Here’s a minor paradox for you: the most exciting and encouraging poll of a governor’s race from the last week made the polling average in that race worse. In fact, it single-handedly affected the overall bottom line; unlike last week, when the Democrats were leading in the averages in eight Republican-held states, they’re now currently leading in only seven.
Let me explain: South Dakota has the longest one-party gubernatorial winning streak in the nation. The last time they elected a Democrat as governor was in 1974. (The lucky guy was named Dick Kneip, if you’re ever looking to wow people with your trivia knowledge.) Until the last month or so, it was widely assumed that this year would be no different, and Republican Rep. Kristi Noem’s big challenge was winning the primary. Since then, though, her Democratic opponent — state Sen. Billie Sutton, a one-time rodeo star in a state where that sport is loved — issued an internal poll in late September showing him leading the race 45-42, one that was never rebutted by Noem. No public polls confirmed that, but rumors started abounding that Noem had been caught napping.
That Sutton internal poll (and an earlier one that had him narrowly losing) were the only things in our average, and that gave Sutton the lead there … but, as you know, internal polls made public are not necessarily trustworthy on their own; whether they’re intended for the media or for donors, the purpose of their purpose is narrative-setting and only the ones showing the candidate’s best foot forward get released. So it was extremely good news when on Wednesday, Mason-Dixon, on behalf of local TV affiliate KELO, released a poll showing a 45-45 tie between Sutton and Noem. It confirmed that Sutton wasn’t blowing smoke with his internal poll, and that this is a real race in one of the reddest of the states.
The downside of that release, though, is that now our South Dakota average sees the race as tied, so — on paper, at least — that made things worse. If you look at the race qualitatively, though, and not just as a poll-averaging robot, the new poll made things better; we can feel more confident that Sutton’s in a good position to actually win here.
Let’s take a look at this week’s “totem pole:”
STATE |
D CAND. |
D AVG. |
R CAND. |
R AVG. |
DIFF. |
FLIP? |
ILLINOIS |
Pritzker |
48 |
Rauner (inc.) |
30 |
+18 |
D FLIP |
COLORADO |
Polis |
50 |
Stapleton |
38 |
+12 |
|
RHODE ISLAND |
Raimondo (inc.) |
43 |
Fung |
33 |
+10 |
|
MICHIGAN |
Whitmer |
47 |
Schuette |
38 |
+9 |
D FLIP |
MAINE |
Mills |
45 |
Moody |
38 |
+7 |
D FLIP |
FLORIDA |
Gillum |
49 |
De Santis |
43 |
+6 |
D FLIP |
MINNESOTA |
Walz |
46 |
Johnson |
40 |
+6 |
|
NEW MEXICO |
Lujan Grisham |
47 |
Pearce |
42 |
+5 |
D FLIP |
CONNECTICUT |
Lamont |
43 |
Stefanowski |
38 |
+5 |
|
OREGON |
Brown (inc.) |
43 |
Buehler |
39 |
+4 |
|
WISCONSIN |
Evers |
48 |
Walker (inc.) |
45 |
+3 |
D FLIP |
IOWA |
Hubbell |
39 |
Reynolds (inc.) |
37 |
+2 |
D FLIP |
GEORGIA |
Abrams |
46 |
Kemp |
46 |
0 |
|
KANSAS |
Kelly |
40 |
Kobach |
40 |
0 |
|
OHIO |
Cordray |
40 |
De Wine |
40 |
0 |
|
SOUTH DAKOTA |
Sutton |
44 |
Noem |
44 |
0 |
|
NEVADA |
Sisolak |
41 |
Laxalt |
45 |
-4 |
|
OKLAHOMA |
Edmondson |
41 |
Stitt |
49 |
-8 |
|
ARIZONA |
Garcia |
40 |
Ducey (inc.) |
50 |
-10 |
|
NEW HAMPSHIRE |
Kelly |
38 |
Sununu (inc.) |
50 |
-12 |
|
MARYLAND |
Jealous |
35 |
Hogan (inc.) |
54 |
-19 |
|
SOUTH CAROLINA |
Smith |
34 |
McMaster (inc.) |
56 |
-22 |
|
ALASKA |
Begich |
?? |
Dunleavy |
?? |
?? |
|
As I said, there are seven races that the Democrats are poised to flip, at least according to the current polling averages. But here’s an important caveat; there are four more races in Republican-held states that are precisely tied today in our averages: not just South Dakota, but also Georgia, Kansas, and Ohio. The smallest shift in the prevailing winds could push the Democratic haul up to 11, which would tie 1994 as the largest number of flipped states in the post-WWII era. (Even seven is extremely good, and the kind of result that only happens in big waves.)
The other one to add the ranks of the tied races is Georgia, which moved up only slightly from a 2-point deficit for Stacey Abrams (thanks to a 48-48 tie in an Opinion Savvy poll released this week). Georgia also has an unusual quirk you might not know about, that it shares with only (sort of) Louisiana: if neither Abrams nor Republican opponent Brian Kemp can round up enough undecided voters to hit 50 percent, they’ll proceed to a runoff which would be held on December 4. (This is possible thanks to the presence of a Libertarian on the ballot this year.) Runoff elections in Georgia are rare — and, when they’ve occurred, in the last few decades they usually haven’t gone well for Democrats, as turnout falls off a lot from the general election. Though, given the general pro-Democratic enthusiasm gap that we’ve seen in polls this year, it seems possible that turnout trend could wind up benefitting Abrams.
Georgia was by no means the biggest gainer this week, on the Democratic side, though. Florida saw Tallahassee mayor Andrew Gillum bouncing up from a 1-point lead in the averages last week to 5 points this week, thanks to a trio of new polls (from CNN, Quinnipiac, and SurveyUSA) all showing both Gillum and Senator Bill Nelson posting larger-than-usual leads. Florida is the most-heavily polled race in the country; there have been 29 polls since the primary in late August, and Gillum has led in 28 of them (with the sole exception being a Republican internal poll from OnMessage).
Finally, there’s one state that’s a big question mark on the table — literally, as you can see above. It’s Alaska, which got shaken up significantly when unpopular independent incumbent governor Bill Walker withdrew from the race at the eleventh hour and endorsed Democratic candidate Mark Begich, with whom he was previously splitting the non-Republican vote. That leaves us with nothing to average, as there have been no polls released since Walker’s dropout (that meet our standards) polling the two-way matchup between Begich and Republican Mike Dunleavy. (There was one poll that surfaced showing Dunleavy leading only narrowly, but that was a message-testing poll, which we can’t use.)
The reduction of the race from a three-way car wreck to a standard D vs. R matchup certainly improves Begich’s odds in Alaska. It remains to be seen, though, whether Begich can convert enough of Walker’s supporters to overcome Alaska’s reddish hue, and we’ll be eagerly awaiting some sort of helpful data from this very difficult-to-poll state.