Will Super Tuesday be the last stand for the not-Romneys? Well, maybe not all of them, as there's no real downside to Ron Paul and Newt Gingrich continuing to stumble along, zombie-like, on their nationwide lecture tours. But it's looking like Rick Santorum's last chance to make an impact that will change the Romney-inevitability narrative... and that chance seems to be slipping away by the day, with Mitt Romney pulling into the polling lead in Ohio and starting to come close in Tennessee.
Ohio isn't the biggest prize of the night (that would be Georgia, which seems poised to give a big margin to hometown favorite Gingrich), but it's the second biggest delegate haul, and has really become the crux of Santorum's hopes. He'd been leading by wide margins in polls here pre-Michigan, but thanks to the reversal of momentum (and a lot of self-inflicted damage to himself, and a whole lotta Romney ad dollars) he's been losing ground ever since; currently Nate Silver at 538 is projecting Ohio to go 37 Romney, 35 Santorum.
Ohio is not winner-take-all: Ohio gives a much bigger share (15) of its 63 delegates on a statewide basis than did Michigan; the winner would get all 15 only in the unlikely event that he crests 50%. If he doesn't, everyone who finishes above 20% (which will probably only be Romney and Santorum) get a proportional share. The remaining delegates are allotted to the winner of each of the 16 Congressional districts under the new map, at 3 apiece.
As for the CDs, unless he continues to accelerate his collapse, look for Santorum to win some of the more rural CDs. The 5th seems the likeliest for a Santorum win; looking at the 2008 map, the state's rural NW corner was Mike Huckabee's best area. The 4th and 8th in the state's western flatlands, and the 6th in its Appalachian-flavored southeast, also seem part of the Santorum demo (blue-collar white and/or evangelical Republican voters).
And here's the model. Note that it's a Santorum-based model, not a Romney-based one; that's largely because Ohio wasn't a Super Tuesday state last time, and Romney had already dropped out, getting only a few percent in the state. John McCain won easily, beating Huckabee one-on-one 60-31 and winning every county. I don't think Romney can hope to get anywhere near 60% and McCain isn't the best proxy for him anyway (since Romney was running to McCain's right last time)... but Huckabee is an excellent stand-in for Santorum, so I'm projecting 2012 Santorum numbers up from the "other" number (which is almost entirely Huckabee votes, though it also contains a few Fred Thompson and Ron Paul strays). "Other"'s 37% is close to the 35% that Santorum is projected to get, and that Santorum would need to hit, in order to have any shot at beating Romney. If Santorum is falling short of that mark on Tuesday evening, the fork can be prepared for sticking.
| County | % of 2008 statewide vote |
McCain vote share in 2008 |
Other vote share in 2008 |
For Santorum to hit 35% |
Cong. Dists. |
|---|---|---|---|---|---|
| Statewide | 100.0 | 60 | 37 | 35 | -- |
| Franklin | 7.7 | 65 | 32 | 30 | 3, 12, 15 |
| Hamilton | 7.3 | 67 | 30 | 28 | 1, 2 |
| Cuyahoga | 5.7 | 73 | 24 | 22 | 9, 11, 14, 16 |
| Montgomery | 4.3 | 58 | 39 | 37 | 8, 10 |
| Stark | 3.8 | 60 | 37 | 35 | 7, 16 |
| Butler | 3.7 | 58 | 39 | 37 | 8 |
| Summit | 3.6 | 63 | 33 | 31 | 11, 13, 14, 16 |
| Warren | 2.6 | 56 | 41 | 39 | 1 |
| Clermont | 2.6 | 59 | 38 | 36 | 2 |
| Lucas | 2.3 | 59 | 35 | 33 | 5, 9 |
| Delaware | 2.3 | 64 | 33 | 31 | 12 |
| Greene | 2.3 | 56 | 41 | 39 | 10 |
| Licking | 2.1 | 59 | 38 | 36 | 12 |
| Fairfield | 2.0 | 61 | 37 | 35 | 10, 15 |
We'll look at Tennessee over the flip...