KEY POINTS FROM THIS ARTICLE
— A race-by-race analysis of Democratic House targets shows the party is close to winning the majority, but they do not have it put away, in our judgment, with Election Day less than three weeks away.
— Barring a big, positive late change in the political environment in favor of Republicans, the bare minimum for Democratic House gains is in the mid-to-high teens. The needed 23-seat net gain is not that far beyond that and there are many different paths Democrats can take to achieve it. So the GOP is still at a disadvantage overall.
— There are 11 ratings changes this week, seven in favor of Democrats and four in favor of Republicans.
— Note: With the election so close, and with the Crystal Ball planning to eventually offer a projection in every general election House, Senate, and gubernatorial race (as per our tradition), we are working to reduce the number of Toss-ups in our ratings, not add to them.
— We are not making any changes to the Senate and gubernatorial ratings this week, but we are including our current assessment of the state of play in Maps 1 and 2 before focusing exclusively on the House this week.
Table 1: Crystal Ball House ratings changes
Table 2: Crystal Ball House ratings
Map 1: Crystal Ball Senate ratings
Map 2: Crystal Ball gubernatorial ratings
The race for the House, district by district
For all the talk of the House generic ballot, President Trump’s approval rating, and other big-picture factors that point to the overall direction of the fight for the House, the battle for the majority comes down to a district-by-district slog.
And it is a district-by-district assessment that shows the Democratic path to the House majority — and the remaining avenues for Republicans to block it.
We have previously described Democrats as being “soft favorites” in the race for the House, and that’s basically where we’re still at. As this roadmap hopefully will demonstrate, Democrats may yet blow open the battle for the House and win the majority comfortably. But there are paths for the Republicans to hold on as well, and a Democratic House majority is not yet written in stone.
This is the third update of a piece we first published in February and updated in May. For consistency’s sake, we’ve kept the same districts in the same groupings, but to reflect the changing battlefield, we’ve added a 13th category (there were 12 originally) as a way to account for some prime Democratic targets that we did not account for previously. We also have added a few races to the various preexisting categories where warranted.
Also, this listing does not take into account every single Democratic House target. Remember, this is just a way of assessing the Democrats’ path to a bare majority, or a 23-seat net gain. If Democrats gain significantly more than that, they very well might win districts that are not mentioned here. As has become clear in recent days, even Democrats in longshot districts are running real races and raising tons of money. We’re not going to obsess over the overwhelming third quarter Democratic fundraising here, but one overall point is this: If the Democrats don’t win the House, it won’t be for lack of resources. Many Republicans used to safe races will still win but with reduced vote shares and higher stress levels. Some may even get caught unaware, producing shocks on Election Night.
Still, we think the 66 Republican-held districts listed as part of this roadmap are the Democrats’ best targets, collectively, and the ones that are likeliest to make (or break) their quest for the majority.
We went with the name “Drive for 25” as an homage to a 2012 House Democratic slogan and to reflect the fact that while Democrats need to net 23 seats, they probably are going to lose at least a couple of seats they currently hold to Republicans. Even in years where one side makes substantial gains in the House, the other side almost always flips at least a seat or two. This year, Republicans appear to be guaranteed at least one pickup: PA-14, an open seat in southwestern Pennsylvania that is a more Republican-leaning version of the old PA-18, the seat now-Rep. Conor Lamb (D) won in a special election in March. Lamb is running against Rep. Keith Rothfus (R) in a new district, PA-17 (more on that race below).
Republicans also are targeting two open seats in Minnesota, MN-1 (covering southern Minnesota) and MN-8 (covering the northeastern part of the state, including the traditionally Democratic but GOP-trending Iron Range). The president carried both of these districts by about 15 points. We’re moving one, MN-8, from Toss-up to Leans Republican this week after the Democratic Congressional Campaign Committee appeared to write off the district. The GOP may very well end up picking up MN-1 as well, which remains a Toss-up. If so, that would mean Democrats would have to win 26 currently Republican-held seats to win the House. Additionally, two competitive, Democratic-held open seats in Nevada, NV-3 and NV-4, seem highly competitive. We still favor Democrats in both but it wouldn’t be a shock if the GOP picked off one or the other. Again, Democrats would have to make up any GOP gains through gains of their own among current Republican seats. To us, the five seats mentioned above are the only truly real GOP offensive targets; two other districts, the open, frequent battleground district NH-1 and a Trump-won seat held by Rep. Matt Cartwright (D, PA-8) in northeast Pennsylvania, move from Leans Democratic to Likely Democratic in this update, as we feel Democrats are in better shape to hold both even as each may be very competitive in future cycles.
With that, let’s go through the 13 categories of Democratic targets. Remember, every district that follows is already controlled by Republicans:
1. Win all four open seats where Democrats already are favored
Seats: AZ-2 (Open), CA-49 (Open), FL-27 (Open), and NJ-2 (Open)
Democratic progress: Progressing but could fall short.
Democrats appear well on their way to putting away at least three of these four seats. We are upgrading the Democrats’ chances in the open AZ-2, moving it from Leans Democratic to Likely Democratic, after the National Republican Congressional Committee stopped spending there. NJ-2 is one of the safest Democratic pickups in the country, and CA-49, the seat from which Rep. Darrel Issa is retiring after a very close call in 2016, has long been the Democrats’ best bet in California.
But the one race giving the Democrats heartburn in this category is FL-27, an open South Florida seat that Hillary Clinton won by 20 points but where there’s a close race between former University of Miami President and Health and Human Services Secretary Donna Shalala (D) and former journalist Maria Elvira Salazar (R), who is hoping to hold the seat as long-serving Rep. Ileana Ros-Lehtinen (R) heads into retirement. The race is probably something close to a tie, with Salazar running a more spirited campaign than Shalala: Mason Dixon-Telemundo 51 had Salazar up 44%-42% in a recent poll. We’re keeping it at Leans Democratic for now, but if Democrats flub any of these four races, FL-27 is the one they would lose, and the uncertainty there is what could cause the Democrats to come up short in this category.
Later on in this piece, we’ll mention another GOP-held seat in South Florida, that of Rep. Carlos Curbelo (R, FL-26). Just like FL-27, FL-26 appears to be very close: Mason Dixon-Telemundo 51 pegged Curbelo’s lead at just 46%-45% recently (Republicans believe Curbelo is in better shape; Democrats believe it’s tied). One could argue that both FL-26 and FL-27 should be Toss-ups; as it stands right now, and because we want to avoid excessive Toss-ups with Election Day approaching, we’re holding FL-27 as Leans Democratic and FL-26 as Leans Republican.
2. At least three more open seats
Seats: CA-39 (Open), MI-11 (Open), NJ-11 (Open), WA-8 (Open), or WI-1 (Open)
Democratic progress: Probably on track.
Based on our ratings, Democrats are favored to reach their target here: We rate MI-11, NJ-11, and WA-8 as Leans Democratic. Of those three, WA-8 is the shakiest: One recent poll from Crosscut/Elway Research showed Dino Rossi (R), a former statewide candidate, up 49%-39% on physician Kim Schrier (D). We still like Schrier’s chances despite the poll, although we very well may be wrong on that. Here’s the thing, though: Even if Rossi wins, Democrats could make up for that in this category by winning CA-39, a true Toss-up in another Clinton-won district. For the GOP to come out ahead here, they probably need to win both CA-39 and WA-8. We see Democratic odds as fairly strong in both MI-11 and NJ-11, two traditionally GOP suburban seats that don’t seem to like the president very much and where Democrats appear to have a candidate edge; if either slips away, Democrats are probably having a bad night relative to expectations.
The one race here where we see the Republicans as favored is departing House Speaker Paul Ryan’s seat, WI-1, which is a clearly right-of-center district in southeast Wisconsin we rate Leans Republican.
3. At least three net seats from Pennsylvania
Seats: PA-1 (Brian Fitzpatrick), PA-5 (Open), PA-6 (Open), PA-7 (Open), PA-10 (Scott Perry), or PA-17 (Keith Rothfus)
Democratic progress: Overperforming.
Bonus district not included in previous roadmaps: PA-16 (Mike Kelly)
When we first set this benchmark of a three-seat net gain for Democrats in Pennsylvania back in February, Lamb had not yet won the special PA-18 House election and, perhaps more importantly, the state’s Democratic-controlled state Supreme Court had not yet replaced a Republican-drawn gerrymander with a map that was much more favorable to Democrats. While Democrats might have made significant gains in Pennsylvania this year even without a new House map in place, the altered district lines have eased their path.
Here’s one place where Democrats seem likely to run ahead of our roadmap, even taking into account their likely loss of PA-14, the new version of the district Lamb won in March. Democrats seem certain to win PA-5 and PA-6, two open seats in Greater Philadelphia that were redrawn to become significantly more Democratic. The Democrats are also favored in an open seat in the Lehigh Valley, PA-7, and Lamb seems like an increasingly bigger favorite over Rothfus in the member vs. member battle in Greater Pittsburgh’s new PA-17. We’re moving that race from Leans Democratic to Likely Democratic as outside Republican groups have abandoned Rothfus. So that means Democrats are in good shape to win at least four current GOP seats in Pennsylvania. And that doesn’t even account for the possibility of a Democratic victory in PA-1, where Rep. Brian Fitzpatrick (R) is in a Toss-up race with philanthropist Scott Wallace (D); they have alternated leads in two recent nonpartisan public polls. The Democrats also have outside shots against Reps. Mike Kelly (R, PA-16) in an Erie-based district and Scott Perry (R, PA-10) in a Harrisburg-based district. Kelly’s district was not included in our May assessment but we’re making note of it here.
In other words, Democrats started the cycle with just five seats in Pennsylvania and are poised to end it with at least nine, and maybe more.
4. Beat at least three of five vulnerable California incumbents in Clinton-won districts
Seats: CA-10 (Jeff Denham), CA-21 (David Valadao), CA-25 (Steve Knight), CA-45 (Mimi Walters), or CA-48 (Dana Rohrabacher)
Democratic progress: Not there yet.
Bonus district not included in previous roadmaps: CA-50 (Duncan Hunter), although Trump carried this district, unlike these others, which Clinton carried.
While Valadao should be fine even though Clinton won his district by double digits — outside groups on both sides have pulled money out of the race, a testament to Valadao’s local strength — the other four contests here are all live Democratic targets. Somewhat surprisingly, the most imperiled of these incumbents appears to be Walters, who has trailed in a couple of recent public polls and whose own internal polling as of a couple of weeks ago appeared to show her in something of a tie. We’re moving Walters from Toss-up to Leans Democratic. She may yet recover, but this move is also a way of reflecting in our ratings our belief that the Democrats should be able to defeat at least one of these incumbents, and quite possibly more, be it Walters and/or one or more of the others.
Denham, Knight, and Rohrabacher all remain very vulnerable as well. That said, some Democrats and other observers believe Rohrabacher, who has a much longer history in his district than the others, may be the hardest to dislodge, in part because his district was the closest in the presidential race (Clinton won it by less than two points — she won the others in this category by more). The parties have vastly differing views on the California races: Some Democrats seem to believe they’re positioned to sweep all the most competitive races (setting aside Valadao), while Republicans believe they are holding up OK given the circumstances.
One other California Republican incumbent who is somewhat newly in danger, at least compared to the last update of this roadmap, is Rep. Duncan Hunter (R, CA-50), who was indicted over accusations that he and his wife misused campaign funds for personal expenses. His San Diego-area district is GOP-leaning, but former Obama White House fellow Ammar Campa-Najjar (D) argues he’s tied with the damaged Hunter (a couple of independent polls show Hunter up by a little or a lot).
California is very important to Democratic House hopes, and remember on Election Night that not only do polls close late there (11 p.m. on the East Coast), but if races are very close, the vote counts could take weeks to finalize.
The bottom line: Democrats should win at least one and probably more of these races, but three out of five is not certain at this point.
One other thing: This roadmap splits the California races into three different categories (the open CA-39 and CA-49 are accounted for elsewhere above). Overall, the target for Democrats in California is probably a net gain of four-to-five seats. That remains a question mark but is also achievable for Democrats.
5. Defeat three of these six Clinton-district incumbents
Seats: CO-6 (Mike Coffman), FL-26 (Carlos Curbelo), IL-6 (Peter Roskam), MN-3 (Erik Paulsen), TX-23 (Will Hurd), or VA-10 (Barbara Comstock)
Democratic progress: On track with potential to overperform.
The category illustrates the different movement in different kinds of districts this year. Four of these six districts — CO-6, IL-6, MN-3, and VA-10 — are all highly affluent, highly educated suburban seats where Hillary Clinton improved on Barack Obama’s 2012 showing. Trump’s standing is weak in all four, and it appears all four incumbents are underdogs. That list now includes Roskam, who we are moving from Toss-up to Leans Democratic. However, Curbelo and Hurd both represent less affluent, more diverse districts (both FL-26 and TX-23 are majority Hispanic), and both are holding up better than the others. We’re moving TX-23 from Toss-up to Leans Republican, as Hurd has held solid leads in several polls and the NRCC has diverted funds from his race in a sign of confidence. As noted above, Curbelo is closer to being in a Toss-up race than Hurd, but we’re holding him at Leans Republican, along with Hurd.
Still, this category is otherwise going well for Democrats. If our Leans Democratic ratings hold up in the other four districts, they will have beaten our benchmark in this category.
6. Win one of these three Clinton-won, historically Republican seats
Seats: NJ-7 (Leonard Lance), TX-7 (John Culberson), or TX-32 (Pete Sessions)
Democratic progress: Probably on track.
Bonus district not included in previous roadmaps: NJ-3 (Tom MacArthur), although Trump won that district.
All three of these races remain true Toss-ups, but it seems like the odds are good Democrats will end up capturing one of the three (these districts are similar to the ones in the previous category, as they are all affluent, highly-educated, suburban seats where Clinton did well compared to prior Democratic performance).
If Republicans hold on to all of these seats, Democrats might be able to make up for it by beating Rep. Tom MacArthur (R, NJ-3), another vulnerable New Jersey incumbent in a Toss-up race. He doesn’t quite fit this category because Trump won his district after Obama carried it in 2012 (these other three districts went from voting for Mitt Romney to backing Clinton), but this is another very close race that needs to be accounted for in this roadmap.
7. Defeat one of two Trump-district freshmen, who were narrow winners in narrow districts
Seats: MN-2 (Jason Lewis) or NE-2 (Don Bacon)
Democratic progress: On track.
These two races have gone different ways in our ratings, with MN-2 now Leaning Democratic and NE-2 now Leaning Republican. Democrats only need one of the two, so they are on track here, but Lewis also won as an underdog in 2016 so he shouldn’t be written off. Bacon’s opponent, nonprofit executive Kara Eastman (D), was an upset primary winner over former Rep. Brad Ashford (D, NE-2), who lost to Bacon in 2016. Democrats cooled on the race after that, but that doesn’t necessarily mean she is a sure loser. But Bacon is favored.
8. Net at least two seats from “Trump York”
Seats: NY-19 (John Faso) and NY-22 (Claudia Tenney), but also possibly NY-1 (Lee Zeldin), NY-11 (Dan Donovan), NY-21 (Elise Stefanik), NY-23 (Tom Reed), or NY-24 (John Katko)
Democratic progress: Not there yet.
Bonus districts not included in previous roadmaps: NY-2 (Peter King), NY-27 (Chris Collins)
Republicans hold nine of 27 districts in New York, and all nine of them are hypothetically competitive, although seven of the nine GOP incumbents are favored in our ratings (Stefanik and King are safe). The two Toss-ups are NY-19 and NY-22, and realistically to meet our benchmark, Democrats would have to win both. They certainly could, but they could also win just one or come up empty-handed. So the Democrats, to us, are behind in this category at the moment.
A wild card here is Collins in Western New York. Collins, who like Hunter in California is running under indictment after being charged with insider trading, represents the most Republican district in New York, but a recent Spectrum News/Siena College poll had Collins up only 46%-43% on Grand Island Town Supervisor Nate McMurray (D). Collins, like fellow indicted incumbent Hunter, remains a favorite in our ratings. We haven’t seen a compelling reason to think of any of the others as Toss-ups, although many of these New York Democratic challengers have raised gobs of money.
9. Win two of these four Trump seats with down-ballot Democratic DNA
Seats: IL-12 (Mike Bost), KY-6 (Andy Barr), ME-2 (Bruce Poliquin), or UT-4 (Mia Love)
Democratic progress: Not there yet.
All of these races seem tight, but it seems like Democrats are having a harder time in these kinds of Trump-won districts as compared to ones that Clinton won or came closer to winning. If Republicans hold on to the House, our best guess is that winning all four of these races would be part of that, and they very well could do so. At the same time, recent polling in all four districts, be it from partisan or nonpartisan sources, have shown very close races. We still give Love an edge in our ratings; the other three are Toss-ups.
10. Net at least one seat from Iowa
Seats: IA-1 (Rod Blum) or IA-3 (David Young)
Democratic progress: On track.
Blum appears to trail state Rep. Abby Finkenauer (D) in an Obama-to-Trump district in Northeast Iowa, and yet there are some conflicting signs. The Democratic Congressional Campaign Committee cut its ad reservation here, but then later went back with a modest buy. Congressional Leadership Fund, the GOP group, also just went on the air with a sizable $1 million ad buy, an indication that it believes the race is still winnable in a district where paid media is fairly cheap. So this one doesn’t seem put away yet for Democrats, although Blum is still the underdog: He trails in his own internal polling (45%-43%), never a good sign for an incumbent.
Meanwhile, IA-3 looks like a genuine Toss-up.
11. Net at least one seat from Kansas
Seats: KS-2 (Open) or KS-3 (Kevin Yoder)
Democratic progress: On track.
It’s unclear whether Yoder has been written off by national Republicans or not: The National Republican Congressional Committee is no longer spending there, but CLF remains engaged. We think Yoder is an underdog, and the KS-2 race could go either way. Trump won KS-2 by double digits, while Clinton narrowly carried KS-3, which covers some of the Kansas City suburbs. Given the trajectory of this cycle, it makes sense that Democrats seem to have a better shot in the suburban Clinton-won seat as opposed to the Trump-won seat, but Democrats have an outside shot to sweep both of these seats.
12. Net at least one of these Trump-won seats in North Carolina, Ohio, or Virginia
Seats: NC-2 (George Holding), NC-9 (Open), NC-13 (Ted Budd), OH-1 (Steve Chabot), OH-12 (Troy Balderson), VA-2 (Scott Taylor), VA-5 (Open), or VA-7 (Dave Brat)
Democratic progress: Probably on track.
Democrats don’t have any single one of these seats locked up. We do still favor them in NC-9, a district that covers some Charlotte suburbs before extending south and east along the South Carolina border. But it’s the shakiest Democratic prospect of any Republican-held seat we currently list as Leans Democratic, and we may very well favor former pastor Mark Harris (R) over Marine veteran Dan McCready (D) in the end. Additionally, Republican prospects seem better in some of these seats than they were several weeks ago. Rep. Scott Taylor (R, VA-2) appears to be leading former Navy commander Elaine Luria (D) in a historically competitive but GOP-leaning seat in Hampton Roads based on two recent independent polls, one from New York Times/Siena and another from Christopher Newport University’s Wason Center for Public Policy. We’re holding at Toss-up but the race may be moving in Taylor’s direction. And we are moving Rep. Steve Chabot (R, OH-1) from Toss-up to Leans Republican; his opponent, Hamilton County Clerk of Courts Aftab Pureval (D), has been struggling with a story involving using his local political campaign account to pay for an ad dealing with his congressional race. OH-1 is also a district designed specifically to elect a Republican, just like the 11 other Republican-held seats in Ohio, but keep an eye (still) on OH-1 and also OH-12, where Balderson narrowly beat Franklin County Recorder Danny O’Connor (D) in a nationally-watched special election but won’t have the same level of outside support he enjoyed last time (but a bigger midterm electorate in that district might be more GOP-leaning than the special electorate).
With all of the targets in this category, it stands to reason that the Democrats should be able to clear the relatively low bar of winning just one seat from it. We’re just not exactly sure which one of these districts would be that one (or more).
13. Bonus category: Prime Democratic targets unaccounted for above
Seats: FL-15 (Open), GA-7 (Rob Woodall), IL-14 (Randy Hultgren), MI-8 (Mike Bishop), NM-2 (Open), SC-1 (Open), and WV-3 (Open)
Democratic progress: Probably will win at least one.
The districts we identified in earlier versions of this framework still make up the lion’s share of the top Democratic targets, but the playing field has expanded over the past several months. Some of the districts that have gotten more competitive are noted above, but there are a handful of others that didn’t neatly fit into the categories above. They are included here.
Of these, Bishop, who represents some exurban Detroit turf as well as Michigan’s state capital of Lansing, is the most clearly endangered. He’s become a poster child for the GOP’s money troubles this cycle, as former Obama administration Defense Department official Elissa Slotkin (D) has blown him away in fundraising. Yet the race is still a Toss-up, if only barely. Additionally, a typically Republican open seat in southern New Mexico, NM-2, has turned into a jump ball. We are moving WV-3, a seat where Democrats have down-ballot strength despite being one of Trump’s best districts in 2016, from Toss-up to Leans Republican this week, but while state Delegate Carol Miller (R) seems to have taken a small lead over state Sen. Richard Ojeda (D), the race remains very competitive. Becoming more competitive down the stretch are two other open seats, FL-15 and SC-1, the latter of which we are moving from Likely Republican to Leans Republican this week, and two other typically Republican but well-educated and affluent suburban district representatives, Woodall in suburban Atlanta (GA-7) and Hultgren in Greater Chicago (IL-14).
Our best guess is that the Democrats will get at least one seat from this category.
Conclusion: Setting a Democratic floor with less than three weeks to go
Let’s quickly go back over the 13 categories above and check on the Democrats’ progress. Based on our best guess, here’s the lowest number of gains Democrats are likely to get from each category:
1. Favorable open seats: +3
2. Slightly less favorable open seats: +2
3. Pennsylvania: +4
4. California incumbents: +1
5. Clinton-district incumbents: +3
6. Clinton-won, historically Republican: +1
7. Trump-district freshmen: +1
8. New York: 0
9. Trump-district Democratic DNA: 0
10. Iowa: +1
11. Kansas: +1
12. North Carolina, Ohio, and Virginia: +1
13. Bonus category: +1
That works out to a Democratic gain of 19 seats, six short of the goal of 25 assuming that the Republicans pick up at least two current Democratic seats. And, remember, Republicans may pick up more than just two Democratic-held seats, although probably not many more than that. Another way of looking at this is that we see the floor for net Democratic gains as 17, about three-quarters of the way to 23 net seats, which would flip the House to a Democratic majority.
Still, this district-by-district roadmap does show that the Democrats are knocking on the door of the House majority. And even though this list includes 66 GOP-held House seats, there are still others that Democrats might end up picking up that we may be missing. For instance, there are several districts in the Leans Republican column that didn’t even make this list, which just illustrates the potential for massive Democratic gains if things break their way on Election Night.
So: The House majority is within the grasp of Democrats. But there’s the potential for them to come up short.
In terms of our ratings, this week’s changes leave 212 seats at least leaning to the Democrats, 201 at least leaning to the Republicans, and 22 Toss-ups. Democrats need to win six of the Toss-ups to win the House, and all the other seats that currently lean to them (some of which are still very much in play), to win the House.