Worksheet for the 2018 Midterms (If There Is a Blue Wave, What Next for Democrats?)

Subscribe to Post Comments

43 comments

1. apberusdisvet October 25, 2018 at 8:17 am

   A lot of the analysis seems to be based upon polling which is highly problematic; especially given the 2016 polling fiasco. Too many of the 2016 polls were based upon disingenuous weighting of D responses over Rs; certainly a thumb on the scale to suppress the R vote. In this election, the early voting indicates a greater R response than the polls suggest, especially in the State of Florida.
   1. Yves Smith October 25, 2018 at 8:49 am

      No, the polling in 2016 was accurate. It showed Trump as anywhere from even with Hillary to as much as down 11 on a national basis, with Hillary typically having a 2 to 4 point lead with the margin of error being 2 to 4 points depending on the poll. Hillary was up 4 post Labor Day but the margin tightened in the ten days before the election to IIRC 2 points. She won the popular vote by 3 points, which was completely line with the polls.

      Moreover, Lambert was in a decided minority in focusing on electoral college paths to victory. The supposed experts for the most part couldn’t be bothered to look at both hard data and telling anecdotes in swing states, like turnout at Trump rallies. As we know now, Trump was keenly focused on what it took to win, which was victory in the swing states and used his private jet to great advantage to do far more trips to battleground states, and then also in formats (rallies) designed to get local media coverage, while Clinton too often did her wrapped in tissue paper routine and met in small and/or elite groups.

      Lambert was updating the electoral college game close to daily in Water Cooler. He saw Trump go in the six weeks before the election from having hardly any electoral college paths to victory to having many new routes open up as election day approached. Lambert kept warning that a Trump win was way more likely than MSM reports were suggesting.

      Shorter: Lambert has a very solid track record and you diss him at your peril.
      1. XXYY October 25, 2018 at 10:40 am

         As long as we’re on the subject of polling: The conventional wisdom is that predictive polling is somewhat unreliable because turnout is hard to predict; one can know with great accuracy how a particular segment will vote, but be wildly wrong on what fraction of the electorate consists of that segment. The wisdom is also that exit polling is the gold standard because it includes actual turnout effects.

         Is this in fact what we actually saw in 2016? I have never read a.clear conclusion or report on this.
         1. Skip Intro October 25, 2018 at 2:57 pm

            Raw polling information must always be weighted by demographic turnout models to make predictions. These models are rarely discussed but contain a number of important assumptions. I believe that the assumptions of Dem. turnout were based significantly on the 2008 and 2012 presidential elections, which everyone but the Clintongang/DNC recognized had a very different dynamic from the 2016 election which featured the two of the least liked/trusted candidates ever to run for the office. These mistaken assumptions, combined with the use of polling as performative rather than predictive, led to massive complacency. It is the embodiment of the Democrat habit of taking certain demographics for granted.

            I believe this was confirmed by various studies, including exit polls, which showed depressed turnout and a number of Obama voters going for Trump. If you take places like Detroit for granted, after their experience, then you are asking for unpleasant surprises.
   2. Lambert Strether Post authorOctober 25, 2018 at 2:22 pm

      > A lot of the analysis seems to be based upon polling

      I think the best way to think about the method I’m using is that it’s a meta-analysis of the work of horse-race analysts. (I like Inside Elections and Cook Political Report a lot better than FiveThirtyEight because they’re openly subjective, instead of wrapping their judgment calls in a numeric veneer.) Much like horse race touts, they’re not only using polls as input, but Beltway gossip, reporting from the ground, contacts in the campaigns, and good old seat-of-the-pants judgement. I picture them smoking cigars. So it’s not just polls.)

      I should also say, because the caveats I wrote in didn’t seem to do the job, that my main focus is institutional. I want to use these midterms as a lens to understand the institutional structure of the Democrat Party, which is a harder problem than one might think. So, Tables 2 and 3 are the results that are important to me, for that project. However, people want to know about the horserace! So I put that material first, adding the caveat that it’s repurposing a sample that really was not meant for that purpose. (The districts mentioned in Note [1], for example, may provide the Democrat margin of victory, which I’m not seeing because my sample does not see it.)

      Two weeks is a long time in politics! We’ll see what happens!
2. a different chris October 25, 2018 at 8:47 am

   Ok I’ve been too embarrassed to ask this question for like a decade now, but here it is:

   Does the rating go “likely, lean, tilt, toss-up” or “likely, tilt, lean, toss-up”???
   1. Lambert Strether Post authorOctober 25, 2018 at 2:25 pm

      I have to mentally doublecheck this myself; it triggers my dyslexia. It is:

      Toss-up -> Tilt -> Lean -> Likely (-> Safe)
3. allan October 25, 2018 at 9:04 am

   FiveThirtyEight has pointed out that there are four races that are at least somewhat competitive
   but (as of Oct. 22) have seen no public polling, and none of these are on Lambert’s list:

   CO-3 Toss-up
   IN-9 Lean R
   NY-2 Lean R
   WI-6 Lean R

   The candidate in NY-02 Liuba Gretchen Shirley, supports Medicare for All:

   I would support HR 676 Medicare for All. Health care is a human right.

   There may be some surprises on Nov. 6
   1. Lambert Strether Post authorOctober 25, 2018 at 2:29 pm

      > There may be some surprises on Nov. 6

      Entirely possible!

      CO-03 and NY-02 are in Note [1] (the “expanded map”). See that note for context; it would take a long time to add those districts to the level of detail I’ve done for all the other districts; time I did not have. And while doing the expanded map would help with the horse-race, I don’t think it’s a requirement for understanding the Democrat Party as an institution, which is my primary goal.
4. cm October 25, 2018 at 9:19 am

   I see WA-03 mentioned in the footnote, but here on the ground, it looks like Long (D) has a good chance of beating Herrera-Beutler.
   1. Lambert Strether Post authorOctober 25, 2018 at 2:30 pm

      Can you say why?
      1. cm October 25, 2018 at 2:46 pm

         This is a good article.

         JHB is in her fifth term, and has been phoning it home for years. She has one of the worst attendance rates in the House, missing many votes. Part of that is due to her child’s health problems, which she is happy to use her Congressional health plan, while trying to gut the ACA, medicaid, and other health programs for the rabble.

         Worst of all, she is too scared to host town halls and actually face her constituents. Instead she brags about her impromptu conference calls, where all questions are pre-screened. This stance has driven even Republicans to want her out.

         In contrast, Long is very well spoken (a college politics prof) and has made pressing the flesh (even w/ hostile crowds) her hallmark.
5. Utah October 25, 2018 at 9:25 am

   Lambert, you’ve got two people winning UT-04 on your list, and then you attribute that win to McDonald, which is wrong. McAdams won that primary, the DCCC candidate. Though in my opinion McDonald would have made a better challenger.
   1. edmondo October 25, 2018 at 11:55 am

      Ditto PA-07. Edwards was the Bernie acolyte who lost (naturally) to Wild who could be the female version of Hillary Clinton.
      1. Lambert Strether Post authorOctober 25, 2018 at 2:30 pm

         Thank you both. I will fix.
         1. edmondo October 25, 2018 at 5:03 pm

            One last fix:
            Richard Ojeda is the Dem nominee in WV-03. He beat the DCCC choice (Davis) like a drum in the primary. I find Ojeda to be one of the more interesting Dems running. He voted for Trump in 2016 because Hillary. He supported Red for Ed teachers strike in WV and believes that government should provide “concrete material benefits” (where have I heard that phrase before?) to citizens. Should he win, I could see him being a natural ally with AOC even though he is from one of the reddest states in the Union. He would give Pelosi agita at the very least.
6. Eduardo October 25, 2018 at 9:43 am

   Sorry if I missed something, but in Table 2, “Clinton (OT)” means what?
   1. Lambert Strether Post authorOctober 25, 2018 at 2:32 pm

      Clinton’s endorsements are nominally those of Onward Together, her Astroturf organization.
7. Llewelyn Moss October 25, 2018 at 10:01 am

   I sure hope the Dems take over the House. After McConnel said out loud on teevee that he plans to Gut Social Security and Medicare to fix the deficit (created by the Trump taxcuts for the Rich), Repubs have become a frightening breed. And what else will they attack? The Trump presidency has turned from awful to Nightmarish. I’m not even a fan of the corporate Dems but Congressional gridlock is our only hope.
   1. Other JL October 25, 2018 at 12:04 pm

      If I’m completely honest with myself, I think it would be better for Rs to keep the house. The D/R charade just gives hope to leftists while preventing meaningful institutional reform. IMO things need to get worse before they can get better, and having a split Congress will delay that. I think it’ll take 3-4 terms of solid R rule before the left has a chance to make meaningful change.

      Here’s a thought experiment: suppose the Dems had solid control of both houses: what would they do? If you aren’t excited about that outcome, why vote for it?
      1. Prairie Bear October 25, 2018 at 1:00 pm

         I have had similar thoughts in wondering what would be best. Maybe a complete humiliation for the Ds in the House, like the GOP gaining 10 seats, but then a flip of the Senate, which doesn’t seem likely. It would have to be by several seats to counter Manchin, etc. I voted straight D. It’s all just speculation on my part; damned if I even know anymore what would be best.
         1. ape October 25, 2018 at 1:41 pm

            Historically, “the worse the better” hasn’t worked out, unless you’re hoping for revolutionary conditions.

            Otherwise, most people are pretty unprincipled at the end of the show — they’ll run to join the crowd.

            And the “revolutionary solution” is really, really bad historically. Really bad.

            What you really want is the Dems to kick-ass, even if they’re total sell-outs, to create space on the left. But if they lose? You get a whole lot of people becoming radical right wingers to be on the side of the winners.
            1. Other JL October 25, 2018 at 7:31 pm

               I don’t think we need revolutionary conditions. Rather I think the current D leadership is too cozy with a neoliberal agenda, and they’ve got more in common with R congress critters than the majority of the populace. Hence they’re mostly happy to lose, as the desired agenda is still being implemented.

               What would change with a complete loss of power is two things: first, more leftists would realize that the strategy isn’t working at all, and hopefully agitate enough to drive grassroots change.

               Second, the current strategy only works while there’s a semblance of a balance of power. If D’s become an obvious losing brand, money will dry up as lobbyists don’t need to hedge elections and personal donors get tired of throwing away donations. With no money, levers like the DCCC have much less power, creating a leadership vacuum.
   2. Big River Bandido October 25, 2018 at 12:07 pm

      If McConnell does attempt to got Social Security and Medicare, he’ll get plenty of help from Democrats. All of the Democrat leadership, and close to half the caucus in each house would support that…among Democrats in the Senate, there would probably be no more than a dozen who would oppose that if push came to shove.
      1. drumlin woodchuckles October 27, 2018 at 3:27 am

         Well . . . McConnell himself said in his statement that “reforming” “entitlements” will have to be a bipartisan effort. The Republicans can not do it alone. And he said that “divided government” would be the necessary condition for Democrats to bipartisanly support the ” entitlements reform” agenda.

         If McConnell is right about that, then a Democratic Takeover of the House would be disastrous for our Social Insurance programs. Because then “just enough” Catfood Democrats would work with the Catfood Republicans to vote Catfood Reform up out of the House and over to the Senate. Whereas if the Dems can be kept to a minority in the House and Senate, the Catfood Dems will pretend to support Social Security and Medicare to keep fooling their voters into re-voting for them.

         The ideal outcome would be for every Catfood Democrat to be defeated and for the SanderSocial Democrats to establish a strict and total monopoly of all Democratic-officeholder-held House and Senate seats.
8. Mark Gisleson October 25, 2018 at 10:30 am

   MN01 (my new home) is a bit of a surprise, but a yard sign crew told me that a key legislative race is being flipped due to high negatives for an R rep who is being opposed by the husband of a popular county official.

   R’s have been very nasty at the state level in MN, and it’s made them vulnerable. Ironically, the race Minnesotans are now most worried about is Keith Ellison’s. His accuser’s BS has fallen flat (not credible based on her own words which read like they were copied from a handbook), but DFLers were too quiet and the usual Hillbots are pushing venomous memes on social media. Meanwhile a trove of racist/sexist blog posts from his opponent have emerged, leading to desperate pleas on social media from people who’ve forgotten just how unforgiving Hillbots are.

   Heading down to Randy Bryce’s CD in a couple of days; I’ll be counting yard signs.
   1. Liz October 25, 2018 at 1:57 pm

      Moved to MN01 in May from CA-49. Any chance you might want to host a southern MN NC meetup?
      1. Lambert Strether Post authorOctober 25, 2018 at 2:34 pm

         You might want to bring this up in Links or Water Cooler, where more people might see it. Then you could organize together, perhaps.
   2. Left in Wisconsin October 25, 2018 at 7:53 pm

      Ironstache had a poll done the other day that shows him +1. That would be a huge win.
9. flora October 25, 2018 at 12:19 pm

   KS-02

   Paul Davis (D)
   vs
   Steve Watkins (R) (Jenkins is retiring, not running again.)

   with a libertarian candidate thrown in as a 3rd party.

   Trump was in town to rally with Watkins a short while ago. Lot of moderate Rs won’t vote for far-right* Watkins, even though this is an R district. Should be an interesting election. Third party candidates appear to have popped up in important KS races where far-right candidates might not get enough R votes, but where a 3rd party candidate could draw off moderate R votes that might otherwise to go the D candidate. Who is funding these 3rd party candidates remains a mystery.

   *on the same spectrum as Kris Kobach, imo.
10. Big River Bandido October 25, 2018 at 12:20 pm

    I think your approach of filtering out who the real candidates are from the left is correct. Dana Balter and Kara Eastman have been particularly disheartening as general-election candidates; Eastman, especially, talked a great game on health care back in the primary. Since getting the nomination, it seems that they caved to the establishment and diluted their platforms to tripe — Eastman did it within days of winning her primary. Same is true in solid Democrat districts that were never part of this series — I can’t even view the change in MA-07 as much of a win, since on policy at least, Presley appears to have defeated Capuano from the right, not the left. I’m not at all surprised that this process leaves only 2 genuine leftists remaining, plus AOC.

    OTOH, I’m not really discouraged either, for I didn’t expect otherwise. Lambert is right that the big potential for gains on the left is at the state and local levels.

    On that note…perhaps post-election, you could do a special wrapup of races at those lower levels where the left does make gains? It’s really hard to find that info, I know…but having that kind of picture would really be helpful. And probably would give the left some well-needed cheer post-election. I’d like to know what happens to the DSA members running for the PA and NY legislatures, for example…and whether there are similar opportunities on city councils around the country.
    1. Lambert Strether Post authorOctober 25, 2018 at 2:37 pm

       > You could do a special wrapup of races at those lower levels where the left does make gains?

       Our Revolution has a really impressive page on this.

       Among other things, running for office is a skill that can only be acquired by doing it. It’s not easy at all. So from the standpoint of building a left infrastructure (assuming that electoral work is something you think the left should be doing) getting out there and campaigning is a good thing, win or lose. Next time comes the win!
11. Prairie Bear October 25, 2018 at 1:05 pm

    I don’t know if it’s an error or I’m just not understanding what’s supposed to be in the tables, but Iowa 03 Axne: she is in tables 1 an 2 as “fM” but not in table 3a as a supporter of MfA/single payer. Maybe table 3a is only supposed to contain “likely winners” from 2a? Not entirely clear to me from the text preceding 3a; sorry if I am not reading correctly.

    Very interesting work, though. And a lot of work I bet!
    1. Lambert Strether Post authorOctober 25, 2018 at 2:39 pm

       fM = faux Medicare for All. Verbiage, not the real thing.

       Figure 3 is all winners. See the status column.

       Yes, it’s a lot of work!
12. Tom Doak October 25, 2018 at 1:09 pm

    The premise that the Left would make its impact in districts now under Republican control seems wrong to me. Those are the ones where the DCCC is spending all its money to try and push a DINO over the line and “win” a seat, even though that seat may not represent a reliable vote in close votes. (They’ll have to vote conservatively to keep their seat, and keep their powder dry for when they have a full majority.)

    The true Left candidates seem more likely to come from solid D districts where the longtime supporters are fed up with DINO incumbents, as with AOC. The real action was all in the primaries – and in the state elections andcredistricting battles.

    PS. I am just traveling back from Houston where Beto fever is in full swing. Lots of Obama girl types (but not so young) swooning over how handsome and well spoken he is; lots of yard signs; not a word about his policies. I doubt he’ll beat Cruz (they’re running in TX, not IL), but if he does he will be a shoo-in as the 2024 nominee for President.
    1. Lambert Strether Post authorOctober 25, 2018 at 2:40 pm

       Do Beto’s yard signs have sweat on them?
    2. JohnnyGL October 26, 2018 at 4:18 pm

       “The true Left candidates seem more likely to come from solid D districts where the longtime supporters are fed up with DINO incumbents, as with AOC.”

       I’m not sure this is correct. AOC’s opportunity came at least partly from the power of strong organizing work in a place with nearly-Jim-Crow-era-level restrictions on the electoral apparatus.

       Turnout isn’t nearly that reliably low in most true-blue districts. A lot of these places are in coastal citadels where dem party machinery is a lot stronger and is able to squash insurgents more effectively, at least partly because there’s lots of comfortable, relatively wealthy residents who like Team D just fine the way it is.

       I think there’s a number of scenarios that present opportunities for the left, not least of which is competing in communities that have been abandoned by the democratic party almost entirely. JD democrats have won some primaries unopposed. Then, the only challenge is to win the general election.

       As Ben Jealous is finding in true-blue MD, winning the primary is only half the battle. A lot of democrats are defecting over to the Republican side in an attempt to undermine Jealous in the general election. It looks like Hogan’s going to win re-election as Governor.
13. a different chris October 25, 2018 at 2:31 pm

    Oh, everybody seemed to have ignored Lambert’s question right in the header “if…what’s next for Democrats”.

    But Nancy has already answered that: Paygo! And probably a lot of “reaching across the aisle” and all that good stuff. If I wake up on Nov 7th and see the wave I expect, I won’t feel good because I just want everything in DC brought to a halt, basically. And that’s not what I will get.
    1. Lambert Strether Post authorOctober 25, 2018 at 2:41 pm

       If the left doesn’t control the party machinery (see Figure 3), “more of the same.” I should have stated that explicitly. My bad.
14. Jeremy Grimm October 25, 2018 at 3:21 pm

    Looking at what’s on offer for my area I’m afraid there will be a lot of undercounts on my ballot.

    The local sheriff’s race seems to have a lot of play but when I did some web searches and talked to my neighbor I didn’t gain any insight into what was going on or who was the best candidate. To find out about candidates and their politics the League of Women Voters suggests examining who endorses candidates, the media coverage, campaign literature, candidate websites, friends and family. The only reliable source I know of is Lambert’s guide.
    1. DonCoyote October 26, 2018 at 12:20 pm

       Endorsements as a source of information can be tricky. While they can be telling (i.e. we “learned” from HRC getting endorsed by all the newspapers in 2016 that they and she are neoliberals, and Trump is not), they may be given without being solicited, and, at least in the primary, organizations like Justice Democrats would only endorse one candidate (the first one that asked that met their criteria). And many are just due to the letter after your name, or idpol, not policy.
15. chicagogal October 25, 2018 at 4:53 pm

    Here in the IL-06, what’s surprised me the most is the number of Casten signs I’m seeing vs Roskam signs – especially in uber-conservative Wheaton. Just drove thru there and it’s amazing to see the sign fight amongst neighbors on well-traveled main streets – and not so well-traveled side streets – not to mention the normal sign gathering locations on empty lots. Also interesting is the size of the signs toward the larger end. It’s truly neighbor vs neighbor here, but I honestly don’t hold out any hope that the Dems will beat Roskam since they couldn’t (refused to?) when that seat was completely open in 2006.
16. DonCoyote October 26, 2018 at 1:15 pm

    Table 2(a), the horserace, you seem to have missed a few {select * from elections_analyis where IE .NE. [IE (Old)] or CR .NE. [CR (Old)] }, although maybe you didn’t want to get the red highlighter out :

    IL-06: IE moved two places, from Tilt-R to Tilt-D
    NE-02: [Red highlighter] IE moved from Tilt-R to Lean-R
    OH-12: [Red highlighter] CR moved from Toss-Up to Lean-R
    TX-23: [Red Highlighter] IE moved two places, from Toss-Up to Lean-R (bad news for Ortiz-Jones)

    Trivia: the VA-02 race features not just two m’s against each other, but Navy vs Navy! The Dem, Luria, actually graduated from the Naval Academy.

    Ballotpedia has an interesting report on wave elections. Looking at the data from the last 100 years, Dems would need to be +48 for this to constitute a wave (although if you restrict it to last 70 years, this drops to +30).

Comments are closed.