Via an approving tweet from Ezra, Public Policy Polling makes much of the idea that Sarah Palin runs much worse than other Republicans against Barack Obama in Texas.
Look, this kind of polling is just meaningless. Republicans are extremely unlikely to nominate a candidate who is unpopular among Republicans, including Republican-leaning independents. Not just because such a candidate won't win primaries and caucuses -- but because such a candidate that does win will receive highly favorable coverage from the Republican partisan press, and therefore become popular among most Republicans.
It is, I suppose, theoretically possible that a party could nominate a candidate intensely supported by 51% of the party but hated by the other 49% so much that even positive coverage in the partisan press won't matter; it's even theoretically possible that a factional candidate with an even smaller base of intense supporters could win a nomination with even less support if that faction turns out to vote in disproportionate numbers. However, nothing like that has happened in the last thirty years, and there is good reason to believe that it is at the very least highly improbable.
Now, it is of course true that there may be some variation in how popular a nominee might be, but for out-parties the odds are that it won't make a whole lot of difference. Presidential approval matters a whole lot more than how well-liked the challenger is.
As long as I'm at it, the even more obvious reason to ignore a lot of this kind of polling is that while it may be theoretically possible that a party could nominate a candidate who remains unpopular, it certainly isn't possible for a party to nominate a candidate who remains unknown. Even testing Mitt Romney at this point is probably going to undervalue his strength because low-information voters may not recognize his name and therefore be reluctant to say they'll vote for him, and that's even more the case with lesser-known candidates.
Bottom line is that head-to-head general election polling at this point just can't do what PPP wants, which is to tell us which candidates will do better against Obama in fall 2012 -- because we really can't know what the nominee will look like after another year of invisible primary, after Iowa and New Hampshire, after the rest of the primaries and caucuses, after spending months as the nominee pre-convention, and after getting the three day propaganda-fest that is the modern national party convention.