Odd polling trends: USA

Over on WSJ's "Best of the Web Today", James Taranto asks why Kerry isn't gaining ground on Bush despite "right track/wrong track" and approval numbers that would normally mean trouble for an incumbent.

Yesterday he found answers in some other poll numbers -- very low "strong support" numbers for Kerry among his own supporters (38% vs. 68% for Bush), and an indication that the Democrats are overplaying their hand (60% of voters think Rumsfeld should stay on the job). And there's a great Kerry quote that illustrates the personality problem he has with voters.

Today Taranto presents some readers' explanations for the phenomenon. Here's a sample from Ray Newton:

I think that you are missing one important point on the polls. If a pollster asked me if I approve of the job Bush is doing I would have to say no. Too apologetic, not strong enough.

Do I approve of his handling of Iraq? Again no. Need to get tougher.

Do I approve of his handling of the economy? Again no. Too much spending. Too much appeasing the Dems. Tax cuts must be permanent.

For all of these reasons Kerry is a much worse choice. That is why polling can be very confusing. When you disagree, they never ask if you tend more towards the right or the left.

Taranto breaks the last six elections involving an incumbent into two categories -- in 1972, 1984, and 1996, where the incumbent was a polarizing figure with strong opposition from the other party but solid support in his own party; in 1976, 1980, and 1992, where the incumbent was not as polarizing a figure, but had unenthusiastic support from his own party, and was challenged for the nomination. Incumbents in the first category won landslide victories, incumbents in the second category all lost. Taranto puts W. in the first category and says that this election comes down to Bush vs. anti-Bush, and if you're not already a Bush-hater, it's unlikely that anything is going to happen in the next six months to make you one.