Less than a month ago, Jazz Shaw wondered whether Client #9 could be stopped.  Eliot Spitzer had a 17-point lead over Scott Stringer in the New York City Democratic primary race for comptroller and presumably a clear path to a political comeback.  In just three weeks, though, his nineteen-point lead over Stringer (56/37) in the Quinnipiac poll series has flipped to a 45/47 deficit — a change of 21 points in the gap in nearly as many days:

The Democratic primary for New York City Comptroller is too close to call, with 47 percent of likely primary voters for Manhattan Borough President Scott Stringer and 45 percent for former Gov. Eliot Spitzer, according to a Quinnipiac University poll released today.

This compares to a 46 – 46 percent dead heat in an August 29 survey by the independent Quinnipiac (KWIN-uh-pe-ack) University and a 56 – 37 percent Spitzer lead in an August 14 survey.

It’s an astounding turnaround in a race that most people assumed Spitzer has sewn up in August. The collapse of Spitzer’s commanding lead comes out of nowhere, too; Spitzer hasn’t committed any major gaffes, and his earlier scandals have been well known all along. The result is technically a virtual tie, as it falls within the margin of error, but having the kind of name recognition Spitzer has as former governor and media star and only 45% against a local politician is a very bad sign with the primary just five days away.

Quinnipiac’s Maurice Carroll points out that the African-American vote is all that’s keeping Spitzer viable, with a 61/32 lead over Stringer.  However, that’s probably a bad sign for Democrats if Spitzer wins, because the Republican nominee is an African-American conservative with a sterling resumé in the field, John Burnett, who might do better with that demographic than Spitzer in a general election. Spitzer also does better with voters under 50 and strong Democrats, but he’s getting clobbered among those with weaker party affiliation.

There’s more bad news for Spitzer in voter loyalty. Almost twice as many of Spitzer’s voters say there is a good chance they’ll switch than Stringer’s, 17% to 9%.  Stringer’s voters are more likely to claim their choice as definite, too, 81% to 74%.  With five days left, Spitzer has more downside and is perhaps already trailing.

What happened?  Absent any particular event in the past three weeks, it may be that voters just initially identified with Spitzer because of name recognition until they got comfortable with Stringer.  The novelty may have worn off, too, and with it the hipness factor of supporting a disgraced politician looking for a comeback.  However, I’d bet that the continuing antics of Anthony Weiner have done a lot of damage to the comeback narrative in New York City, and had voters asking themselves why they’d bother to support disgraced former officeholders at all.  Weiner’s implosion may have taken Spitzer’s candidacy down with him.