The new PPP Michigan poll has some bad news for Santorum:

Romney will go into election day with a large lead in the bank. Only 16% of Michigan voters say they’ve already cast their ballots, but Romney has a whooping 62-29 advantage over Santorum with that group. Santorum actually leads Romney 39-34 with those who are planning to cast their votes on Tuesday, but he’d need to win election day voters by even more than that to neutralize the advantage Romney’s built up.

The last week of the campaign in Michigan has seen significant damage to Santorum’s image with GOP voters in the state. His net favorability has declined 29 points from +44 (67/ 23) to now only +15 (54/39). Negative attacks on Romney meanwhile have had no negative effect with his favorability steady at +20 (57/37) [bold mine-DL]. Two weeks ago Santorum’s net favorability in Michigan was 34 points better than Romney’s. Now Romney’s is 5 points better than Santorum’s.

Santorum boosters might object that this is what was supposed to happen all along. The spin goes something like this: of course Romney was always going to win Michigan, but the outcome should never have been in doubt, and the fact that it was briefly in doubt should make Republicans elsewhere realize that Romney is a bad candidate. On the other hand, the fact that Santorum was briefly leading in Michigan with extremely strong favorability and then managed to lose the lead and see his favorable numbers collapse suggests that Santorum is an even worse candidate than Romney. Santorum has been imploding as completely and rapidly as Gingrich did in Florida, and like Gingrich a lot of the damage has been self-inflicted.

P.S. Contrary to all assumptions about Santorum’s ready-made appeal to Michigan laborers, Romney is winning members of union households by eleven points.