It’s our second-to-last installment in the Daily Kos Elections Senate Forecast, and the top lines still haven’t changed in the last few weeks: Democratic candidates still have leads in the Arizona, Colorado, Iowa, Maine, and North Carolina races, which is adequate to take back a Senate majority. The Iowa and North Carolina races are still showing relatively close margins, but those are two are heavily polled races, where almost all polls from high-quality pollsters show something very similar, so it’s OK to feel pretty confident about those two as well.
There are, however, two other races where Democrats are playing defense, that have been mostly afterthoughts for the last few months, that have suddenly shown some interesting movement in the last week. One is in Minnesota, where most previous polls showed a Democratic advantage of around ten points, but a couple of surprisingly close polls have popped up recently. The other one is Alabama, which has been an afterthought simply because most polls have shown Democratic incumbent Doug Jones getting crushed; however, a new poll from the Jones campaign this week giving him a narrow lead had a big impact on our averages.
Alabama Senator Doug Jones, as you hopefully remember from the 2017 special election, is in the Senate representing one of the country’s reddest states largely because he had the good fortune to run against Roy Moore, who’s terrible enough to actually lose an election in Alabama while running as a Republican. It’s been taken as something of an article of faith that Jones would probably lose re-election if he faced off against a more normal Republican who, for instance, hadn’t been kicked off the state Supreme Court and who hadn’t been banned from his local mall for being a sex pest.
This time, Jones is running against former college football coach Tommy Tuberville, who generally comes off as a standard-issue Republican, despite showing only a mild familiarity with actual policy issues and a personal financial track record of failed hedge funds and questionable charities. And, indeed, most polls have shown Jones losing to Tuberville by double digits. (Daily Kos Elections has rated this race from the outset as “Likely Republican,” a nearly unheard-of position for a race with an incumbent.)
However, the race got an unexpected jolt last weekend when the Jones campaign released an internal poll from Democratic firm FM3 giving Jones a 48-47 lead over Tuberville. Taken together with all the other polls in our average, that still wasn’t enough to put Jones into the lead, though it did push Tuberville’s lead down to low single digits (3 points).
There are several reasons you should take that with a little salt, though. One is that public polls haven’t shown anything like that. One caveat to that caveat is that this a rarely-polled race (since Alabama obviously doesn’t have a competitive presidential race); the most recent public poll of the race, from Auburn University at Montgomery — which gave Tuberville a 54-42 lead — had field dates from Sept. 30 to Oct. 3, so it’s a legitimate question to wonder whether something has genuinely changed in this race.
One thing we always advise at Daily Kos Elections, though, is if a candidate drops a surprisingly good internal poll, to wait and see what the other candidate does. If there’s radio silence on the other end, that’s usually a good indication that the internal poll is close to the mark. (In fact, that’s one of the main ways we evaluate House races, since often the only polls you ever see in House races are the ones released by candidates themselves.) If the other candidate responds with a poll showing the direct opposite, that tends to call the first poll into question. And that’s what happened in Alabama; Republicans responded by a poll from GOP firm Moore Information giving Tuberville a 55-40 lead.
The below-the-radar race where Democrats are playing defense and the movement hasn’t been positive is in Minnesota, where incumbent Tina Smith faces off against Republican ex-Rep. Jason Lewis. Smith won a special election in 2018 easily enough (by 11 points), after having been appointed to replace Al Franken; this race is for a full term.
Smith had usually been maintaining a low-double-digits lead lead over Lewis in our averages, basically no different than in 2018. Last week the average fell to +8, which seemed like a bit of a blip at the time, but two subsequent polls have now pushed her lead down to five points, in the same spot as the Michigan race where Democratic incumbent Gary Peters faces a competitive race … and that’s without Trafalgar putting their thumbs on the scale they way they have in Michigan averages.
One of those polls was a SurveyUSA poll that gave Smith only a 43-42 lead (though compared with most polls, that’s with a strangely high number of undecideds and third-party voters). Ordinarily, I’d shrug that off and throw it in the heap with all the other polls, but SurveyUSA is an unusually good pollster (though not immune to the occasional outlier), and there was also a Change Research poll in a similar timeframe giving Smith a slightly-better 48-44 lead.
One caveat is that Change Research isn’t a pollster we’re very fond of; they’re only a few years old and put up some erratic results during the Democratic presidential primary process. In addition, other pollsters that we trust have found the same old low double-digits leads in the same time period; Civiqs, for instance, saw the race at 54-43 last week. And this race isn’t polled with anywhere near the same frequency as Michigan, since Minnesota is on pretty much the outer bound of being a swing state. (The same SurveyUSA poll also had Joe Biden with “only” a six-point lead, suggesting maybe that SurveyUSA stumbled into an unusually-red sample.)
The main reason, though, that the Minnesota race probably isn’t as much of a problem as our polling average makes it to be, is that the alphabet committees aren’t acting like this is a top-tier race. In other words, the DSCC and the NRSC, and the leadership Super PACs that mirror them (Senate Majority PAC and Senate Leadership Fund) aren’t jumping into this race with advertising buys. That contrasts with Michigan, where SMP and SLF have been hitting the airwaves. (In fact, it even contrasts with the New Jersey Senate race in 2018, where there was last-minute SMP money deployed, but which turned out to be a nothingburger in the end, as New Jersey races usually do.) The committees see much more extensive private polling than we do, and as I’ve reiterated many times, the best way to get a sanity check on the polls is to follow where the big money is going, and at least so far, it’s not going to Minnesota.
With that in mind, let’s take a look at the whole “totem pole”:
STATE |
DEMOCRAT |
D AVG. |
REPUBLICAN |
R AVG. |
DIFF. |
FLIP? |
NEW HAMPSHIRE |
Shaheen (i) |
54 |
Messner |
39 |
+15 |
|
COLORADO |
Hickenlooper |
51 |
Gardner (i) |
40 |
+11 |
D FLIP |
ARIZONA |
Kelly |
51 |
McSally (i) |
41 |
+10 |
D FLIP |
MAINE |
Gideon |
47 |
Collins (i) |
41 |
+6 |
D FLIP |
MICHIGAN |
Peters (i) |
48 |
James |
43 |
+5 |
|
MINNESOTA |
Smith (i) |
46 |
Lewis |
41 |
+5 |
|
IOWA |
Greenfield |
47 |
Ernst (i) |
44 |
+3 |
D FLIP |
NORTH CAROLINA |
Cunningham |
47 |
Tillis (i) |
44 |
+3 |
D FLIP |
|
|
|
|
|
|
|
GEORGIA |
Ossoff |
45 |
Perdue (i) |
45 |
0 |
|
KANSAS |
Bollier |
43 |
Marshall |
44 |
-1 |
|
SOUTH CAROLINA |
Harrison |
45 |
Graham (i) |
46 |
-1 |
|
MONTANA |
Bullock |
47 |
Daines (i) |
48 |
-1 |
|
MISSISSIPPI |
Espy |
42 |
Hyde-Smith (i) |
43 |
-1 |
|
ALASKA |
Gross |
44 |
Sullivan (i) |
46 |
-2 |
|
ALABAMA |
Jones (i) |
45 |
Tuberville |
48 |
-3 |
R FLIP |
TEXAS |
Hegar |
42 |
Cornyn (i) |
47 |
-5 |
|
KENTUCKY |
McGrath |
39 |
McConnell (i) |
51 |
-12 |
|
Another reason for Democratic optimism is that right below the Red Line there are half a dozen races where the Democratic challenger is tied or down just a point or two, and where a strong tailwind on Election Day could be the difference maker between a gain of “only” four or five seats versus eight or nine. With the exception of Georgia, these, however, are states where the presidential race isn’t expected to be close, so these candidates face the disadvantage of not really being able to rely on coattails to pull them across the finish line. (Of course, the Republican failures at the top of the ticket nevertheless had a lot to do with these races getting to the point where they’re competitive in the first place.)
As we discussed last week, there isn’t much time left on the clock for David Perdue’s racist gaffe in the Georgia Senate race to take effect, and based on the current polling average, it still hasn’t had much of an impact yet. Democratic opponent Jon Ossoff has pulled back into a tie with Perdue after a two-point deficit last week, but that may simply be normal within-the-margin-of-error bumping around. Meanwhile, the second race in Georgia, for the special election, is currently Democrat Raphael Warnock at 32, appointed Republican incumbent Kelly Loeffler at 22, Republican Doug Collins at 21, and Democrats Matt Lieberman and Ed Tarver at 7 and 3.
Georgia’s election law requires a runoff when no candidate reaches 50% in the general election, so it’s possible, if not likely, that both races will proceed to overtime in early January. If Senate control is still hanging in the balance at that point, fundraising would be absolutely epic in terms of fundraising and national attention. However, with Democrats still looking likely to net at least four pickups on Election Day, it’s probable control of the Senate will have already been decided at that point.