海角大神

Mitt Romney鈥檚 fundraiser comment that he doesn鈥檛 worry about the 47 percent of Americans who pay no income taxes, and that those people are dependent on government and consider themselves 鈥渧ictims,鈥� has been Topic 1 in US politics for days now. Polls are now starting to appear that take these words into account, and at first glance they don鈥檛 look good for the GOP standard-bearer.

For instance, according to a just-released 36 percent of respondents who knew about the flap said Mr. Romney鈥檚 comment makes them less likely to vote for him. Twenty percent said it made them more likely to vote for him, while 46 percent said it made no difference.

鈥淭he immediate impact of Romney鈥檚 comments appears to be more negative than positive, which suggests that the comments could hurt Romney鈥檚 ultimate chances of winning the election,鈥� writes Gallup editor Frank Newport.

But Mr. Newport added that the ultimate effect of the comment remains unclear, in part because today鈥檚 polarized media outlets are spinning the comments in very different ways. It鈥檚 possible, too, that Romney won鈥檛 lose as many votes over this issue as Gallup鈥檚 initial numbers imply.

One reason is that polls that ask whether particular things make someone more or less likely to do something aren鈥檛 that definitive. What鈥檚 鈥渕ore鈥�? What鈥檚 鈥渓ess鈥�? Ten percent more? Twenty percent less? What the question really measures is whether respondents believe the item in question is positive or negative.

And that response, in turn, is colored by what respondents were already leaning toward doing. You can see this in the details of Gallup鈥檚 numbers. The poll finds that 68 percent of Democrats say they鈥檙e less likely to vote for Romney due to the 鈥�47 percent鈥� stuff, for instance. But pretty much all those people weren鈥檛 going to vote for him anyway.

Only 4 percent of Republicans said the comment would make them less likely to pull the lever for the former Massachusetts governor. Forty-four percent said it make them more likely to vote for him. But again, most of those Republicans were going to vote for their party鈥檚 nominee in any case.

The results for independents were arguably more indicative. Of these self-described swing voters, 53 percent said Romney鈥檚 recent words made no difference. Twenty-nine percent said they made them less likely to vote GOP, while 15 percent said it made them more likely.

That鈥檚 not a positive result for Romney, but it鈥檚 far from a disaster. According to these numbers, he鈥檒l lose a few independent votes at the margin. But it鈥檚 still weeks until the election, so it鈥檚 possible even that effect won鈥檛 persist.

That point leads to the second reason Romney鈥檚 words won鈥檛 swing the election: Gaffes seldom do. As we鈥檝e already pointed out, stumbles that seem game-changing to pundits on cable news often don鈥檛 make much difference to large numbers of real voters.

John Sides, a George Washington University associate professor of political science, has graphed poll responses to various 2012 stumbles, such as President Obama鈥檚 鈥減rivate sector doing fine鈥� statement, and he鈥檚 found they generally result in no movement.

鈥淗asn鈥檛 the 2012 campaign taught us not to jump the gun with various 鈥榞affes鈥�?鈥� he writes on the .

That does not mean the 鈥�47 percent鈥� words won鈥檛 have an impact. It does mean they are but one gust in the windstorm that is a presidential campaign, and it is the whole storm that finally blows one candidate or another over the finish line first. (Can you think of a better metaphor? Feel free to let us know.)

Look at it this way: A new finds that Mr. Obama has an astounding 43 percentage point advantage over Romney on the question, 鈥淲hich candidate connects well with ordinary Americans?鈥� Sixty-six percent of respondents answered that the incumbent US chief executive does. Twenty-three percent said Romney does.

The 鈥�47 percent鈥� comment won鈥檛 help Romney close this gap, will it? In that sense it only solidifies the picture many voters have of him as a wealthy man who does not understand their problems.

One warning sign for Romney is that battleground state polls have not been good for him in recent days. New Fox News polls in , , and have all shown Obama with substantial leads: 50 percent to 43 percent in Virginia, 49 percent to 44 percent in Florida, and 49 percent to 42 percent in Ohio.

Interviews for these Fox surveys were conducted Sept. 16-18. The 鈥�47 percent鈥� story broke on the 18th, meaning they may also reflect 鈥� a bit 鈥� a negative initial reaction to the comments.