone had to say in The Washington Examiner on Saturday. His state-by-state analysis deserves close attention.

I would disagree with Michael in only one particular. I believe, as I have believed since early August, that Romney’s victory will be much more impressive than Michael imagines. What happened in 2010 is going to be confirmed in spades. The electorate will break for him on Tuesday as they broke for Ronald Reagan on the first Tuesday in November in 1980.


I hope Paul Rahe is right, obviously, as I hope Michael Barone is right. One thing about this election -- one has a broad menu from which to choose who one hopes is right. I concede I don't have the patience or the mathematical skills to go deep into the weeds with those looking at state by state polls and attempting to unskew them in sophisticated ways. I am able to be skeptical of a poll that gives you a top line based on an assumption of D+7 say when 2008 was D+5. There has certainly been a lot of that. These assumptions have been based on a Democrat GOTV machine working as well as it designers say it will -- but whether they should be taken at their world is unclear. My skepticism is increased by knowing that most media outlets and probably pollsters too are much closer in the social network to Democrats than they are to, say, evangelical Christians or Mass-attending Catholics in southwest Ohio. Thus claims of Dem GOTV prowess are likely to resonate with pollsters much more than GOP claims of enthusiasm. This is a long winded way of saying, I am not comfortable assuming that the state poll top lines do anything like capture who actually is likely to turn out on election day. The early voting numbers I am simply unable to interpret apart from the spin given to them by the people I'm getting them from, mostly GOP sources. Though on their face, the claim that Mitt is doing well in EV seems plausible, though in Ohio what is going on in EV seems obscure. OTOH Karl Rove's analysis of Ohio EV seems persuasive, but I don't know what assumptions his is based on. So what it seems to come down to is the GOP/Mitt supporters' claim that something is going on at ground level (maybe something historical, as Rahe claims) that will lead to a (perhaps even decisive) Mitt victory and the claim that state polls are not capturing this something. This, combined with some sound reasons for thinking that if it is actually going on, the state polls might well not capture it. To capture it, poll designers would have to have enough exposure to, e.g., evangelical enthusiasm in Ohio to set it up against Dem GOTV prowess, which they don't. Of course, absence of evidence is not evidence of presence. There is enough evidence of widespread Mitt enthusiasm that if he does win, it will seem in hindsight that we should have known it all along. But I don't know it now. I will say the Dem confidence seems somewhat contrived to me, based on assumptions of a very large social organization performing exactly as planned in a complex environment, which they often don't. So wrapped in a thick cloud of uncertainty we have a core of . . . uncertainty. But there is plenty of room for reasonable hope. --TS

| Permalink