Outlier. Gotta be an outlier. Either that or the polls are suddenly skewed and must be uncucked forthwith.

The poll, released Thursday, found 44 percent of respondents said they would vote for Bredesen, compared to 43 percent for Blackburn…

Bredesen is bolstered in the survey thanks to support from women. Forty-nine percent of female respondents said they preferred Bredesen, while 37 percent favored Blackburn.

Fifty percent of male respondents said they preferred Blackburn, while 37 percent said they’d vote for Bredesen.

Women favor by 12 points one of the very few Democrats in the country who admitted he’d have voted to confirm Kavanaugh? What a searing betrayal of #MeToo.

How can it be that the last three polls of the state had Blackburn ahead by five, eight, and 14(!!) points, only for this one from Vanderbilt University to suddenly tilt blue? Are the previous polls outdated, perhaps? Not really: The one from the Upshot showing Blackburn ahead by 14 was conducted over almost the same period as the Vandy poll putting Bredesen up a point. Both went into the field on October 8th; the Upshot wrapped up on the 11th while Vandy wrapped on the 13th. Nothing ultra-dramatic happened on the 12th that would explain a 15-point swing. If you’re looking for a silver-bullet explanation for the difference in the results (and you’re not content with nervous “m-m-must be an outlier!” dismissals), you might find one in the samples. The Upshot poll looked at likely voters, which all surveys should be doing three weeks out from an election. The Vanderbilt poll looked at … registered voters. It may be that Bredesen is competitive with Blackburn among the general Tennessee population, just not so much with the people who’ll actually, you know, show up to vote.

But wait. There’s other evidence that the Kavanaugh effect might be — well, maybe not “fading” so much as “not really the gamechanger Republicans are hoping for.”

In the polls, provided to Secrets, it is the Democratic candidates in Indiana and Nevada who are benefiting and in West Virginia, Democratic Sen. Joe Manchin’s decision to cross party lines to back Kavanaugh does not appear to be helping him…

The poll has Donnelly over Braun 55 percent to 45 percent. Some 35 percent said Donnelly’s vote [against] Kavanaugh will make them more likely to support the Democrat, another 28 percent said it would have no impact, and 27 percent said it would make them less likely to vote for reelection…

In Nevada, the poll has Democratic challenger Rep. Jacky Rosen over Sen. Dean Heller 51 percent to 49 percent. Voters are split 34 percent to 34 percent over whether Heller’s vote to confirm Kavanaugh makes them more or less likely to support the Republican.

Those polls also look like outliers. Rosen over Heller is quite possible, as Nevada has been nip and tuck the whole way — although Heller had led the last three polls (one of them by, er, seven points) before this. Donnelly by 10 in Indiana, with fully 55 percent of the vote in the bank, seems bananas, though. Of the four polls taken in the state before this, the largest lead for Donnelly was just four points, within the margin of error. That was also the most recent poll taken before today’s. You can find good news or bad news in that as you like. The good news: Donnelly by 10 is far-fetched. The bad news: There *is* a bit of evidence that the state is trending bluer, not redder, in the wake of the Kavanaughpocalypse, with Donnelly increasing his advantage slightly. How could that be happening in a state like Indiana if the Kavanaugh effect is real?

Here’s the latest ad from Donnelly, showing off his pride at being a … conservative Republican, basically. If you’re a fan of the show “Veep” and this looks familiar, there’s a reason why.