The presidential race continued well within the margin of error in polls released Saturday October 20. The RealClearPolitics Average continued to show Mitt Romney trailing Barack Obama infinitesimally, 47-47.1. Gallup continued to be the outlier, both in terms of its estimated large Romney lead and his (imputed) white share. Saturday's jump in Rasmussen's estimate of Romney's share of the white vote (see below) might be a confirmation of Gallup, but it's so sudden that it could also be statistical noise. Looking at all the polls, it's clear that a higher white share is the key to a Romney lead—and also that he still has not broken decisively out of the GOP's mediocre post-Reagan white share range. Remember, Reagan got 64% in 1984. More white share comparisons here—scroll down.

• Gallup continues (Oct 20) to show Romney leading Obama, 51-45, the same six-point margin as yesterday.

Gallup has not published its racial breakdown since Oct. 16, when it showed Romney getting a record 61% white share. That translated into a four-point national lead over Obama.

• IBD/TIPP (Oct 20) shows Romney falling further behind Obama, 44.0-46.6.

Romney's white share: still stalled at 54%.

• Rasmussen Reports now (Oct. 20) shows Romney back one point ahead of Obama, 49-48.

Rasmussen's Premium Platinum subscribers learn that Romney's white share has suddenly jumped three points since yesterday, to 59%. This seems to be the highest white share that Rasmussen has shown.

Rasmussen also offers regular state polls, enabling us to document Romney's chronic failure to mobilize Northern whites. (For example, see Washington State. See Michigan. See Ohio.)

Now, however, Rasmussen shows Romney ahead 50-46 (Oct. 20) in what the pollster defines as "Swing States":

11 key states won by President Obama in 2008 and thought to be competitive in 2012. The states collectively hold 146 Electoral College votes and include Colorado, Florida, Iowa, Michigan, Nevada, New Hampshire, North Carolina, Ohio, Pennsylvania, Virginia and Wisconsin.

In this mixture of South and North, Romney's white share is 56%.

Other recent national polls that RealClearPolitics doesn't seem to count:

• Public Policy Polling, released Oct. 19, shows Romney trailing Obama 47-49.

Romney's white share: 57%. PPP puts Obama's white share at 39%, saying he is holding Romney below what it calls "the key 20 point lead threshold with white voters."

• Ipsos/ Reuters, released October 18, has Romney trailing Obama 44-47.

Ipsos does not release its racial breakdown for Likely Voters, but told me that on Oct.7, when Romney trailed 45-47, his share of Registered Voters was a very low 50%.

• UPI/C-Voter, released Oct. 20, shows Romney ahead of Obama 48-46.

UPI apparently doesn't think the racial breakdown is worth reporting.

Footnote: a curious article in the Washington Post (Romney is winning the white vote — by a lot, by Aaron Blake, October 18 2012) makes the crucial point, verboten in the Main Stream Media, that the white vote is still the key to American politics. But it completely misstates how well (poorly) Romney is doing. There's no substitute for VDARE.com!