The good news from the last week of Senate polling is that Jon Tester, the incumbent Democratic senator from Montana, is looking in pretty good shape as we enter the closing weeks. Instinctively, going into this cycle his situation felt more like Joe Manchin’s situation (in West Virginia) than, for instance, Claire McCaskill in Missouri or Joe Donnelly in Indiana: he’d been around for a long time and had established his own “brand,” in a state that still does a lot of ticket-splitting. And more importantly, he didn’t luck into a win in 2012 by drawing a hopelessly flawed opponent (like McCaskill or Donnelly did … though in McCaskill’s case, it may have been more skill than luck, thanks to some helpful meddling in the GOP primary).
The problem was, there just wasn’t much polling evidence for that proposition. Of course, there wasn’t much evidence that he was in trouble, either; except for perhaps North Dakota, this was one of the least-polled of the competitive Senate races. Over the summer, all we saw here was a handful of polls from either online pollsters of questionable reputation, or Republican internal pollsters trying to signal to donors that his Republican opponent, state auditor Matt Rosendale, was a good investment.
Surprisingly, we still haven’t seen any polls of Montana from any national-level “gold standard” pollsters. At the very least, though, the last few weeks saw polls released by local universities, the University of Montana and Montana St. University-Billings, which gave Tester a lead of 10 and 9 points, respectively. That pushes his lead in our polling average up to 8 points this week, up from a 3-point lead at this point last week.
We’re not really sure how far to trust the University of Montana poll; their previous poll of the race, from August, had visible stink lines radiating off it (it spent over two weeks in the field, and came back with a 24-point lead for Tester, completely unlikely any other poll’s results). But their newest poll spent a more reasonable one week in the field, and moreover it came up with almost exactly the same result as the MSU poll which was taken simultaneously. So we’re feeling, let’s say, uncomfortably confident about Montana being more securely in the Democratic column.
In fact, an eight-point lead in Montana moves it ahead of two races that most people really aren’t that concerned about and that we’re keeping in the “Likely Democratic” column in our qualitative ratings: the Senate races in New Jersey and the Minnesota special election (for Tina Smith to fill out the rest of her term, after having been appointed to fill Al Franken’s seat). Our polling average puts New Jersey’s senator Robert Menendez up seven points, and Smith up six points.
The Menendez race, in particular, has given some people heartburn in the last week, when the Senate Majority PAC (the super PAC linked to Senate Democratic leadership) spent several million dollars on advertising against Menendez’s opponent. That kind of spending doesn’t really seem merited based on the publicly-released polls, so one possible assumption is that the campaigns are seeing much closer results than what the public pollsters are seeing, and they’re having to spend millions on Menendez that they’d really rather spend in a different race.
The other possibility, though, is that one or several deep-pocketed donors who want to remain anonymous and/or avoid individual contribution limits are simply using Senate Majority PAC as a pass-through for a pro-Menendez ad buy; in other words, it’s money that was given specifically with the condition that it be used on Menendez’s behalf and isn’t coming out of SMP’s general pool of funds to help more-endangered candidates. We don’t know for sure that’s what’s happening, but there’s some precedent for that very thing in New Jersey politics, such as when House Majority PAC (the House equivalent to SMP) spent significant money in 2014 on Democratic Rep. Donald Norcross’s election in New Jersey’s dark-blue 1st district.
Smith’s seat, by contrast, has generated almost no heartburn on anyone’s part, despite several polls in the last few weeks finding a mid-single-digits race. Here, the lack of interest on the part of the DSCC or NRSC (or SMP, or its GOP-equivalent super PAC, the Senate Leadership Fund) suggests that, no, the parties aren’t seeing a race that close in their internal polling. In a different year, Smith could’ve been in some real trouble (senators can be at their most vulnerable when they’ve been appointed and are running for the first time; see, for instance, Luther Strange in Alabama last year, who couldn’t even survive his primary). But fellow senator Amy Klobuchar, and Rep. Tim Walz, in an increasingly uncompetitive gubernatorial race, seem to have the coattails to carry Smith over the line this year.
The bad news, as the headline alludes to, though, is that Indiana’s Democratic incumbent senator, Joe Donnelly, has fallen narrowly behind his Republican opponent, state Rep. Mike Braun. The current average puts Braun up by two, thanks to two polls this last week giving small leads to Braun: one from YouGov and one from a little-known Republican pollster called Mason Strategies.
Somewhat like Missouri (which I talked about last week), there’s a lot of filler in the Indiana trendlines, though. Most of what’s there is either Republican internal polls, or polls by online-only pollsters. The only polls from what you’d think of as “gold standard” pollsters (SurveyUSA, Fox News, and Marist) have all given Donnelly a small lead, which is an encouraging factor.
That, by itself, isn’t a reason to think that the average is necessarily wrong, however. If you’d asked me offhand, at the start of the cycle, which Democratic incumbent was the most vulnerable, I probably would’ve gone with Donnelly (rather than McCaskill or Heitkamp), not just because of Indiana’s red-state status but also because so much of his 2012 victory was dependent on drawing the hapless Richard Mourdock as an opponent. So Donnelly’s consistent (small) leads so far might not have been built on a solid-enough foundation (similar to the leads Phil Bredesen was posting in Tennessee until lately), and we might be seeing previously-undecided Republican voters finally reverting to form, as one might expect, in the final weeks. On the other hand, though, unlike what we’ve been seeing in Tennessee, Braun is still far from the 50 percent mark.
Here’s this week’s totem pole:
STATE |
D CAND. |
D AVG. |
R CAND. |
R. AVG. |
DIFF. |
FLIP? |
OHIO |
Brown (inc) |
50 |
Renacci |
34 |
+16 |
|
WISCONSIN |
Baldwin (inc) |
54 |
Vukmir |
40 |
+14 |
|
WEST VIRGINIA |
Manchin (inc) |
49 |
Morrisey |
38 |
+11 |
|
MONTANA |
Tester (inc) |
49 |
Rosendale |
41 |
+8 |
|
NEW JERSEY |
Menendez (inc) |
49 |
Hugin |
42 |
+7 |
|
MINNESOTA (SP.) |
Smith (inc.) |
47 |
Housley |
41 |
+6 |
|
ARIZONA |
Sinema |
46 |
McSally |
44 |
+2 |
D FLIP |
FLORIDA |
Nelson (inc.) |
46 |
Scott |
45 |
+1 |
|
RED LINE |
RED LINE |
|
RED LINE |
|
|
|
INDIANA |
Donnelly (inc.) |
42 |
Braun |
44 |
-2 |
R FLIP |
NEVADA |
Rosen |
43 |
Heller (inc.) |
46 |
-3 |
|
MISSOURI |
McCaskill (inc.) |
44 |
Hawley |
48 |
-4 |
R FLIP |
TEXAS |
O’Rourke |
45 |
Cruz (inc.) |
50 |
-5 |
|
TENNESSEE |
Bredesen |
42 |
Blackburn |
50 |
-8 |
|
NORTH DAKOTA |
Heitkamp (inc.) |
39 |
Cramer |
55 |
-16 |
R FLIP |
Mississippi (SP.) |
Espy |
27 |
Hyde-Smith + McDaniel |
49 |
-22 |
|
Taken together, we’re back to the point where the Democrats are poised to lose net seats. That’s partly because of Donnelly, but also because the Nevada Senate race, which was seen as a tie in the averages last week, has dropped back to a small lead for Republican incumbent Dean Heller. As I’ve said before, the polls might be underestimating Democratic nominee Jacky Rosen, who will depend heavily on Latino turnout (which may not be showing up in the polls because of the difficulty involved in contacting Spanish speakers, but who have a well-organized GOTV operation through the powerful Culinary union). (Keep in mind, of course, that they also might not be underestimating her support.)
And we’re at the point in the cycle where, at least in Nevada, early voting numbers start to take on more importance than the polls. Local expert Jon Ralston seem to think the current numbers, as of Sunday, are fairly encouraging for the Democrats, though there’s a predictable ebb-and-flow to the voting patterns there, and the question remains as to whether the Democrats have built up enough of an early voting edge this week to hold back Republican momentum later.