The 2020 House Elections: How did I do?

On Wednesday, I took a look at back at my predictions for the 2020 U.S. Presidential election, and Saturday it was the Senate’s turn. Today: how did I do at predicting elections for the House of Representatives?

As happens every even-numbered year, all 435 seats in the House of Representatives were up for a vote. I predicted the outcomes of them all on a map, and discussed detailed predictions in part 2 and part 3). How did I do in predicting the results of those elections?

What We Still Don’t Know

The answer is, “we don’t know for sure yet,” because there are four results still unknown.

The weird one

  • Louisiana-5: As with the two Senate Elections in Georgia, we don’t know the winner in Louisiana’s 5th congressional district, which covers the whole northeastern quarter of the state and includes Monroe and Alexandria. Like Georgia, Louisiana uses a “jungle primary” system where all candidates appear on the ballot regardless of party. If one candidate gets 50 percent of the vote, they win the election; if not, the top two candidates face off in a runoff. The top two finishers were Republicans Luke Letlow (33.1% election day vote share) and Lance Harris (16.6%), leaving us in the weird position of knowing that a Republican will represent LA-5, but we still don’t know which one.

The exciting ones

And there are also three house races where both winner is still unknown, and might be from either major party:

  • California-21 (the southern third of the Central Valley and the east side of Bakersfield): with 95 percent of ballots counted, Republican former representative David Valadao (83,564 votes; 50.5%) leads Democratic incumbent T.J. Cox (81,946; 49.5%). CURRENT RESULTS
  • California-25 (rural-ish Los Angeles and Ventura Counties, including Palmdale, Santa Clarita, and the San Fernando Valley): with 87 percent of ballots counted, Republican incumbent Mike Garcia (169,060 votes; 50.1%) leads his former special election Democratic opponent Christy Smith (168,660; 49.9%). That’s a difference of just 400 votes, smaller than my high school graduating class. CURRENT RESULTS
  • Iowa-2 (the southeastern quarter of Iowa, including Davenport and Iowa City): the most mind-blowing one of all… with all votes counted and currently being recounted by hand, Republican Marianette Miller-Meeks (196,862 votes; 50.006%) leads Democratic candidate Rita Hart (196,815; 49.994%) BY JUST 47 VOTES. That’s literally less people than would fit on a single bus! CURRENT RESULTS

Iowa’s second district race is headed to a recount, but recounts almost never change election results, so the most likely outcome is a ridiculously narrow win by Miller-Meeks. What is especially remarkable about this race is that neither I nor anyone else had this on our radar as a close race; Hart was expected to win easily. But so it always goes on election night – there is always at least one race that defies expectations. That’s the nature of trying to predict 435 different things; enough opportunities for outcomes to occur virtually guarantees that there will be a rare outcome somewhere, somehow.

So there are those four uncertainties, but that still leaves 431 certainties. How did I do in my predictions?

Hits and Misses

My top line result for the House of Representatives was that the party count would be a good result for the Democratic Party: 237 seats going Democratic and 198 Republican, for a net swing of five seats Democratic.

In reality, the Democratic majority remains, but the House elections feel very much like a win for the Republicans. Not including the three seats yet to be decided, the count is 230 Democratic, 202 Republican. Republicans have a narrow lead in all three undecided races, meaning the result most likely result is a swing of seven seats in favor of the Republicans.

I have some ideas about why the 2020 House and Senate elections were so good to the Republicans, but that’s a topic for a different post. What matters here is: how right or wrong were my predictions? And for evaluating my predictions, the important question is not the swing, but rather the individual election results.

Here is my the map of the outcome of my predictions. Colors indicate the party of the winning candidate, using the traditional yet arbitrary red for Republican and blue for Democratic. Districts whose calls I missed are shown with their winner’s name in BIG BOLD ALL-CAPS.

Checking my predictions for the 2020 U.S. House of Representatives elections. Click for a larger version.

The map shows that my predictions were correct for 417 of the 432 decided House seats. That’s a 96.5 percent success rate, although it’s less impressive than it sounds when you realize how gerrymandering has made most House elections completely uncompetitive. It’s also less impressive when you realize the systematic offset: I mistakenly predicted one Republican win in a seat that went Democratic, and FOURTEEN Democratic wins in seats that went Republican. If Garcia holds on in California-25 and Miller-Meeks in Iowa-2, the latter number goes up to SIXTEEN.

Here are the districts where I got it wrong. Click the district names to see the district’s Wikipedia page, which includes a map; click on RESULTS see the results from the most official source I could find (and many thanks to the websites of the Secretaries of State of California and Utah for making their official state counts so easy to find and link to):

  • California-39 (eastern suburbs of LA, including Fullerton and Yorba Linda): Republican state assembly woman Young Kim (173,312 votes; 50.6 percent vote share) defeated Democratic incumbent Gil Cisneros (169,089; 49.4 percent). RESULTS
  • California-48 (coastal Orange County, including Huntington Beach and Laguna Beach): Republican county councilwoman Michelle Steel (201,660 votes; 51.1 percent vote share) defeated Democratic incumbent Harley Rouda (193,305; 48.9 percent). RESULTS
  • Colorado-3 (the western half of Colorado, including Aspen, Grand Junction, and Pueblo): Republican bar owner and gun rights acitvist Lauren Boebert (220,502 votes; 51.4 percent vote share) defeated Democratic state representative Diane Mitsch Bush (193,980; 45.2 percent). RESULTS
  • Florida-26 (southern Miami-Dade County and the Florida Keys, including Homestead and Key West): Carlos Gimenez, the Republican county commissioner of Miami-Dade County (177,223 votes; 51.7 percent vote share) defeated Democratic incumbent Debbie Murcasel-Powell (165,407; 48.3 percent). RESULTS
  • Florida27 (cities of Miami, Miami Beach, and Coral Gables): in another race that was absolutely not at all on anyone’s radar, Spanish-language TV news host Maria Elvira Salazar (176,141 votes; 51.4 percent vote share) shocked long-revered Democratic incumbent Donna Shalala (166,758 votes; 48.6 percent). The unexpected result, like Trump’s unexpected advantage in Miami-Dade County, is further evidence that that Trump’s messaging resonated in Miami. RESULTS
  • Georgia-7 (northeastern suburbs of Atlanta, including Duluth, Norcross, and Suwanee): The Democratic 2018 candidate, Professor Carolyn Bourdeaux (190,900 votes; 51.4 percent vote share) defeated emergency room doctor and Marine Corps veteran Rich McCormick (180,564; 48.6 percent). I predicted this race would be within-one-percent close, but I picked the wrong winner. It was the only race where I pick d a Republican and the seat went Democratic. RESULTS
  • Indiana-5 (city of Indianapolis and its northern suburbs, including Carmel and Marion): In another election I predicted would be within-one-percent close – but was not – Republican state senator Victoria Spartz (208,212 votes; 50.0 percent vote share) defeated Democratic former state representative Chrstina Hale (191,226; 45.9 percent). RESULTS
  • Iowa-1 (northeastern quarter of Iowa, including Cedar Rapids and Dubuque): in another race that I thought would not be at all competitive, I was blindsided by a loss for the Democratic incumbent. Republican former Cedar Rapids TV news anchor Ashley Hinson (212,088 votes; 51.3 percent vote share) defeated Abby Finkenauer (201,347; 48.7 percent). RESULTS
  • New Mexico-2 (the southern part of the state, including the southern part of Albuquerque, Las Cruces, and Roswell): a rematch of the 2018 election goes the other way, as Republican state representative Yvette Herrell (142,169 votes; 53.8 percent vote share) defeated Democratic incumbent Xochtil Torres Small (122,314; 46.3 percent) by a surprisingly large margin. RESULTS
  • New York-2 (the South Shore of Long Island, including Islip, Massapequa, and Ronkonkoma): New York state assemblyman Andrew Garbarino (164,272 votes; 56.4 percent vote share) won the election to replace the retiring senior Republican Peter King, by defeating Democratic challenger Jackie Gordon (124,071 votes; 42.6 percent). RESULTS
  • New York-11 (Staten Island and far southwestern Brooklyn): State assemblywoman Nicole Malliotakis (136,382 votes; 56.4 percent vote share) returned New York City’s most conservative district to Republican control, defeating one-term Democratic incumbent Max Rose (99,224 votes; 41.1 percent vote share). RESULTS
  • New York-22 (the bridge of the nose, including Binghamton and Utica): in a rematch of 2018 with a different result, Republican attorney and publisher Clauida Tenney (149,769 votes; 50.5 percent vote share) unseated Democratic incumbent Anthony Brindisi (140,644; 47.4 percent). RESULTS
  • Utah-4 (rural central Utah plus the southern edge of Salt Lake City because gerrymandering): As part of 2018’s “blue wave,” Ben McAdams made this the most Republican-leaning district with a Democratic representative – but like the Alabama Senate, it was always going to be tough to hold the seat. Republican candidate and former Jets and Raiders American football safety Burgess Owens (179,688 votes; 47.7 percent vote share) defeated McAdams (175,923; 46.7 percent). Someday I will look up former American football players in Congress and decide an all-star team, but this post is already long and is about to get longer. RESULTS

Margins of victory

Elections I predicted would be close,
and their actual margins

As I had discussed when I evaluated my Senate picks on Saturday, I predicted not only the winner of each contest, but also which would be close (within 5% vote share) and very close (within 1% vote share). How did I do?

I predicted that 26 races would be close – decided by a vote share of more than 1 percent but less than 5 percent. I looked up the actual vote share margin of those 26 races, defined as the percent vote share for the winning candidate minus that for the losing candidate. The table to the left (or above if you’re on a small screen) shows the results of those 26 races.

Because of votes for third-party candidates, the vote shares do not always add up to 100 percent; but that doesn’t matter because I only considered the difference. For a few of the close races I predicted, I missed not only the margin but also the winner. Those races are described in more detail in the Hits and Misses section above, and I somewhat arbitrarily decided to double the penalty for missing the margin. Of my 26 predicted close races…

NINE were actually close, decided by between 1 and 5 percent once I applied the double-penalty for picking the wrong winner. The one that I missed most hilariously was New York-11, described above, where I predicted a close win for Max Rose but Nicole Mallonakis won by a shocking 15.8 percent.

What about the races that I predicted would be very close, within one percent vote share or less? The table to the left (or below if you’re on a small screen) show the results.

Elections I predited would be
very close, and their
actual margins

Of the 11 races that I predicted would be very close, two actually were very close – the two in California that are so close we don’t yet know who won. One more (OK-5) was within 5 percent. The other eight were decided by even more than 5 percent. Most hilariously wrong were the elections in NY-2 and NY-24, where I picked the wrong candidate, and the other candidate won by 13.8 percentage points.

Correctly calling of 26 close and 2 of 11 very close races is not great, but even that doesn’t tell the full story, because there were races that I did not expect to be close that turned out to be close. The most obvious example of a miss to that side was Iowa-1, where I picked incumbent Abby Finkenauer to win easily, but actually she lost by 3.6 percentage points. In addition to her and Donna Shalala in Florida-27 (who lost by 2.8 percentage points), there were an additional 17 where my predicted candidate won, but by less than 5 percentage points. Those unexpectedly close races are shown in the table below.

What do all 19 of these districts have in common? They were all occupied by Democratic representatives. Two of them have changed parties and are now represented by Republicans. The other 17, like the House as a whole, remain Democratic, but by a much smaller margin than expected.

District Margin
Races that were unexpectedly close,
with their actual margins

That was supposed to be the end of this post – an evaluation of how I did in predicting the 2020 House elections. Not great, but not bad for a first attempt. I’ll take it.

But this is not the end of this post, because in fact, doing this comparison has made me realize something that I hadn’t realized before, and I haven’t heard anyone else observe either: the Republicans came very close to winning the House. Once all the seats are decided, it looks like the balance of power will be Democratic 220, Republican 205. If eight elections had gone the other way, the balance would be Republican 213, Democratic 212.

How many additional Republican votes would it have required to swing the eight closest elections in their favor? Those are the districts in the second through ninth row of the table to. the right (or above if you’re on a small screen), from Illinois-14 to Texas-15. I’m calculating that number now, and I’ll let you know on Wednesday.

And if there is one ironclad rule in American elections, it’s that the President’s party suffers significant losses in Congress in the first midterm election. Which means that there is a good chance that the Republicans will control both houses of Congress in 2022. What impact would that have on our system?

Hang on folks, because I think I found a game-breaking glitch in democracy.

More on Wednesday.

The 2020 Senate Elections: How did I do?

On Wednesday, I took a look at back at my predictions for the 2020 U.S. Presidential election to see how I did. I successfully predicted every electoral vote except for the 29 in Florida and one in Nebraska’s second congressional district. But as important as this year’s Presidential election was – after all, everyone said it was The Most Important Election Of Our Lifetimes(TM) – it was not the only election this month, nor was it the only election I predicted.

I also made predictions for the 35 open seats in the 2020 U.S. Senate elections, and the 435 open seats in the 2020 U.S. House elections (including more detailed predictions in part 2 and part 3). Today, I’ll evaluate my predictions for the Senate. Up next, the House.

How did I do in predicting the results of those elections?

What We Still Don’t Know

The answer is, “we don’t know for sure yet,” because there are a few elections whose results are not yet known. Both of Georgia’s senate elections finished with no candidate claiming more than 50 percent of the votes, meaning they are headed to a runoff on Tuesday, January 5th, 2021. One runoff election will match Republican David Perdue (49.7% election day vote share) and Democratic candidate Jon Ossoff (47.9%); the other will match the Democratic winner Raphael Warnock (32.9% election day vote share) against the higher vote-getter of the two major Republicans, Kelly Loffler (25.9%). Republicans will be favored to win both elections.

So there are those two uncertainties, but that still leaves 33 certainties. How did I do in my predictions?

Hits and Misses

My top line result was that the Senate would be split evenly, 50 seats to 50, and that the Democratic Party would win tiebreakers thanks to Vice President Kamala Harris. That prediction could still come true, if both runoff elections in Georgia go Democratic. That is definitely possible, and as a Democratic not-in-Georgia Party Member I will do what I can to make it happen, but it’s unlikely. I think it’s more likely that the final Senator count will be Republican 52, Democratic 48.

The map below compares my predicted Senate results to the actual results as we know them now. Each hexagon represents a senator. Their colors show the party affiliation of the election winner; red is republican and blue is democratic. The pink, cyan, and lavender hexagons are incumbents (Republican, Democratic, and Independent respectively) who were not up for re-election this year. The as-yet-undecided races in Georgia are dark purple. Races that I mis-predicted are shown in BOLD ALL-CAPS.

Checking my predictions for the 2020 U.S. Senate elections. Click for a larger version.

Not counting my as-yet-unconfirmed Georgia predictions, I missed on two, both Republican incumbents. In North Carolina, Thom Tillis (R-NC) beat lawyer Cal Cunningham; in Maine, Susan Collins (R-ME) beat state house speaker Sara Gideon.

Margins of Victory

What about my predictions for which races would be close? Not including Georgia, I predicted that eight races would be decided by five percent vote share or less, and that two would be decided by one percent or less. In reality, none of the elections were that close: none were decided by one percent or less, and only three were decided by five percent or less. Although was wrong about Tillis’s win in North Carolina, I did predict the election would be within five percent (the final margin was 1.8 percent). I was also right about Mark Kelly in Arizona (2.4 percent) and Gary Peters in Michigan (1.7 percent).

But I missed badly on some other predictions of close elections. I had predicted a less than 1 percent margins for Gideon in Maine; in reality, Collins won by 8.9 percent. I predicted a Joni Ernst win in Iowa by less than 1 percent; the actual margin was 7.9 percent. And I missed on five of the five-percent races as well: John Hickenlooper won by 9.3 percent in Colorado, Steve Daines by 10 percent in Montana, Lindsey Graham by 10.3 in South Carolina, Roger Marshall by 11.9 in Kansas, and Tommy Tuberville by 20.4 in Alabama.

The lesson here? Predicting a close election is good for pageviews, but they can’t all be close.

The bigger lesson? Revisiting predictions to see what has come true and what has not is a powerful way of learning how to do better and better. It’s so easy, I wish more people were willing to do it.

Coming up next: my predictions for the House of Representatives.

The 2020 Presidential Election: How did I do?

Two weeks and a day after the 2020 Presidential election, we finally know the results of each state. The last to be called was Georgia, and thus the final electoral map looks like this:

The final electoral map for the 2020 Presidential election: Biden 306 Trump 232

So now that we know the final results, how did my prediction go?

Permit me a humblebrag, except maybe without the humble part, because I got

48.8 out of 50 states right!

Awwwwwwwww yeah.

Photo of The Dude from The Big Lebowski
Joel Zimba, probably

I missed on only Florida (29 electoral votes) and Nebraska’s second electoral district (1 electoral vote). I finished tied with Nate Silver, who also missed only Florida and Maine’s second congressional district going to Trump. My electoral vote total was farther off because of Florida’s 29, so the winner of the predict-off is… Joel Zimba! Congratulations Joel!

What my map proves, and Joel’s, and Nate Silver’s, is that once all the votes were counted, the polls were pretty much right on, with one major exception. Polling strongly underestimated Trump’s performance in Miami-Dade County, Florida, where Trump’s messaging was far more successful with Cuban-Americans than anticipated.

A histogram peaking at 45
The probability of successfully predicting different numbers out of 50 states at 89% probability for each

But remember that because of the electoral college, it’s not one election, it’s 50 separate elections. When you predict 50 different states with Nate Silver’s expected 89 percent probability (and yes, I know it wasn’t the same for every state, this is just a guesstimate), the probability of getting at least one state wrong is more than 99 percent.

The point of this is not to excuse the states that I missed – because seriously folks, I missed very little – but to remind us all that predicting future events is hard. All we can do is the best we can with the information that we have at the time.

And sometimes we do very well indeed.

We have a President!

Photo of Joe Biden in front of a flag and campaign sign
President-Elect Joseph R. Biden

2:05 PM ET: Congratulations to Joe Biden on being elected the 46th President of the United States!

I know that many of us wished the decision could have been announced much sooner, but I am glad we took the time to make sure we got it right. In a Democracy, the people choose their leaders according to established rules , no exceptions.

Now that we know who the President is, I can finally share some insights into the 2020 U.S. federal elections.

3:23 PM ET: On Election Night, it appeared that Trump had opened a big lead, and that Biden slowly cut into the lead in multiple states before being declared the winner this morning. That is completely false.

In reality, the difference is solely due to the fact that it takes some time to count ballots. For example, the earliest projections on Tuesday night had Trump leading the vote in Rhode Island, causing me to literally shout “WTFRI” at my TV. But when it was finallly announced that Biden won 59 percent of votes in Rhode Island, it was because Biden had ALWAYS won 59 percent of votes in Rhode Island. Every one of those votes was cast on or before Election Day.

3:50 PM ET: When I looked at the preliminary map on Wednesday morning, and it looked like the entire race came down to a few thousand votes in Nevada, I was scared. My biggest fear was that a close election would lead to a long and rancorous legal battle that would make Bush v Gore look like a pillow fight, and that Trump would refuse to leave. And the difference is that in 2000, I had faith that the Supreme Court was a nonpartisan body that could make a decision in the best interests of the country. In 2020, I no longer believe that.

An astronaut with an American flag patch looks at a preliminary map of the election, while an astronaut with a Nevada flag patch points a gun at him. Caption: "Wait, it all comes down to... Nevada?"
Fortunately it also comes down to Pennsylvania and maybe Georgia

Fortunately, Biden also won Pennsylvania, was confirmed by a wider margin in Arizona, and seems more likely than not to also win Georgia. Those are important because it means that Trump’s lawyers would have to come up with four separate strategies for challenging the election, one for each state. When Trump realizes how much work that will be, he will do what he always does: give up.

I’m not even going to address the content of Trump’s accusations of voter fraud. There is absolutely no evidence of any large-scale irregularities in any election in any state anywhere. I will say that of the people I know who supported Trump, most have graciously accepted the result and have encouraged Trump to do the same. Unfortunately some have not, and their reasoning seems to be “Trump lost, so the election could not possibly have been fair.” I heard the same from a very small number of Clinton supporters in 2016, and they were wrong also. Both-sides-ism is stupid, but I am willing to say “both sides are wrong” on those occasions when both sides actually were wrong.

5:53 PM ET: Of course, the Presidential election wasn’t the only game in town on and after Tuesday, and for the Congressional elections, Republicans have reason to be pleased. With four Senate seats yet to be decided, Republicans have lost two (Colorado and Arizona) and gained one (Alabama). Votes are still being counted in North Carolina and Alaska, but the Republican incumbents are nearly certain to win there. That leaves the Senate evenly split at 48-48, with two seats in Georgia remaining. Because of course they do, Georgia has a bizarre system where if no one receives more than 50% of the vote, the top two candidates go to a runoff election on January 5, 2021.

If Democratic candidates David Ossoff and Raphael Warnock win those runoffs, they would become the two U.S. Senators from Georgia, the Senate would be evenly split 50-50, and the Democratic Party would control the Senate thanks to a tiebreaking vote from Vice President Kamala Harris. That is… unlikely.

In the House of Representatives, the news for the Democratic Party is both better and worse. With votes still being counted for 26 seats, the current distribution is Democratic 214, Republican 195. The good news is that they will almost certainly maintain their majority. The bad news is that they massively underperformed expectations. They were expected to gain about 5 seats; instead it looks like they will likely lose about 5 seats.

Even with a fairly decisive electoral and popular vote win for President-Elect Biden, the losses in Congress will make it difficult for the Democratic Party to claim a legislative mandate. And the dream of expanding the Supreme Court is most certainly dead.

Check back here throughout the day, and beyond, for thoughts.

2020 U.S. House forecast part 3: close races, New Jersey to Texas

So far I’ve brought you several forecasts of the 2020 U.S. Presidential election (the most recent), and a forecast for the nail-bitingly-close Senate election, which I predict will messily resolve into the thinnest possible margin for a Democratic majority. Now the House elections, which I predict will result in a slight Democratic gain.

This was originally going to be a single post, then a 3-part series. So, welcome to my 4-part series of 2020 House election predictions!

Today, previews of close races – those where I predict the vote will finish within 5 percentage points – in the alphabetical first half of states, from Alaska to Minnesota, then skipping over the “New” states directly to North Carolina because apparently I cannot into alphabet.

The National Map

Below is my forecast map for what the entire House will look like when the 117th U.S. Congress opens on January 3, 2021 (for comparison, here’s what the map looks like now). See the guide below for what each part of the figure means, and definitely click on it for a version you can actually read.

My too-small-to-read prediction of what the U.S. House of Representatives will look like after next month’s election. Click for a version you can actually read.

How to read the map (skip if you know it already)

As is traditional yet completely arbitrary, Republican representatives are shown by red hexagons and Democratic representatives are shown by blue hexagons. The text labels show the names of the incoming representative; plain text means that I predict the incumbent will be re-elected, bold means that I predict a new representative will be elected from the same party, and bold all-caps means that I predict a party switch. A larger font size and a single asterisk(*) mean the election is likely to be close, within about 5%. An even larger font size and double asterisk(**) mean the election is likely to be very close, maybe within 2%.

Preview of close races (New Jersey to Texas)

In my Senate prediction, I offered at least a passing comment on each of the 35 races. With 435 House races, I’m obviously not going to do that, but I have looked at them all to get a sense of which will be most exciting.

Here they are, starting with the close races marked with a single asterisk (*) in the map above.

House races likely to vote within 5 percent (52%-47% or closer)

New Jersey 7

This district covers several of the rural and exurban communities of northwestern New Jersey. With a median household income of $104,000 per year, it is the richest congressional district on the close races list, and fifth-richest in the country. In 2018, Democratic challenger Tom Mailnowski upset five-term Congressman Leonard Lance. He is running for re-election against State Senator Tom Kean. There have been no polls in this district since March 11th, when dinosaurs roamed the Earth. So who knows. But incumbent + slight Democratic lean to the district + probably a good year for the Democratic Party overall =

Prediction: Malinowski re-elected

New Mexico 2

New Mexico, like the Mongols, is the exception. It’s a western state, mostly rural, with only one metro that feels like it’s bigger than it really is – the Albuquerque metro area has a smaller population than those of Fresno, California or Greenville, South Carolina. And yet, because of its significant Mexican-American and Native American populations, it tends to vote Democratic in national elections, enough that it has moved completely out of swing state territory.

New Mexico’s second district encompasses the southern half of the state, plus a gerrymandered tentacle (gerrycle?) reaching up to crack Kirkland Air Force base in half.

In 2018, longtime Republican representative Steve Pearce (not that one) stepped down to run unsuccessfully for Governor or New Mexico, and Democratic candidate Xochtil (pronounced SHOCK-til) Torres Small won an extremely close election against Republican Yvette Herrell – so close that it appeared Herrell had won on election night, but Torres Small pulled ahead once all the votes were counted. The same two face off again this time, with Torres Small as the incumbent. Polling is once again extremely tight, with Torres Small leading by 3 in the average.

Prediction: Torres Small re-elected.

New York 11

New York has 27 House districts, 10 of which are entirely within New York City. One of those is the 11th, which includes all of Staten Island and some neighborhoods at the southern tip of Brooklyn; it is the only district in New York City that voted majority for Trump in 2016 (at 53%). It is currently represented for the Democratic Party by Max Rose. He is running for re-election against state assemblywoman Nicole Malliotakis. Rose is ahead by 5 points in aggregate voting and is likely to win.

Prediction: Rose re-elected.

New York 22

This district runs north-south through the bridge of New York’s nose, between the Adirondacks and Lake Ontario. It includes the cities of Ithaca, Binghamton, Newburgh, and Poughkeepsie. In 2018, Anthony Brindisi won the seat back for the Democratic Party by defeating then-incumbent Republican Claudia Tenney. The same two face off again this time, with Brindisi 4 points ahead in aggregate polling.

Prediction: Brindisi re-elected.

Ohio 1

Ohio-1 is another gerrymanderiffic district looking like a pair of sexy nunchucks to crack apart Cincinnati. It has been represented by Republican Steve Chabot in 12 of the last 13 Congresses, missing only on term after losing in the blue wave of 2008 but getting voted back in in 2010. His opponent is Cincinnati public health worker Kate Schroeder. The race is tightening, but Chabot is up by 2.5 percentage points with two weeks left.

Prediction: Chabot re-elected.

South Carolina 1

This is another gerrymanderific district, centered on the city of Charleston, South Carolina. It includes the southern half of South Carolina’s coastline, plus one of those beefy arms for good measure. The district typically votes Republican in the Presidential election, by double-digit margins. But in 2018, the district elected its first Democratic representative by less than 4,000 votes. That representative is Joe Cunningham, and he is running for re-election for the first time against state representative Nancy Mace. Normally it would be an uphill battle for Cunningham in such a solidly Republican district, but Mace’s popularity is fading along with Trump’s, and Cunningham is now up by 3 in aggregate polling.

Prediction: Cunningham re-elected

Texas 7

In the 2020 House elections, the eyes of the world are on Texas. The first of five close races we’ll look at is Texas’s 7th congressional district, which covers the western suburbs of Houston. Democratic incumbent Lizzie Fletcher is running for re-election against real estate agent and former Army helicopter pilot Wesley Hunt. Fletcher led by 4 when I made the map, but her lead has extended to 5 since.

Prediction: Fletcher re-elected

Texas 21

I don’t even know how to describe the shape of this gerrymandered district. Evil Massachusetts? A snuffleupagus with a goatee smoking a pipe? Regardless, it exists for only one reason: to crack the cities of Austin and San Antonio by shoving as many of their voters as possible into a weird district, while also including enough rural area to the west so that rural voters will maintain a slight edge in numbers.

But I digress.

In 2018, 16-term Republican incumbent Lamar Smith retired, and Republican Chip Roy won the seat. His opponent this time is a hero of the Texas Democratic Party: Wendy Davis, whose 13-hour filibuster succeeded in stopping passage of a restrictive abortion bill (although it passed in the next legislative session). Roy is ahead by just under 4 points in aggregate polling.

Prediction: Roy re-elected

Texas 22

Here’s another weirdly gerrymandered district, covering the southern suburbs of Houston. Painfully asymmetric breasts?

Six-term incumbent Republican Pete Olson is retiring, opening the seat to a new occupant. The new occupant will be either Fort Bend County Sheriff Troy Nehls or foreign service officer and national security expert Sri Preston Kulkarini. The candidates’ signature issue issues are law enforcement and health care. Guess which is which!

The race is tightening, but aggregate polling has Nehls 3 points ahead.

Prediction: Nehls wins, holding the seat for the Republican Party

Texas 23

This district covers a vast rural area of western Texas, from the outskirts of San Antonio to the outskirts of El Paso. It takes 8 hours to drive from one end of the district to the other. The district is majority Mexican-American, but the inclusion of the highly conservative San Antonio suburbs, plus the cities of Midland and Odessa, makes for a hotly contested district. It is currently represented by Will Hurd, of the most moderate and bipartisan-willing Republicans in Congress. Presumably sick of the bullshit, Hurd surprised many by announcing he would not seek re-election, issuing what may be the most politely shade-throwing retirement statement of all time:

I’m leaving the House of Representatives to help our country in a different way… It was never my intention to stay in Congress forever, but I will stay involved in politics to grow a Republican Party that looks like America.

Soon-to-be-former Rep. Will Hurd (R-TX-23)

I like this guy.

His replacement will be either his 2018 opponent, Air Force veteran and national security consultant Gina Ortiz Jones, or conservative populist Navy veteran Tony Gonzales. Ortiz Jones is ahead by 5 points in polling.

Prediction: Ortiz Jones is elected, switching the seat to the Democratic Party. And also becoming the first openly LGBT person elected to Congress from a southern state.

Party Switches

If you’re keeping score at home, we’ve had three party switches after the review of close races, with a net gain of +1 for the Republicans:

MI-3: Justin Amash (L) -> Peter Meijer (R)
MN-7: Collin Peterson (D) -> Michelle Fishbach (R)
TX-23: Will Hurd (R) -> Gina Ortiz Jones (D)

But if you remember my overall prediction from the map, it was +5 Democratic. So where will the Democratic Party get those extra six seats from?

Tune in Friday for the epic conclusion, the House races that I think could be decided by 2 points or fewer!

2020 U.S. House forecast part 2: close races, Alaska to North Carolina

Continuing today my forecast for the 2020 U.S. elections, now just 15 days away!

Previously, I gave my Presidential election forecast (the most recent), a Senate forecast, and an overall House forecast. The House forecast was originally going to be a single post, then a 3-part series. So hey, welcome to my 4-part series of 2020 House election predictions!

Today: the first half of previews of close races.

The National Map

Below is my forecast map for what the entire House will look like w2hen the 117th U.S. Congress opens on January 3, 2021 (for comparison, here’s what the map looks like now). See the guide below for what each part of the figure means, and definitely click on it for a version you can actually read.

My too-small-to-read prediction of what the U.S. House of Representatives will look like after next month’s election. Click for a version you can actually read.

How to read the map (skip if you know it already)

As is traditional yet completely arbitrary, Republican representatives are shown by red hexagons and Democratic representatives are shown by blue hexagons. The text labels show the names of the incoming representative; plain text means that I predict the incumbent will be re-elected, bold means that I predict a new representative will be elected from the same party, and bold all-caps means that I predict a party switch. A larger font size and a single asterisk(*) mean the election is likely to be close, within about 5%. An even larger font size and double asterisk(**) mean the election is likely to be very close, maybe within 2%.

Preview of close races (Alaska to North Carolina)

In my Senate prediction, I offered at least a passing comment on each of the 35 races. With 435 House races, I’m obviously not going to do that here, but I have looked at them all to get a sense of which will be most exciting.

Here they are, starting with the close races (forecast to be within 5 percentage points, meaning a 52.5%-47.5% vote). These races are marked with a single asterisk (*) in the map above.

I’m running through previews alphabetically by state name, and today I preview states from Alaska through Minnesota. I am then skipping over all the “New” states directly to North Carolina, because apparently I cannot into alphabet.

Alaska at-large

A moose resting in a field
Rep. Don Young (R-AK).
Click to see what he actually looks like.

This district covers the entire state of Alaska, and it’s the second-largest electoral district by land area in the world (only the Nunavut district for the Canadian House of Commons is bigger).

Republican Don Young has represented the district since 1973, and considering his long service it’s shocking that he’s in any race at all with challenger Alyse Galvin, an Anchorage-based education consultant and activist.

The race was within 5% when I made the map, but Young has pulled more than 6 points ahead in polling, so he is virtually certain to return for his 25th term in Congress.

Prediction: Young re-elected.


This district covers central Arkansas, including the state capital of Little Rock (aerial photo below). Like most districts containing urban centers, it is far more friendly to Democratic candidates than the rest of the state. Except it’s Arkansas, so it’s still not at all friendly to Democratic candidates.

An aerial photo of Little Rock, Arkansas. A metal bridge over a river is in the foreground, skyscrapers are in the middle of the picture and forests in the background.
Downtown Little Rock, the heart of Arkansas’s 4th congressional district

Republican third-term incumbent French Hill currently leads Joyce Elliot (who represents Little Rock in the State Senate) by 5 percentage points. The race is getting closer, but time is getting short.

Prediction: Hill re-elected.


Hiral Tipirneni's campaign logo. It says "Dr. Hiral Tipirneni for Congress," with an outline of Arizona next to "for Congress" and a stethoscope wrapped around the "Dr."
Dr. Hiral Tipirneni’s campaign logo. Did you know she is a doctor?

This district covers the northeastern suburbs of Phoenix, including the cities of Scottsdale and Paradise Valley.

Four-term Republican David Schweikert faces emergency room doctor and health care advocate Hiral Tiperneni. Schweikert leads by 3 points in a tightening race.

Prediction: Schweikert re-elected.


Gil Cisneros (left) and Young Kim (right) – a picture label, not a political commentary

This district covers the foothills east of Los Angeles -parts of Los Angeles, Orange, and San Bernardino counties – including Yorba Linda, the hometown of Richard Nixon.

In 2018, a longtime Republican Congressman retired and Gil Cisneros sniped the seat for the Democratic party, narrowly winning the seat in a recount against state assemblywoman Young Kim. The same two are running again, and Cisneros currently leads by 5 points.

Prediction: Cisneros re-elected.


Supergirl, from the CBS TV series
Candidate for CA-48 Michelle Steel (click to see what she actually looks like)

This district covers the coast of wealthy, traditionally Republican Orange County, from Huntington Beach to Laguna Beach. In 2018, Harley Rouda became the first Democratic representative in the district’s history. Rouda is running for re-election against Orange County Councilwoman and self-proclaimed “Tax Fighter,” the impressively-named Michelle Steel. Rouda is ahead by a bit under 3 points. I’m less sure about this one – but when the incumbent is ahead with less than a month left, go with the incumbent.

Prediction: Rouda re-elected.


A map of Florida with the 26th District highlighted in green - the southern tip and the Keys
The Tip of the Wang: Florida’s 26th Congressional District

This district covers the Tip of the Wang, from Homestead in southern Miami-Dade County to all of the Florida Keys. I’m amazed that in the ridiculously close President/Senate state of “It All Comes Down To” Florida, this is the only remotely close House race.

First-term Democratic incumbent Debbie Murcasel-Powell goes up against Miami-Dade County mayor Carlos Giménez (note that “mayor” here is equivalent to County Commissioner in other counties. Not to be confused with the mayor of the city of Miami, who is just as Cuban-American, young, and good-looking as you would expect).

Like Alaska, this is a race that has separated even since I made my map, and Murcasel-Powell is virtually certain to win re-election.

Prediction: Murcasel-Powell re-elected.


Portrait of German composer Georg Frideric Handel (1685-1759)
Former Rep. Karen Handel (R-GA-6). Click to see what she actually looks like.

This district includes the northern suburbs of Atlanta – and as always, as go the suburbs, so go the nation. In a 2017 special election, the district famously gave Democratic pundits hope when Jon Ossoff lost to Karen Handel by about 9,000 votes – thus demonstrating just how desperate for hope Democratic pundits were in 2017.

But in 2018, Democratic prospects were better, and Lucy McBath captured Handel’s seat by only 3,264 votes. It’s McBath vs. Handel again this year, and McBath is just over 3 percent ahead in current polling.

Prediction: McBath re-elected.


Illinois District 13 shown in green on a map of Illinois. It runs across the center of the state from the mid-east to the western border, with weird tentacles to the northwest and south
Illinois’s 13th Congressional District. Gerrymander? I ‘ardly know er!

This is a gerrymaderific district that mostly includes rural west-central Illinois, but has weird squid tentacles that include small parts of the state capital of Springfield and the college towns of Champaign and Bloomington because reasons.

The district hasn’t voted Democratic since ’92…….. 1892, that is. In 2018, Republican Rodney Davis won re-election against challenger Betsy Dirksen Londrigan by a ridiculously thin margin of 2,058 votes.

Davis and Dirksen Londrigan play it again this time, with Davis ahead by about 3 percent in the polls.

Prediction: Davis re-elected.


An artistic drawing of a unicorn standing next to a cliff
A third-party member of Congress (Justinus Amishii), a species hunted to extinction. Click to see what Justin Amash (L-MI-3) actually looks like.

This district includes the city of Grand Rapids and the surrounding area of central Michigan. Justin Amash was elected as a Republican in 2010, but ran afoul of President Trump and switched parties – first to Independent and then to Libertarian, becoming the first third-party representative in Congress since 1950. th

Understandably, Amash has had enough and is not running for re-election. The seat will either go back to the Republicans via Peter Meijer, heir to Michigan’s Meijer supermarket chain – or else will flip Democratic with immigration lawyer and education activist Hillary Scholten. Polling is absolutely all over the place, but the average has Meijer ahead by about 3 points.

Prediction: Meijer wins, capturing the seat for the Republican Party.


A photo of a moose resting in a field
Rep. Jim Hagedorn (R-MN-1). Click to see what he actually looks like.

This is a rural district that stretches across the whole southern border, and includes midsize cities like Mankato, Rochester, and Winona.

The district has traditionally leaned very, very slightly Democratic, but it currently represented by massively conservative, massively controversial Republican Jim Hagedorn. In 2018, Hagedorn defeated Democratic candidate Dan Feehan by just 1,311 votes. Feehan, an Iraq War veteran and former middle school math teacher, is running again. Some polls have Feehan ahead, but on average, Hagedorn leads by 4 points.

Prediction: Hagedorn re-elected.


A photo of a moose resting in a field
Rep. Collin Peterson (D-MN-7; click to see what he actually looks like).

This is another rural district, eh? It is the largest district by area Minnesota, covering the whole western side of the state except the far south, with the largest city being Moorhead. It’s the most conservative part of Minnesota, with a long history of Representatives with wonderfully Scandinavian names like Haldor Boen, Ole Juulson Kvale, and Odin Langen.

The district has been represented since 1990 by Collin Peterson, one of the founders of the Blue Dog Democratic Coalition. But thanks to the USA’s increasing partisanship and a strong candidate – Lieutenant Governor Michelle Fishbach – the Republicans seem poised to capture the seat from the 15-term incumbent. When I made the map, Fishbach was 4.5 points ahead when I made the map, but her lead has increased to six points.

Prediction: Fishbach captures the seat for the Republicans.

Montana at-large

Photo of Rep. Greg Gianforte (R-MT), a balding, smiling white man in a black suit
Rep. Greg Gianforte (R-MT)
Yes, I know Montana has moose, but YOU WERE EXPECTING IT THIS TIME, WEREN’T YOU?

This district represents the entire state of Montana, and it’s actually the largest district by population in the country. Incumbent Greg Gianforte is stepping down to run for Governor of Montana.

The Republican candidate is state auditor Matt Rosendale; the Democratic candidate is former state representative Kathleen Williams. Rosendale has led the entire way, but Williams has pulled within 5 points – not very close with so little time remaining.

Prediction: Rosendale wins, retaining the seat for the Republican Party.

North Carolina-8

A moose resting in a field
Click to see what Richard Hudson (R-NC-8) actually looks like.

This district covers rural south-central North Carolina, including the midsize cities of Concord, Albemarle, and Pinehurst.

The district is currently represented by Republican Richard Hudson. He is running or a fifth term against Patricia Timmons-Goodson, an Associate Justice of the state Supreme Court. Polls are tightening, with Hudson currently holding a 3-point lead.

Prediction: Hudson wins re-election.

Party switches

If you’re keeping score at home (can’t tell the players apart without a program!), that’s two seats that I predict will switch parties:

  • Michigan-3: From Libertarian to Republican (although Justin Amash was originally elected as a Republican), captured by Peter Meijer
  • Minnesota-1: From Democratic to Republican, captured by Michelle Fishbach

That’s two seats over the Republicans, but remember that my overall prediction was five seats turning Democratic. That means that there are some switches that way still to come in my House forecast previews.

Coming Wednesday: Close races from New Jersey to Texas.

Coming Friday: An additional ten races that I predict will be very close – within two percent at 51%-49% or closer. Which means that they will likely require a recount, bringing us more drama for weeks after the election. Getcha popcorn.