The Republican roller coaster ride continues on to Michigan and Arizona tonight. While Arizona doesn't seem like it'll present much challenge for Mitt Romney, polling in Michigan in the last few days has shown a dead heat between Romney and Rick Santorum, and we'll probably be watching the count for a looooooong time tonight.
In Michigan, Romney and Santorum have both been hovering in the high 30s; in fact, for a brief time on Monday night, 538's rolling projection had them both tied at 37.6%. As of today, Nate Silver is projecting Romney 39.0, Santorum 38.1.
That's very handy for purposes of establishing benchmarks, because Mitt Romney also got 39% in Michigan last time, in 2008. In other words, I don't need to make any adjustments; I can simply put up 2008's numbers county by county and they can be Romney's benchmarks for viewers at home to see if he's maintaining the pace he needs to hit 39% statewide. There's, of course, a caveat that comes with that, in that Romney was running to the right of John McCain in 2008, while he's certainly to the left of Santorum this year, meaning that some of the Romney votes in 2008 might turn into Santorum votes this year, while some of the McCain votes in 2008 might become Romney votes this year, so the county numbers won't necessarily extrapolate the same.
(Continue reading under the fold)
Looking at the 2008 map, though, I'm seeing a pattern where a lot of the McCain votes may, in fact, transfer to Santorum, as the McCain areas of strength last time are the more rural, downscale parts of the state that, conventional wisdom says, will go for Santorum this time, with those voters attracted to his more social-conservative and populist pitch. Romney's areas of strength will probably continue to be the suburban arc around Detroit, especially Oakland County, where Romney's country-club positioning and family ties to the auto industry may give him a built-in advantage.
The most interesting question, and where the election probably pivots, is what happens in the Grand Rapids area (Kent and Ottawa Counties). These went narrowly for Romney last time, which stands out because most of the rest of the state's conservative SW part went for McCain. The Grand Rapids area is much more socially conservative than the Detroit suburbs, given its Dutch-American flavor and the presence of the Reformed Church (and the Church of Amway) there... but it's also more blandly middle-class than the rest of the state's smaller cities, so Romney might not lose too much ground here. Another possibility, though, is that Santorum might make up some ground in the Flint (Genesee Co.), Saginaw, and Bay City (Bay Co.) areas, blue-collar industrial areas where most voters are Democratic but where the Republicans tend to fit Santorum's more working-class profile (and also where many of them are probably Catholic).
I've added a special feature to this table, also, since half of the state's delegates will be awarded on the basis of each congressional district. Remember that they're using the new congressional districts for 2012, not the existing map (despite the fact that the state BoE will apparently be reporting results divided up by old CDs!), so now would be a good time to familiarize yourself with the new map. The interesting districts to watch will be the 2nd and 3rd (Grand Rapids area) and the 5th (Flint and Saginaw), but also the strongly-Dem districts of the Detroit area. (Romney will probably get the delegates from those ones, but with so few GOPers present, and Operation Hilarity in force, anything's possible; Nate also posits the possibility of new MI-12, with college students in Ann Arbor at one end and the Arab-American community in Dearborn at the other, giving delegates to Ron Paul.)
County |
% of 2008
statewide vote |
McCain vote
share in 2008 |
Romney vote
share in 2008 |
McCain or
Romney? |
Cong.
Dists. |
Statewide |
100.0 |
30 |
39 |
R |
-- |
Oakland |
15.3 |
27 |
47 |
R |
8, 9, 11 |
Wayne |
11.4 |
26 |
44 |
R |
11, 12, 13 |
Macomb |
8.9 |
25 |
45 |
R |
9, 10 |
Kent |
7.8 |
31 |
38 |
R |
2, 3 |
Ottawa |
4.3 |
30 |
35 |
R |
2 |
Washtenaw |
3.2 |
34 |
37 |
R |
7, 12 |
Genesee |
3.0 |
27 |
34 |
R |
5 |
Livingston |
2.7 |
26 |
45 |
R |
8 |
Kalamazoo |
2.6 |
37 |
31 |
M |
6 |
Ingham |
2.4 |
31 |
36 |
R |
8 |
St. Clair |
1.9 |
26 |
40 |
R |
10 |
Berrien |
1.7 |
36 |
28 |
M |
6 |
Saginaw |
1.7 |
27 |
43 |
R |
4, 5 |
Jackson |
1.5 |
30 |
36 |
R |
7 |
Allegan |
1.4 |
33 |
33 |
M |
2, 6 |
Muskegon |
1.4 |
36 |
33 |
M |
2 |
Monroe |
1.4 |
25 |
37 |
R |
7 |
Grand Traverse |
1.3 |
32 |
41 |
R |
1 |
Calhoun |
1.3 |
36 |
35 |
R |
3 |
Eaton |
1.2 |
30 |
35 |
R |
7 |
Lapeer |
1.1 |
24 |
35 |
R |
10 |
Midland |
1.0 |
32 |
36 |
R |
4 |
Bay |
1.0 |
27 |
43 |
R |
5 |
Lenawee |
1.0 |
25 |
35 |
R |
7 |
In the interest of thoroughness, there's also a race in Arizona tonight. It hasn't been strongly contested, with the state's large Mormon population giving a boost to Romney and all polls giving a lead to him. In addition, it's kind of pointless to try and model anyway, since a) it had the McCain favorite son factor in 2008 and b) most of the state's vote is globbed together in Maricopa County, so there's no geographic precision possible... which is fine, since Arizona is a winner-take-all state and there's little question Romney will be the winner. So, I'm just presenting the 2008 results without any special sauce added:
County |
% of 2008
statewide vote |
McCain vote
share in 2008 |
Romney vote
share in 2008 |
McCain or
Romney? |
Statewide |
100.0 |
47 |
35 |
R |
Maricopa |
64.7 |
48 |
33 |
M |
Pima |
14.3 |
47 |
38 |
M |
Yavapai |
4.6 |
49 |
31 |
M |
Pinal |
3.8 |
43 |
39 |
M |
Mohave |
3.2 |
49 |
31 |
M |
Cochise |
2.0 |
41 |
40 |
M |
Yuma |
1.7 |
44 |
33 |
M |
Coconino |
1.5 |
51 |
28 |
M |
Michigan polls close at 8 pm local time. The vast majority of the state is on Eastern time, but small parts of the Upper Peninsula are on Central Time, so statewide exit poll numbers may not be released until that point (which would be 9 Eastern, 6 Pacific). Polls in Arizona close at 7 pm local time (Arizona is on Mountain time and doesn't observe Daylight Savings, so... uh... you do the math).
With that, let's turn this over to you guys, for your predictions (and assorted other trash-talking)! Does Santorum manage to pull out the surprise victory here, thanks to a coalition of social conservatives, disgruntled blue-collar guys, and Democratic pranksters? Or does Romney ride the establishment train all the way to the end of the line? And what's the spread in Arizona?