Princeton Election Consortium

A first draft of electoral history. Since 2004

Find a competitive House race near you!

September 9th, 2016, 8:51am by Sam Wang


This year, Republicans are going to lose seats in the House of Representatives – this is certain. How many seats, we don’t know. As analyzed by Dave Wasserman of the Cook Political Report, their majority is protected by aggressive redistricting efforts in 2010 (see my Stanford Law Review article and David Daley’s book Ratf***ed) and poor candidate recruitment by Democrats this year. Nonetheless, there remains some chance that a strong enough national popular vote win can flip the House.

Even if the House does not change control, a closer seat margin increases the ability of the minority to get legislation passed by peeling off votes from the majority. Under a Hillary Clinton Presidency, this will affect legislative priorities that cut across party lines, such as an increase in the minimum wage.

If you want to make a difference for your side, you can volunteer for a campaign in a contested district. In response to my wish, PEC reader Sharon Machlis has developed the Competitive Congressional District Finder, a cool application in which you type in your address or ZIP code and get back a map showing competitive races near you, as identified by the Cook Political Report. Whether you support Democrats or Republicans, these are the races that matter. Give the app a try – and get out there!

Note the general location of competitive districts: in the Northeast, the Midwest, the Southwest, and Florida. Many of these districts are competitive thanks to good-government and/or nonpartisan redistricting practices – see especially Arizona, California, Florida, and New York. This demonstrates the power of redistricting reform, whether pursued through the courts or through voter initiative.

Note on Florida: for a week or two, the app will not show current boundaries, thanks to a court-ordered redistricting and delays in getting the map files. The currently competitive Florida districts are FL-7FL-13FL-18, and FL-26.

Sharon used Google Fusion Tables and the Searchable Map Template, which was created by civic tech builder and open-government advocate Derek Eder. Many thanks to Sharon for her fast work. Talented readers like her are a major source of gratification.

Tags: 2016 Election · House

48 Comments so far ↓

  • Gopa

    Sharon Machlis, your map works like a charm. From the time Sam gave his suggestion to you getting it working is rapid turnaround time. Is there any way to motivate a Democratic House re-taking?

    • Sam Wang

      Isn’t it great that a famous scholar can opine on a quantitative issue without quantitative evidence and get published in a news outlet? I believe that more or less defines the concept of an academic pundit.

  • Shawn Huckaby

    Time to reset the ActBlue contribution goal. Let’s dream a little bigger this time! ;)

  • Rachel Findley

    Thanks, Sam Wang and Sharon Machlis, for this tool. I’ve been overwhelmed with rather random contribution requests from House campaigns, and this will help me choose the ones where my smallish contributions could make a difference.

  • Paul Ruston

    Sam-
    Novice questions about CSV file.
    For Ohio the file currently has:
    CBS C+7 mid-date 9/8
    Q T+1 mid-date 9/2 (Guessing this a two way and the next Q poll is a four way)
    Q T+4 mid-date 9/2
    PPP C+4 Middate 8/26(this poll is listed after the Wa post poll)
    Wa Post T+3 mid-date 8/20
    Wa Post T+3 mid-date 8/20

    1. Do you use both Q and both Wa Post Polls in your calculation?
    2. Is there a reason why the PPP poll is below the Wa poll which has a later mid-date?

    • Sam Wang

      The CSV file is further processed to obtain poll medians from all pollsters, one poll per pollster for each state. It is provided to you simply so you can see the entire gusher of information from HuffPost.

  • Jason

    I’ve been reading through the code, and I’m confused by the following line in EV_estimator.m: ” startrange=round((269-medianEV(1))/1.25)/10-2;”. It looks like the range you examine mostly closely is startrange to startrange+6, but startrange is derived from the median EV? how do you know that a calculation based off of median predicted EV will give you a good range of values to test for the meta-margin? where does that equation come from?

    • Sam Wang

      It’s an approximation that works. There is an approximately linear relationship between Meta-Margin and median number of electoral votes, so I just found a place that is always below the shift needed to generate a tie. Probably I could do that prettier.

  • Jeremiah

    Looking at the Huffington Post feed “2016 National House Race” chart, I prefer to look at their trend line than the median poll trend line you give here. However, I think the weakness in the HuffPost trend line is that they are not plotting the margin of the polls but the absolute numbers for D and R. This introduces noise into the trend because the response rates of the polls varies so greatly. It would be nice to have a trend line based on the margin.

  • Guy

    Thank you for doing this Sam. Really.

  • James Wimberley

    A request for a banner snapshot prediction for control of the House, based on the generic margin model. Obviously it would be much less informative than the Presidential and Senate predictions, but would fill a need. Nobody else is doing this.

  • Telso

    Virginia districts were also redrawn after a court order and the new ones will be in effect, so make sure to update those shapefile as well. Cook rates the new VA-04, an open seat, as likely Dem (which would be a Dem gain) and has for months (I believe since January), though it’s not colored on the map. It might be worth double checking the districts. Still, great resource!

  • James Wimberley

    Clinton’s campaign needs to keep Democrats energised by fear of a Trump victory and hope of Democratic control of Congress. It also needs to lull Republicans into an apathetic or disgusted acceptance of impending defeat so that they stay home. Ticket-splitting is unlikely and does not help the Congressional ambition.

    Call it the Werenfrid dilemma.
    Gene Wolfe, The Shadow of the Torturer:
    Master Malrubius … told us of a certain Master Werenfrid of our guild who in olden times, being in grave need, accepted remuneration from the enemies of the condemned and from his friends as well; and by stationing one party on the right of the block and the other on the left, by his great skill made it appear to each that the result was entirely satisfactory.

  • AySz88

    Is there any comparison yet on the “power” of working to get votes for a House race vs. doing that for the Senate (with the goal of flipping control of the chamber)? The Senate seems slightly closer to the knife-edge overall.

    I wonder if it’s worth marking the competitive Senate races. (Or even both simultaneously? Perhaps different colors for close House, close Senate, and overlapping?)

  • Kevin King

    This is wonderful! Thank you, and good job.

    Unfortunately, I live near Atlanta, Georgia; no competitive district lies within 250 miles of me.

    • Sam Wang

      Yes – if you live in the South, you’re out of the House game this year. Your options are to help in North Carolina (super-exciting election there!) or to donate money to the DCCC via ActBlue or the NRCC.

    • Keith R

      Aw c’mon, Sam. Florida (four districts) is in the South as well as Texas (one district). And there is the competitive race in VA-10 (in the northern part of the state).

    • Olav Grinde

      @Keith: Perhaps it’s more accurate to view Florida as an almost-Caribbean state, rather than a Southern state? ;)

    • Keith R

      In South Florida definitely true these days, but go further north and you’ll find all kinds of southern people.

  • Michael Coppola

    The Hastert rule (“a majority of the majority) doesn’t preclude passing bipartisan legislation. The Boehner rule (“a majority of the house from the majority party”) effectively does. A majority of the current majority is only 124 votes. A majority from the majority allows 30 tea party zealots to block everything.

  • Leading Edge Boomer

    Don’t forget Iowa. They have used a nonpartisan group to do the redistricting since their law was enacted in 1980. Passed nearly unanimously by a Republican legislature.

  • Joan Duce

    I would like to personally donate $1 to each of the even remotely competitive House races this year (in favor of the Dems), and it is surprisingly hard for me to find a definitive list of what those are at the moment. Where would you draw that line?

  • anonymous

    A few competitive districts in New York, guess that should end the complaining of New Yorkers that their vote does not matter.

  • Peter

    This is missing CA-49, which in Lean-R in Cook. Is this out of date? What’s the refresh?

    Would be nice to have the +- values from Cook in the UI.

  • Doug Taylor

    District boundaries in Florida look wrong to me.

  • Carl Nyberg

    “Poor candidate recruitment” is a charitable way of saying the DCCC and national Dems didn’t recruit US House candidates, except in a few districts like IL-10 (a D+8 district held by a Republican).

    The national Dems are not trying to win the US House. Why not?

    • Matt Harrison

      I pulled the shp files for the 114th from here :

      http://cdmaps.polisci.ucla.edu/

      to calculate centroids. I’ll look into what happened to Fla. Just getting to this now. If there is a problem with their shp files, I can let them know

  • Steve

    What happened to Ohio in the presidential power poll on the right-hand side of the homepage?

    • Sam Wang

      It currently favors Trump. Polls would have to swing about 4 percentage points toward Trump to give him a shot at 270 EV, which would make Ohio a safe state for him.

  • Arthur Klassen

    That’s a pretty incredible map. Just wondering about this statement: “Even if the House does not change control, a closer seat margin increases the ability of the minority to get legislation passed by peeling off votes from the majority.”

    Doesn’t that assume that the next speaker of the House, even if a Republican, takes a pass on the “Hastert Rule” about the majority of the majority? It seems to me that so long as that principle holds, only an outright majority is any guarantee that “some other agenda” will have any legislative influence. Unless, perhaps, the minority piles onto a moderate member of the majority party in the election of a Speaker or something else radical like that.

    Thoughts?

    • David vun Kannon

      I agree that the statement is doubtful. A smaller GOP majority strengthens the extreme wing, since they are going to be re-elected from safely gerrymandered (or just safely red) districts, and it is the moderates that will be losing seats.

    • LondonYoung

      I agree with Sam’s take, here’s why:
      Look at something like increasing the debt ceiling – “must-pass” legislation. Failing something like this is typically understood as a “no confidence” vote that brings down the government in a parliamentary system.

      The GOP can’t get its right wing to vote for things like this, so, for the govt to function, dem votes are required.
      Which means that deals must be struck.
      A small GOP majority means more must-pass bills require dem support, and more deals for the dems.