Pew (Pew!) shows Romney ahead by 4
were wondering, like me, if there was a real Romney gain following last week’s debate, wonder no more. In a poll
conducted after the debate, Pew Research, one of the most reliably liberal-slanted polling organizations in the country, shows Romney leading Obamageddon Man by four points among likely voters, 49 percent to 45 percent. In its previous poll, taken in September, Obamageddon Man led Romney by 51 to 43 percent. That’s a gain of 12 points for Romney! I’m stunned. I’m even shocked.
- end of initial entry -
Andrew B. writes:
One of the best things to come out of the past 12 years of electioneering press coverage is the concept of looking at all polls being published in aggregate and as a running trend rather than slavishly looking at just one poll as though it is the most accurate, which can either confirm one’s previous biases when it is in one’s favor, or send one into a panic when it is not.
Of course doing this requires numeracy, which is sadly lacking in most of the public and the press.
Polling very small samples is inherently volatile, and the margin of error often mentioned will come into play in some polls, resulting in a Pew poll, for example, showing Romney leading by 4, or the Bloomberg poll several months ago that showed Obama up by 13. These are obviously to a sober observer what are known as outliers.
The reality of the race is, as I have mentioned before, shown by Romney needing to campaign in Colorado, North Carolina, and Virginia, and Obama not needing to campaign in Pennsylvania, Minnesota, and Oregon. Follow the money. Where the big money is being spent is where the race really stands, and remember we are talking about tens of millions of dollars every week.
In 2000 and 2004, the Electoral College had a conservative bias, in that the Republicans could merely draw even or even be slightly behind and still win. Because of shifts over the past 12 years in Nevada, Colorado, New Mexico, and Virginia due to the continued influx of immigrants the Electoral College now has a Liberal Republicans now need to be ahead about 2 points to win, because it will take that margin to flip most of those states back and win Florida, Ohio, Iowa and New Hampshire.
The reaction in the polls towards Romney is more a solidifying of Romney’s own support within the party (thus the ecstatic and almost hormonal reaction of the National Review crowd towards their new crush) and with Republican leaning Independents, and a temporary dispiriting of Obama’s support (as seen by the hysteria of people like Andrew Sullivan and Chris Matthews) than a movement of Democrats or Democrat leaning Independents toward Romney. This leaves the election depending on turnout to eke out a small Republican victory unless additional events occur to push Romney further up in the polls.
Polling bounces after events such as the conventions and debates can act as a sort of revealed behavior indicator, where marginally attached voters briefly become enthusiastic and give their real preference to the pollster of the candidate instead of acting indecisive and disinterested. Based on that right now, and based on the state polling and campaign events and where campaign ads are being run, one would need to say that the election is roughly a 49-51 Obama, 47-49 Romney, 1-2 Other affair right now, with undecideds looking to break around 2:1 for Romney.
Regarding the need for looking at several polls to get an average, as a general matter you are of course correct. However, when a famously liberal-biased polling organization comes out with a poll with a shift and a spread that spectacular, that is something that in itself deserves attention.
Terry Morris writes:
Assuming Pew’s numbers are correct, if Romney manages to hold this
Posted by Lawrence Auster at October 08, 2012 08:20 PM | Send
eleventh-hour lead into November, and actually wins this election, I
for one will be shocked. And as I’ve said before, I’ll gladly devour
my helping of humble pie for underestimating Romney and prematurely
counting him a probable loser.