Flipping Congressional Districts, Part 1

 

[Previously: Data request, part 0]
[Next: Flipping Congressional Districts, part 2]

To many people, including myself, the most infuriating thing about the GOP is that they don’t try to win elections. I mean, sometimes it even seems that they are actively trying not to win elections. You can have the best candidates, with the best platforms, but it comes to nothing if they aren’t elected.

The Dems understand this, so they take the exact opposite approach and do anything they can to win elections. (Normally one would include examples here, but I don’t think it’s necessary for this audience.) Winning elections requires expensive resources, of course, but it also requires strategy, which is free.

As an engineer, I’m intrigued by the problem of how to efficiently flip congressional districts. I needed a subject for a little programming project, so I grabbed a source of 2018 election data, wrote a program that parsed it, patched up a few issues the data, and ran some metrics on it. (Some boring details here.)

This will take the form of a series of posts as an exploratory conversation with that data.

So … currently, there are 235 Democrat congressional districts and 200 Republican congressional districts.

Wikipedia has an excellent summary of the 2018 election data here.

The first question I have is, how many congressional campaigns were uncontested?

I first thought you could search through the Wikipedia page for instances of the word “unopposed.” But that’s not actually good enough. There are a significant number of cases where a Democrat ran against a Green Party candidate with no Republican in sight. I guess that’s technically not unopposed. But I’m more interested in cases where the candidate ran unopposed by a major party.

So with the data, I am now able to answer questions like that.


There are 38 congressional districts where the Democrat ran unopposed by a Republican in 2018. They are:

AL-7
AZ-7
CA-5 CA-6 CA-13 CA-20 CA-27 CA-34 CA-40 CA-44
FL-10 FL-14 FL-20 FL-21 FL-24
GA-5
LA-2
MA-1 MA-4 MA-7 MA-8
MI-13
MS-2
NY-5 NY-6 NY-7 NY-8 NY-16 NY-17
PA-18
TX-9 TX-20 TX-28 TX-30
VA-3
WA-2 WA-9
WI-2

On the other side, there are 4 congressional districts where the Republican ran unopposed by a Democrat:

CA-8
GA-8
NC-3
ND

I’m also interested in the case where the candidate ran “effectively unopposed,” meaning that one of the major candidates was so nonfunctional that they ended up receiving less than, say, 20% of the vote.

Here are the 22 districts where the Democrat ran “effectively unopposed”:

CA-12 CA-29 CA-37
IL-1 IL-2 IL-4 IL-7
MI-14
MO-1
NJ-8 NJ-10
NY-9 NY-10 NY-12 NY-13 NY-14 NY-15
OH-11
OR-3
PA-3
TN-9
WA-7

And the flip side, here are the two districts where the Republican ran “effectively unopposed”:

TX-11 TX-13


Now, the standard Republican response is that those districts are all heavily Democratic and any resources spent there are wasted. Yeah, yeah, I got that. But that’s also an easy excuse to not do anything.

You know the phrase, “The future belongs to those who show up?” Well, the data shows that the Republican Party is not showing up. And they are not showing up an order of magnitude (that is, a factor of 10) more than the Dems. 38 vs. 4 for unopposed, 22 vs. 2 for effectively unopposed. Together, 60 vs. 6.

This is a serious problem.

Barack Obama basically became president by running unopposed. In 1996, he won his state senate seat when his people had his main competition disqualified. In 1998, he ran for reelection unopposed. And ran unopposed again in 2002. Obama famously won his US Senate seat in 2004 when Republican Jack Ryan withdrew after some divorce papers were mysteriously unsealed. And, as he ran for president, John McCain suspended his campaign for a while. So running unopposed is a proven political strategy.

More to come. Next, I’ll explore applying a cost metric to flipping a congressional district.

Now that I can run arbitrary calculations on the data, I want to encourage suggestions.

Published in General
This post was promoted to the Main Feed by a Ricochet Editor at the recommendation of Ricochet members. Like this post? Want to comment? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

There are 52 comments.

Become a member to join the conversation. Or sign in if you're already a member.
  1. Gary McVey Contributor
    Gary McVey
    @GaryMcVey

    There’s a lot of truth in this post and I agree with its central point: you don’t have a chance when you don’t compete. I don’t agree that we’re demure and helpless while the Dems are shrewd and ruthless. We’re as ruthless as they are, we’re just not as shrewd. The Dems are more than happy to run someone who doesn’t sign on to the full party list when they know the district demands it. If they ran a NYC AOC type in upstate NY, they’d lose. For that matter, if the GOP ran a Steve King type in my district, he wouldn’t get a single vote, even mine. By and large they don’t.

    The President will campaign for GOP candidates, but everyone senses he’s a different kind of Republican from the kind they’re used to, and it means the label “Republican” isn’t as consistent as it used to be. A vague label is not much of an extra selling point. 

    We’d love to stick every Dem campaign with defending AOC and they’d do the same with Steve King, but that usually doesn’t work. Unlike the UK, this isn’t a unified national campaign in either funding or messaging. AFAIK it’s mostly locally driven fundraising with some national help to supplement it. So whoever steps forward to be pioneers in these uncontested districts knows, going in, that they’re going to have to spend a year of their life and raise a lot of money. They’ll do that for a long-shot chance, but not for virtually no chance. The Dems are catching up with the Richie Rich class of donors, plus they have a built-in institutional fundraising advantage in the unions.

    • #1
  2. Vectorman Inactive
    Vectorman
    @Vectorman

    It isn’t the Steve King types that lose elections for Republicans, its the Republicans In Name Only (RINO) that like to lose because they are losers.

    During the eight years of Obama, they couldn’t accomplish much of the real Republican agenda, which is cut taxes, cut spending, cut regulations, cut illegal immigration, etc. So what do they do in 2016 when all three areas (House, Senate, President) are nominally Republican? About 50 retire from the House in 2018 due to Trump winning!

    Some might have been afraid that their corruption and/or Epstein-like escapades would be exposed. But most quit because they were Democrats at heart and ran as a Republican for the power. What the Republican party needs are more Newt Gingrich types – smart, hard working, with a reasonable agenda for running a proper Federal government.

    • #2
  3. WillowSpring Member
    WillowSpring
    @WillowSpring

    Interesting project.  It sounds like the technique I have used to learn new computer languages – that is, instead of doing simple ‘toy’ programs, just force my way into using the language for a ‘real’ problem that I am interested in.

    I have a couple of what I guess are administrative questions:

    * what source (s) of data did you find?

    * what tools are you using for the analysis?

    I look forward to other posts

    • #3
  4. JoelB Member
    JoelB
    @JoelB

    This Republican no-show problem is a sad state of affairs. A portion of the piece might alternately be titled “How the Republicans made Barack Obama President” if you were going for click-bait.

    The Dems are always accusing the GOP of gerrymandering. (I know, they both do it.) Could it be that the GOP is handing the Dems safe districts and hoping to win the marginal ones (and sometimes failing)? Or could it be that the Dems have built a solid base of districts that they don’t need to worry about financially and kept the GOP on the edge?

    • #4
  5. I Walton Member
    I Walton
    @IWalton

    Good article.  Hope some Republicans read it. 

    • #5
  6. Stad Coolidge
    Stad
    @Stad

    namlliT noD:

    You know the phrase, “The future belongs to those who show up”? Well, the data shows that the Republican party is not showing up. And they are not showing up an order of magnitude (that is, a factor of 10) more than the dems. 38 vs 4 for unopposed, 22 vs. 2 for effectively unopposed.

    This is a serious problem.

    It most certainly is.  You can’t win if you don’t run.  Furthermore, it’s hard to win when an incumbent drops out, leaving an election wide open (think there were over 30 House Republicans in 2018 who did this).

    • #6
  7. namlliT noD Member
    namlliT noD
    @DonTillman

    WillowSpring (View Comment):

    I have a couple of what I guess are administrative questions:

    * what source (s) of data did you find?

    * what tools are you using for the analysis?

    I look forward to other posts

    I’m using the MIT Election Lab data at https://electionlab.mit.edu/data .

    And I describe some of the errors in that data in my previous post, here.

    I wrote a small Python program to parse, correct, and work with the data.  I’m happy to share it, but it’s still under construction.

    At the same time, it’s entirely possible to do this analysis in a spreadsheet.  Perhaps even more effectively.  I’m still exploring, we’ll see.

    • #7
  8. rgbact Inactive
    rgbact
    @romanblichar

    Democrat districts are more partisan, on average, than GOP districts. This is due to more clustering of Democrat voters. Hence, it creates districts that the GOP has no chance in……while creating many swing districts that the GOP has a chance in…..and where they should spend resources. So, I totally disagree with your argument.

    • #8
  9. Al French, sad sack Moderator
    Al French, sad sack
    @AlFrench

    You got OR-3 right. And it has been that way for years.

    • #9
  10. namlliT noD Member
    namlliT noD
    @DonTillman

    Anybody know of a source of the population for each congressional district?  I’m not finding it.

    • #10
  11. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    Thank you Don for the time you have spent in creating this post.  It appears that everything you have said is accurate.  However, I assert that you are drawing the wrong conclusions.

    To draw the accurate conclusions, you will need a resource which lists the partisan bent of each district.  The most common one is the “Cook Partisan Voting Index.”  You can find this in The Almanac of American Politics” which comes put every two years.  Michael Barone was its primary editor for years.  Their 2020 edition just came out earlier this month in its paperback form, which costs $89.  (I collect the hardback form every 2 years going back to 1994.  The hardback book costs $164 and will come out in September.  This is the most expensive book I buy every two years.)  Since I haven’t gotten my 2020 hardbound edition yet, I will use the 2018 edition for my analysis.

    The Cook Partisan Voting Index shows that in the 37 of the 38 districts where the Democrats ran unopposed, they had a Cook Parisian Voting Index of D+ 20, 21, 21,  21, 40, 23, 16, 35, 33, 35, 11, 7, 31, 9, 34, 34, 25, 12, 9, 34, 10, 32, 14, 37, 16, 38, 24, 7, 29, 10, 9, 29, 16, 10, 21, and 18.  (PA-18 was redrawn, so it was not used.)

    The Cook Partisan Voting Index for the 4 districts where the Republicans ran unopposed, show they had a Cook Partisan Voting Index of R+ 10, 15, 12 and 16.

    The Cook Partisan Voting Index shows that in the 21 of the 22 districts where the Democrat ran “effectively unopposed” they had a Cook Partisan Index of D+ 37, 29, 37, 27, 29, 33, 38, 30, 29, 27, 36, 34, 26, 31, 43, 29, 44, 32, 24, 28, and 33.  (PA-3 was redrawn, so it was not used.)

    The Cook Partisan Voting Index shows that in the 2 districts where the Republicans ran “effectively unopposed” they had a Cook Partisan Index of R+ 32, and 33.

    Put bluntly, it would be foolish to run in the general elections in these overwhelming districts.  (Of note the four female “Squad” representatives AOC, Talib, etc. are from these overwhelming districts.  They can’t be taken out in general elections; they must be primaried.  The result will be Democrats, but hopefully not racist, Anti-Semitic Democrats.).

    The unopposed or overwhelming Democratic districts were often drawn to cluster as many Democrats as possible in them so that Republicans can win comfortable margins in as many districts as possible.

    For example in Alabama, AL-7 is an oddly shaped district that pokes into the black portions of Birmingham and Montgomery, yet avoids their white suburbs.  It is 64% African-American.  It has a Cook Partisan Index of D+20.  Alabama’s other six districts have a Cook Partisan Voting Index of R+ 15, 16, 16, 30, 18 and 26 and they all have Republican Representatives.

    In Louisiana, LA-2 is focused in New Orleans.  It has a Cook Partisan Voting Index of D+25.  Louisiana’s other five districts have a Cook Partisan Voting Index of R+24, 20, 13, 15 and 19, and they all have Republican Representatives.

    Democrats play the same games in drawing lines.  In Maryland, there are 7 Democrats and 1 Republican in their delegation.  (As late as 1994, there were four Democrats and four Republicans in the Maryland delegation.)

    I urge Don to run a new analysis based on voting trends based upon the partisan bent of districts such as the Cook Partisan Voting Index.  That would be much more helpful and relevant.  However, I want to praise him for his hard work in developing this post.  I am just suggesting a next step, that’s all.

    • #11
  12. WillowSpring Member
    WillowSpring
    @WillowSpring

    namlliT noD (View Comment):
    At the same time, it’s entirely possible to do this analysis in a spreadsheet. Perhaps even more effectively. I’m still exploring, we’ll see.

    After years of using spreadsheets (starting with VisiCalc!), I switched to ‘R’, a language designed for doing statistics and math.  Another advantage is that is being very actively developed and there are a lot of libraries to obtain data from web sites and clean it up for further processing.

     

    • #12
  13. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    namlliT noD (View Comment):

    Anybody know of a source of the population for each congressional district? I’m not finding it.

    The Almanac of American Politics has this information for every congressional district.  I think that you would really enjoy its analysis and research.

    George Will has called the Almanac of American Politics “The Bible of American Politics” and “The Baseball Digest of American Politics.”  

    • #13
  14. namlliT noD Member
    namlliT noD
    @DonTillman

    Gary Robbins (View Comment):

    namlliT noD (View Comment):

    Anybody know of a source of the population for each congressional district? I’m not finding it.

    The Almanac of American Politics has this information for every congressional district. I think that you would really enjoy its analysis and research.

    George Will has called the Almanac of American Politics “The Bible of American Politics” and “The Baseball Digest of American Politics.”

    I need it in machine-readable form.  

    Right now I could gather all 435 numbers by hand.  But I’m not going to do that.

    • #14
  15. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    namlliT noD (View Comment):

    Gary Robbins (View Comment):

    namlliT noD (View Comment):

    Anybody know of a source of the population for each congressional district? I’m not finding it.

    The Almanac of American Politics has this information for every congressional district. I think that you would really enjoy its analysis and research.

    George Will has called the Almanac of American Politics “The Bible of American Politics” and “The Baseball Digest of American Politics.”

    I need it in machine-readable form.

    Right now I could gather all 435 numbers by hand. But I’m not going to do that.

    I think that the info is out there, I just don’t know where.

    The RNC is well ahead of the DNC in doing this analysis, though the Dems are catching up.  There are computer programs available as to how to gerrymander districts.

    • #15
  16. rgbact Inactive
    rgbact
    @romanblichar

    Gary Robbins (View Comment):

    The Cook Partisan Voting Index shows that in the 21 of the 22 districts where the Democrat ran “effectively unopposed” they had a Cook Partisan Index of D+ 37, 29, 37, 27, 29, 33, 38, 30, 29, 27, 36, 34, 26, 31, 43, 29, 44, 32, 24, 28, and 33. (PA-3 was redrawn, so it was not used.)

     

    Yep. I see no value in running in a D+25 race.  There aren’t many GOP districts that are this partisan. But,  these “lost causes” are thrilling for partisans. Thats why Democrat voters plow millions into defeating Steve King every year.  But, thats not a good election strategy overall.

    Meanwhile, in 2 weeks there is a special election in an actual GOP leaning district…and the Democrat is leading polls.

     

    • #16
  17. namlliT noD Member
    namlliT noD
    @DonTillman

    Hey Gary, 

    I’m familiar with the Cook Partisan Voting Index.  The Wikipedia entry on it is pretty good.

    And I’m familiar with your points.

    The fundamental problem with the PVI, and maybe I should have started with this, is that it doesn’t offer any solutions.  More than that, it suggests that there are no solutions.  As do your points.

    I’m interested in practical solutions, and I believe I’ve found some.  Hence the title.  

    So I am doing something very different here.  I’m actually taking the opposite approach.  There is a lot more to come; hold tight.

    • #17
  18. Gary McVey Contributor
    Gary McVey
    @GaryMcVey

    Gary Robbins is right; some of these districts are not worth spending money on until or unless there’s a cultural revolution that hasn’t happened yet.

    I have to disagree with my friend Vectorman. I don’t think anybody endures a modern congressional campaign because they don’t want to win. In many cases the GOP is blind to what the district is really like, and run candidates who are in tune with national activists but hopelessly wrong for the actual election. If your district is 25% Black and/or Latino, hoping that they’ll start listening to Hannity and reading Jack Posobiec is like hoping that clapping loud enough will bring Tinkerbell back to life. Ain’t happening. 

    After a certain point, people with leadership skills and strong links in the business community start deciding it’s easier to change policy as a moderate Democrat than as a Republican. That becomes easier for them socially when most of their wives have already left the GOP over social issues. I don’t know the answers, but I do know that our approach isn’t working. 

    • #18
  19. namlliT noD Member
    namlliT noD
    @DonTillman

    Gary Robbins (View Comment):

    The RNC is well ahead of the DNC in doing this analysis, though the Dems are catching up. There are computer programs available as to how to gerrymander districts.

    It’s not even close.

    Here’s the famous diagram of the Obama 2012 campaign cloud compute infrastructure:

    https://s3.amazonaws.com/OM-SHARE/AWSOFA-Print-27×240.pdf

    Warning, the diagram alone is 60 MB.

     

    • #19
  20. rgbact Inactive
    rgbact
    @romanblichar

    North Carolina 9th is an R+8 district…….that the Democrats nearly won in 2018, and evidently they are re-running the race on Sept 10th. And the Democrat is up 4pts in the latest poll. Not sure why the GOP is doing so poorly in a southern GOP lean district. Any reports from North Carolina would be appreciated.

    • #20
  21. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    rgbact (View Comment):

    Gary Robbins (View Comment):

    The Cook Partisan Voting Index shows that in the 21 of the 22 districts where the Democrat ran “effectively unopposed” they had a Cook Partisan Index of D+ 37, 29, 37, 27, 29, 33, 38, 30, 29, 27, 36, 34, 26, 31, 43, 29, 44, 32, 24, 28, and 33. (PA-3 was redrawn, so it was not used.)

    Yep. I see no value in running in a D+25 race. There aren’t many GOP districts that are this partisan. But, these “lost causes” are thrilling for partisans. Thats why Democrat voters plow millions into defeating Steve King every year. But, thats not a good election strategy overall.

    Steve King’s IA-4 is R+11.  The other Iowa districts are D+1, D+ 1 and R+1.  Trying to knock off King is not a good use of Dem’s money.  Instead, if memory serves they swung the other 3 districts.   If I had it all my way, we could have all four districts as R+2 or R+3, and could potentially sweep the state.

    Steve King will need to be knocked out in the primary, something that I support.

    Meanwhile, in 2 weeks there is a special election in an actual GOP leaning district…and the Democrat is leading polls.

    NC-9 is a R+9 district.  It should be an easy win.

    [Comment edited at request of post author.]

    • #21
  22. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    rgbact (View Comment):

    North Carolina 9th is an R+8 district…….that the Democrats nearly won in 2018, and evidently they are re-running the race on Sept 10th. And the Democrat is up 4pts in the latest poll. Not sure why the GOP is doing so poorly in a southern GOP lean district. Any reports from North Carolina would be appreciated.

    Comment deleted at request of post author.

    • #22
  23. rgbact Inactive
    rgbact
    @romanblichar

    Gary Robbins (View Comment):
     

    NC-9 is a R+9 district. It should be an easy win. The only reason why the Dem is leading in that district 46-42% is the passionate desire of women, the college educated, the young and the eastern suburbs of Charlotte to send us a message that Trump is toxic, a message that we did not learn in 2018. If the Dem wins a race that he should lose, will we get the message? Frankly, I don’t think so.

    Could be. NC-9 is the 2nd least rural and 2nd highest income GOP district in the state. So, looks like the college educated voter still hates the Trump GOP, even with a good economy.

    Not sure why these suburban candidates don’t make a massive break from Trump. Meanwhile, this NC guy had Trump Jr come in for a fundraiser this week.

     

    • #23
  24. namlliT noD Member
    namlliT noD
    @DonTillman

    Gary Robbins (View Comment):
    The only reason why the Dem is leading in that district 46-42% is the passionate desire of women

    Dude…

    Gary Robbins (View Comment):
    to send us a message that Trump is toxic, a message that we did not learn in 2018.

    Gary, this is a serious discussion on applications of data analysis.  Don’t be crapping all over it with your flaming TDS.

    • #24
  25. Gary McVey Contributor
    Gary McVey
    @GaryMcVey

    The problems existed before Trump, and Trump is not going to be replaced on the ticket.

    If anything, Trump should be the cure to some of them (but he hasn’t been; why? Quien sabe?) He’s not a neo-con enthusiast for overseas wars; he keeps evangelicals friendly while not lifting a finger to overturn Obergefell; he refuses to cut Social Security, breaking the heart of the Wall Street Journal; and unlike endless generations of past Republicans, he doesn’t suggest that middle class workers should eagerly seize the opportunity to drop their wages to the level of people who arrived here a month ago. Those are, or should be, winning positions. If future GOP presidential candidates reverse them, they’ll do worse, not better. 

    But there’s something about Trump that manages to not get credit for it, and no, it’s not all the media’s fault. 

    • #25
  26. Gary Robbins Member
    Gary Robbins
    @GaryRobbins

    namlliT noD (View Comment):

    Gary Robbins (View Comment):
    The only reason why the Dem is leading in that district 46-42% is the passionate desire of women

    Dude…

    Gary Robbins (View Comment):
    to send us a message that Trump is toxic, a message that we did not learn in 2018.

    Gary, this is a serious discussion on applications of data analysis. Don’t be crapping all over it with your flaming TDS.

    I am sorry.  Someone raised the issue of NC-9 and I was following up on that as did Comment 23.   However it can be argued that I added too much commentary so I will edit that out.  I look forward to more of a discussion of data analysis.

    A small point.  I think that I have more TES, or Trump Exhaustion Syndrome than TDS, Trump Derangement Syndrome.  However I am probably not the best judge of that.

    • #26
  27. ctlaw Coolidge
    ctlaw
    @ctlaw

    I used to wholeheartedly agree with the idea of running a candidate in all races.

    After all, a month before the general election, the Dem. might get caught in bed with the proverbial live boy or dead girl.

    But, now, the extreme nature of the modern Dem. voting base (including the vote fraud base), means that there is practically nothing a Dem. can do to lose a significant fraction of the base.

    Also, a corrupt judge will simply choose to allow the Dem. to be replaced at the last minute despite the letter and intent of the law.

    • #27
  28. Petty Boozswha Inactive
    Petty Boozswha
    @PettyBoozswha

    In his last race for Governor in 1990, George Wallace got 90% of the Black vote. In his race for the Senate in 2014, South Carolina’s Tim Scott got 7% of  the Black vote. Spending time and money on majority minority districts is not going to  work, they are Lee Atwater’s pernicious legacy until we do major structural reform. I would recommend we really look at expanding the House so that the smallest state has two Reps and the number of Congresscritters expands proportionately. The number of Reps in Washington could be capped at 500 and the rest would have to live in their districts or state capitals and vote by Skype.

    • #28
  29. TBA Coolidge
    TBA
    @RobtGilsdorf

    rgbact (View Comment):

    North Carolina 9th is an R+8 district…….that the Democrats nearly won in 2018, and evidently they are re-running the race on Sept 10th. And the Democrat is up 4pts in the latest poll. Not sure why the GOP is doing so poorly in a southern GOP lean district. Any reports from North Carolina would be appreciated.

    Tech jobs bringing in libs from out of state? 

    • #29
  30. Petty Boozswha Inactive
    Petty Boozswha
    @PettyBoozswha

    rgbact (View Comment):
    Not sure why these suburban candidates don’t make a massive break from Trump. Meanwhile, this NC guy had Trump Jr come in for a fundraiser this week.

    Because Trump and the Trumpkins have made if very clear that they will devote 500 times more energy in destroying any apostate than they will put in to win a regular contest. Thom Tillis is feeling that right now – if the Dems flip the Senate that will be a big reason why.

    • #30
Become a member to join the conversation. Or sign in if you're already a member.