From today’s Wall Street Journal Political Diary E-mail…

Gallup Delivers a Stunner

As Election Day draws closer, every major public opinion poll shifts from interviewing registered voters to those whom it identifies as "likely" voters. Gallup, the oldest U.S. polling company, first developed the model it uses for identifying likely voters back in 1950 and its final election polls have proven highly accurate.

Yesterday, Gallup delivered its first 2010 "likely voter" poll and the results floored the political community. In the generic ballot question, which asks which party a voter would favor in a generic House contest, Gallup gave the GOP a 46% to 42% edge. But then Gallup applied two versions of its "likely voter" turnout model. In its "high turnout model," Republicans led Democrats by 53% to 40%. In its "low turnout model," the GOP edge was a stunning 56% to 38%. That kind of margin in favor of Republicans has never been seen in Gallup surveys.

What should worry Democrats most is that the "low turnout model" is typical of recent midterm elections. If the Gallup numbers hold up (and the firm cautions that "the race often tightens in the final month of the campaign"), some word more cataclysmic than "tsunami" would be needed for the Democratic losses.

Michael Barone, co-author of the Almanac of American Politics, says either of the Gallup turnout models would produce "a Republican House majority the likes of which we have not seen since the election cycles of 1946 or even 1928." Mr. Barone says the historical parallel might no longer be 1994, when the GOP gained 54 House seats, but instead 1894, when Republicans gained more than 100 House seats in the middle of the economic downturn that engulfed Democratic President Grover Cleveland.

Democrats I spoke with last night downplayed the Gallup numbers, pointing out that Gallup’s surveys have been somewhat volatile this year and other polls (such as those by Scott Rasmussen) show a much smaller GOP edge among likely voters, on the order of three percentage points. That would translate into a GOP House gain of 35 to 40 seats, hovering just around what Republican would need to take control of the Speaker’s gavel.

But regardless of where likely voters are right now, it’s a strange political year when Democrats start consoling themselves with Scott Rasmussen, whose polls they have long disparaged as being biased towards Republicans.

This entry was posted on Tuesday, October 5th, 2010 at 12:00 am and is filed under Blog Posts.