Saturday’s special election in Texas’s 27th Congressional District was unusual in several ways. Nine candidates fought it out for the seat, rather than just one Democrat and one Republican; it was the rare election not to take place on a Tuesday; and unlike other special elections this cycle, it flew way under the national radar. But perhaps most interestingly, the results, while not too surprising for a district as red as this one, were out of character for special elections so far in 2017 and 2018: Democrats did not significantly outperform expectations.
The crowded field of candidates was competing to represent the 27th District for the final six months of the current term, after Rep. Blake Farenthold resigned amid a sexual harassment scandal. In preliminary returns Saturday night, Republican Michael Cloud appeared to have won outright, with 55 percent of the vote.1 Democrat Eric Holguin, who will also be Cloud’s opponent in November’s regularly scheduled election in this district, was Cloud’s closest competition, with 32 percent of the vote; no other candidate received more than 5 percent.
But — at least for those of us who don’t live on Texas’s Gulf Coast — who won the special election is less important than the Republicans’ margin of victory. We’ve been tracking congressional special elections since they began this cycle, back in April 2017, paying special attention to how the results have differed from how a generic Republican candidate “should” perform. So far, we’ve seen Democrats consistently doing better than the partisanship of their districts would suggest. But this time around was a little different.
In Texas’s 27th District, the combined vote share of the three Democratic candidates was 39 percent; for the three Republicans, it was 60 percent. That’s a 21-point combined Republican margin of victory in a district that’s 26 percentage points more Republican-leaning than the nation as a whole, according to FiveThirtyEight’s partisan lean metric.2 That means that the 27th District represented a 5-point margin swing toward Democrats — far less than the average Democratic overperformance in federal special elections going into Saturday, 17 points.
Democratic overperformance in federal special elections
Special elections this cycle, by the seat’s partisan lean and final vote margin
| Year | Date | Seat | Partisan Lean | Vote margin | Dem. Swing |
|---|---|---|---|---|---|
| 2018 | June 30 | Texas 27th† | R+26 | R+21* | +5 |
| April 24 | Arizona 8th | R+25 | R+5 | +20 | |
| March 13 | Pennsylvania 18th | R+21 | D+0.3 | +22 | |
| 2017 | Dec. 12 | Alabama U.S. Senate | R+29 | D+2 | +31 |
| Nov. 7 | Utah 3rd | R+35 | R+32 | +3 | |
| June 20 | South Carolina 5th | R+19 | R+3 | +16 | |
| June 20 | Georgia 6th | R+9 | R+4 | +6 | |
| May 25 | Montana At-Large | R+21 | R+6 | +16 | |
| April 11 | Kansas 4th | R+29 | R+6 | +23 | |
| April 4 | California 34th† | D+69 | D+87 | +18 |
We’re deep enough into the election cycle that you could be forgiven for no longer caring too much about special elections; by this point in the cycle, polls have grown into their predictive power, and the bloodsport of the regularly scheduled November election is well underway. But it’s still true that special elections — and, specifically, Democratic or Republican overperformance in them — have historically tracked closely with the eventual national House popular vote in November.
Special elections have tended to predict midterm outcomes
Swing in average special congressional election from weighted presidential lean vs. the national House vote margin in the next midterm since 1994
| Cycle | Average Swing | National House Margin | Difference |
|---|---|---|---|
| 2014 | R+4 | R+6 | 2 |
| 2010 | R+6 | R+7 | 1 |
| 2006 | D+15 | D+8 | 7 |
| 2002 | 0 | R+5 | 5 |
| 1998 | D+1 | R+1 | 2 |
| 1994 | R+9 | R+7 | 2 |
| Average | 3 |
So what does it mean that Democratic performance was underwhelming (relatively speaking — they still overperformed by 5 points) in Texas’s 27th District? Well, it’s important not to draw conclusions from just one special election. A party’s overperformance in a single race has much less predictive power than the aggregate of all special election overperformances, which remains a very favorable indicator for Democrats: Including the margin swing in Texas’s 27th, the average swing toward Democrats in this cycle’s special elections is now “only” 16 points. That said, it’s difficult to ignore that the 5-point swing toward Democrats in Texas is quite close to Democrats’ current 7-point lead in generic congressional ballot polling. Generic-ballot polling has been telling us almost all cycle long to expect a less dramatic Democratic wave than the one hinted at by special election results. Maybe Saturday’s election in Texas is finally a sign that the two indicators are coming into agreement.
But it could just as easily not be. Like I said, Saturday’s election was unusual. The odd timing and the sleepy campaign added up to an extremely low-turnout election. Only 36,268 ballots were cast, compared with 230,580 in the 27th District general election in 2016 and 131,047 in 2014. This was the third election in the 27th District in four months — the regularly scheduled primary March 6, a primary runoff May 22 and the special election on Saturday — so voter fatigue may have set in, which could mean that the special election is unrepresentative of the mood of the electorate. Basically, this wasn’t a great data point for Democrats, but it probably shouldn’t change your priors much. This cycle’s special elections still imply a Democratic wave of historic proportions, while the generic ballot polling still points to a close race for House control. Lots of uncertainty remains, and you should be prepared for either outcome.