As last week was closing out, the folks at Tom Corbett's gubernatorial campaign released word that the state Attorney General "continues to hold big leads over three potential Democratic rivals in this year’s race for governor in Pennsylvania."

The evidence: a Rasmussen Reports poll showing Corbett up by 20-plus points over Montgomery County Commissioner Joe Hoeffel, Allegheny County Chief Executive Dan Onorato, and state Auditor General Jack Wagner. (All three Dems were within a couple points of each other -- well within the margin of error. The real loser here is Scranton Mayor Chris Doherty, who Rasmussen didn't even bother to ask voters about.) 

The race is very much in play: 14 percent of voters are undecided, and one-third of voters say they don't know enough about any of the Dems to have formed an opinion about them. And in any case, Dems will be quick to dismiss Rasmussen's findings.

The polling firm has long been subject to accusations that it has a pro-GOP bias. (See for yourself here and here -- along with a more nuanced list of grievances from Nate Silver.) While not setting store by any particular poll, I actually think the most important dyanmic here is that Rasmussen polls likely voters. As Politico recently noted:

Rasmussen, for his part, explained that his numbers are trending Republican simply because he is screening for only those voters most likely to head to the polls — a pool of respondents, he argues, that just so happens to bend more conservative this election cycle.

Polling all adults -- a method used by Gallup, another polling firm that conducts a daily tracking poll of Obama -- Rasmussen acknowledged, is "always going to yield a better result for Democrats."

OK, that's what you'd expect Rasmussen to say. But what really worries me is not Rasmussen's defense, but the rejoinder offered by his skeptics:

[C]ritics note that the practice of screening for only those voters regarded as most likely to head to the polls potentially weeds out younger and minority voters -- who would be more likely to favor Democrats than Republicans.

As a general rule, I think that's a legitimate critique -- and deciding which voters are "likely" to vote makes it easy to tweak your sample in all kinds of ways. But none of that should come as much consolation to the Democrats running for governor. These are four middle-aged white guys, and I've seen little sign that any of them have caught fire with younger or minority voters. (Though Hoeffel might come closest.) 

That enthusiasm gap might be the difference-maker come November. It certainly played a role in the recent special election in Massachusetts -- only about 15 percent of people under 30 bothered to vote there. (The article linked here relies on Rasmussen polling -- duh duh DUHHHH! -- but it's totally consistent with other statistical and anecdotal evidence I've seen.) 

Oh, and this will be a problem for Arlen Specter on the Senatorial side as well, assuming he survives the primary. I can't quantify this feeling at all, but everything I see and hear suggests that the GOP is much more motivated to oppose Specter than the Democrats are to support him. That, to me, is the most worrisome trend this year.