Republican Pat Toomey continues to lead Democrat Joe Sestak by 4 points in Pennsylvania, according to the latest Rasmussen poll of likely voters. I've written that the gap in the Pennsylvania Senate race is likely bigger than that which has been indicated by three recent polls, each of which appears to have under-sampled Republicans and/or independents. In that vein, the Rasmussen poll gives with the one hand and takes with the other. On the one hand, it supports the contention that the race likely isn't as tight as the other three polls (the average of which has called the race a tie) have suggested. On the other hand, it does clearly show the race tightening.

Rasmussen shows Toomey's lead shrinking from a 10-point advantage nine days ago to a 4-point advantage now, for two principal reasons: Previously undecided Democrats have, unsurprisingly, come home to Sestak. (Sestak's lead among Democrats was 72 to 14 percent in the earlier poll; it's now 80 to 14 percent.) And Toomey's colossal (62 to 15 percent) lead among independents has become merely a very big lead (50 to 29 percent).

As the tallies among independents suggest, on the whole the numbers still look very good for Toomey. His favorable-unfavorable ratings are noticeably better than Sestak's: 55 percent favorable to 39 percent unfavorable for Toomey; 48 percent favorable to 46 percent unfavorable for Sestak. Voters in Pennsylvania support the repeal of Obamacare by 15 points (56 to 41 percent). And by a corresponding 15-point margin, 45 percent of voters "strongly disapprove" of President Obama's performance, while only 30 percent "strongly approve" of it.

Some commentators will say that Rasmussen merely represents the conservative-leaning edge of a general trend toward this race becoming a dead-heat, but I'd say it's more likely that Rasmussen is being conservative, rather than favoring the conservative. The poll's sample includes a 3-point advantage for Democrats over Republicans (43 to 40 percent), slightly more than the 2-point advantage that exit polls showed Democrats having in Pennsylvania in 2004, when those same polls showed that nationwide turnout was split evenly between the parties. I don't think it's likely that Democratic turnout in Pennsylvania will exceed Republican turnout by 3 points, if at all, on November 2.

In case you're wondering, Democrats haven't made more noticeable gains since 2004 in Pennsylvania than elsewhere. In fact, according to the Pew Research Center, they've made slightly bigger gains nationwide (although it should be noted that all such nationwide gains were measured at just about the point that Obamacare first took center stage). The number of independents, however, has increased in Pennsylvania, from 22 percent in 2004 to 29 percent last year. Yet only 17 percent of those polled by Rasmussen are independents – slightly less than the percentage of independents who voted in the past three general elections in the state. Given Toomey's strong standing among independents, Rasmussen's potential under-sampling of them could well be understating Toomey's (admittedly shrinking) lead. In fact, it is Toomey lead among independents that would likely have to evaporate almost entirely by November 2 for Sestak to prevail.

Next Page