From Victor Provenzano:
Thus far, the 4 national tracking polls have registered an infinitesimal combined net gain of 0.375% for Mitt Romney now that one and a half days (Wednesday night and Thursday) have passed since the debate:
(1) The Rasmussen poll shows the race has not moved in either direction
(2) The Gallup poll shows Obama gaining 1%
(3) The Rand poll shows Obama gaining 1.5%
(4) The online Ipsos poll shows Mitt gaining 4%
Taking all 4 national tracking polls into account, Obama gained a combined 2.5% (in 2 out of the 4 polls) while Mitt gained a combined 4% (in only one poll), which yields a net gain for “Willard” of 0.375%. He won by a long shot! Those 4 polls include a small number of voters who saw the debate on Wednesday night as well as a fuller national sample on Thursday of those who did and did not see the debate.
The only statistical effect thus far is in some of the state polls: in Nevada (where Mitt is now down only 4.6% on average), in Ohio (where Mitt is now down a mere 3%), in Virginia (where Mitt is now behind a minuscule 0.4%), and in Florida (where Mitt is now even with Obama). This movement in the state polls, however, probably has more to do with Rasmussen’s strong Republican bias—-along with the effect of Karl Rove’s recent media spending—-than it does with the recent debate.
Most debates have little if any effect on the outcome of the presidential race since the public has more or less made up its mind before the series of debates begin in the fall. A third of the time, a “win” consists in a net gain of only around 1% for one of the candidates, while a “big win” consists in a net gain of about 3%. In 1980, for instance, Ronald Reagan made a series of major gaffes yet “won” the debate against Jimmy Carter by repeating the phrase “there you go again” gaining a “stunning” 2.8% as a result.