"Is 2010 the Year American Voters Turn Polls on Their Heads?"

[Clio the Leo]

Is 2010 the Year American Voters Turn Polls on Their Heads?
WALTER SHAPIRO

As connoisseurs of presidential primaries know well, New Hampshire voters are notoriously contrarian – mischievously rewriting the media's story line by voting for Hillary Clinton (2008), John McCain (in a 2000 landslide), Pat Buchanan (1996) and Gary Hart (1984). Cosseted by the candidates and pampered by the pundits, these first-in-the-nation primary voters rebel against being taken for granted and revel in proving the dumbness of the conventional wisdom.

All this is prelude to an against-the-grain notion: Is 2010 the year in which all American voters become New Hampshirites?

Never has there been an off-year election with such a perplex of polls, such in-depth public debates over polling methodology, such over-hyped political projection systems and such pseudo-certainty in early October about the outcome. From the proliferation of inside-baseball political Web sites to the ravenous (and sometimes rabid) appetite for political news on cable TV, we have created a nation of armchair campaign consultants.

The expectation of a GOP tidal wave is so engrained in the media and politically sophisticated voters that it is easy to imagine the morning-after headlines: "In Stunning Rebuff to Republicans, GOP Only Picks Up 93 House Seats." Many members of the punditocracy have already moved beyond the election itself to game plan Barack Obama's next two years grappling with a Republican House and possibly a GOP Senate.

<snip>

But for all the glib talk about the next great political tsunami coming in November, it is worth remembering that until recently the House was the embodiment of incumbent-protection win-reelection politics. The 1994 Gingrich Revolution – undeniably an epic wave election – ousted the Democrats from control of the House after a stunning four decades. But every off-year election since then has defied historical patterns. In both 1998 and 2002, the incumbent president unexpectedly gained House seats because of unusual events (Clinton's impeachment and the 9/11 attacks). The 2006 Democratic takeover was a seismic event of far greater magnitude on the political Richter scale than most anticipated after George W. Bush's 2004 victory.

Not too long ago campaign reporters were stripped of their press tags if they did not invoke Tip O'Neill's mantra ("All politics is local") in at least half their stories. Now the dominant orthodoxy is that all politics is national. But one of the biggest questions to be tested in November is whether candidate quality matters in individual races – or whether the adroit and clueless alike get carried off to Congress by the same national wave.

On paper, for example, 76-year-old Leonard Boswell, representing central Iowa, should be a walking definition of a vulnerable House Democrat since he never has won with more than 56 percent of the vote during this century. But Brad Zaun, Boswell's ultra-conservative GOP opponent, is struggling to raise money. And in a telling assessment of Zaun's chances, the National Republican Congressional Committee (NRCC) is running ads in nearly 50 House districts, but is not on the air against Boswell, even though Des Moines TV time is inexpensive.

Washington political consultants are not (shocking revelation ahead) all-knowing. Every year party committees invest heavily in losing races and stiff-arm candidates who prove victorious. (New Hampshire House Democrat Carol Shea-Porter, for example, won in 2006 despite Rahm Emanuel's refusal to assist in her race). So the NRCC's current decision to downplay the Boswell race may not prevent the GOP from picking up the seat. But it does serve as a small illustration of the complicated calculus in handicapping the roughly 100 House and Senate seats in play this November.

This is a strange year – and it is possible that both parties (plus the press pack) are all flying blind. In my off-the-record conversations as a reporter, I have talked with Republicans who are more subdued in projecting GOP gains in November than their Democratic counterparts. Sometimes, in fact, I wonder if the private predictions of political insiders say more about their personal bio-rhythms than any secret knowledge.

So as the campaign days dwindle down to a precious few, it would be wise to remember how cantankerous voters can be – and never assume that polls and prognosticators have taken all the surprise out of politics.

http://www.politicsdaily.com/2010/10/07/is-2010-the-year-american-voters-turn-polls-on-their-heads/

[zbdent]

1. You mean like some other elections in recent years?

Where there were exit polls showing a Democrat leading in the votes, and yet, when the unverifiable electronic voting machines weighed in and the Republican suddenly surged ahead and won?

[karynnj]

3. There is a huge difference between exit polls, taken (obviously)

after people just voted - and these polls. One obvious difference is that there is no need for a likely voter model. On regular polls, it would be silly to use the RV results, but the likely voter model is as much art as science. In addition, you are sampling directly from those who come out to vote - so the only group missed are those who won't talk to pollsters.

[Teaser]

2. this.

polling is in crisis, and has been for a while.

Maybe this is the election that forces it to reboot.

[Imajika]

5. No, polling is not in crisis...

Rejecting the polls is akin to burying your head in the sand.

The art of polling has been refined to the point it is quite accurate - especially when you average many polls (RCP average for example). The sheer number of pollsters competing with each other has also helped make polling even better.

Many people here don't want to hear the polls right now because they tend to indicate a bad year for Democrats, but that does not mean polling is in crisis.

[Teaser]

6. "art of polling has been refined to the point it is quite accurate"

quantify. what's "quite accurate"?


"especially when you average many polls"

no, this is a sign that there is tremendous variability in the underlying data sets.

"The sheer number of pollsters competing with each other has also helped make polling even better."

opinion statement, contains no data or argument. "even better" means nothing without some referent.

[Bluenorthwest]

4. Polling is hocus pocus snake oil fakery

It is done in order to make for a false narrative dynamic which in turn makes for ad sales and consultant's fees having a nice surge. That's it.