For the second time in 2018, voters will go to the polls to replace a Republican congressman who resigned. But Tuesday’s special election in Arizona’s 8th Congressional District — which Rep. Trent Franks vacated in December after reportedly offering one of his staffers $5 million to carry his child — is expected to have a very different outcome from last month’s Democratic upset in Pennsylvania’s 18th District. Prominent nonpartisan handicappers all rate the race between Republican Debbie Lesko and Democrat Hiral Tipirneni as “Likely Republican” — but in terms of reading the tea leaves for November, the winning margin will matter more than the outcome.
1. The district
Arcing through the suburbs northwest of Phoenix, including Surprise, Peoria and Litchfield Park, Arizona’s 8th District is usually a Republican stronghold. According to Daily Kos Elections, President Trump carried it 58 to 37 percent in 2016; Mitt Romney won it 62 to 37 percent in 2012. According to FiveThirtyEight’s preferred method for calculating a district’s default partisanship — we call it a district’s “partisan lean”1 — the 8th District is 25 percentage points more Republican-leaning than the nation as a whole.
That’s comparable to Pennsylvania’s 18th District and most other jurisdictions that have hosted special congressional elections since Trump took office. But, as regular readers of this site know, that hasn’t stopped Democrats from being competitive in — and even winning — many of those races.
|Year||Date||Seat||Partisan Lean||Vote Margin||Dem. Swing|
|2017||April 4||California 34th*||D+69||D+87||18|
|April 11||Kansas 4th||R+29||R+6||23|
|May 25||Montana At-Large||R+21||R+6||16|
|June 20||Georgia 6th||R+9||R+4||6|
|June 20||South Carolina 5th||R+19||R+3||16|
|Nov. 7||Utah 3rd||R+35||R+32||3|
|Dec. 12||Alabama U.S. Senate||R+29||D+2||31|
|2018||March 13||Pennsylvania 18th||R+21||D+0.3||22|
|April 24||Arizona 8th||R+25||?||?|
The margins in federal special elections since Trump’s inauguration have been an average of 17 points more favorable to Democrats than pure partisanship would lead us to expect. Another Democratic overperformance is likely on Tuesday — practically no one thinks Lesko is going to defeat Tipirneni by 25 points or more. But in order for Tipirneni to win outright, she’d not only need to do better than the Democrats’ average overperformance so far, but also better than their overperformance in Pennsylvania.
That’s unlikely. Arizona’s 8th District is especially infertile ground for Democrats. Unlike Pennsylvania’s 18th District, it has no historical tradition of Democratic support. (In fact, quite the contrary — the area is known for its devotion to controversial former Maricopa County Sheriff Joe Arpaio.) Whereas southwestern Pennsylvania had elected a Democrat to Congress as recently as 2000, the West Valley (as this part of Arizona is known) hasn’t done so since 1980.2 And while registered Democrats represent a plurality of voters in Pennsylvania’s 18th District, registered Republicans outnumber registered Democrats 41 to 24 percent among active voters in Arizona’s 8th.
The district’s demographics might be especially bad for Democrats in a low-turnout special election. Arizona Democrats lean on non-white voters, especially Latinos, for a large share of their support, but people of color have historically been among the least likely people to vote in off-year elections. That makes districts like Arizona’s 8th (which is 30 percent non-white, including 19 percent Latino) more vulnerable to a Democratic drop-off in special and midterm elections.
Even worse for Democrats, Arizona’s 8th District is also home to lots of older voters — a demographic that’s both heavily Republican and dutiful about casting ballots even in off-year elections. The West Valley is considered one of the best places in the country to retire, and more than 24 percent of 8th District residents are 62 years old or older. Indeed, predominantly white retirement communities such as Sun City are a distinguishing feature of this district.
2. The polling
The polling average so far (Lesko 47 percent, Tipirneni 44 percent) suggests a close race, but those numbers come with a mountain of question marks. Two of the three public polls were conducted by partisan firms: Perhaps predictably, the Republican one showed a 10-point Lesko lead, while the Democratic one showed a tied race. That combination implies that the true result is somewhere in the middle. On the other hand, the only nonpartisan poll agreed with the Democratic assessment that the race is statistically tied. However, for this poll to be correct, left-leaning demographics would have to constitute an unusually high share of the electorate; the poll’s sample was far younger and better educated than district averages. Turnout could swing that way, but Democrats shouldn’t hold their breath.
|April 14-16||Lake Research Partners (D)||44%||44%|
|April 12-15||Emerson College||45||46|
|April 10||OH Predictive Insights (R)||53||43|
In addition to the public polls for which we have precise numbers, internal Republican polls reportedly put Lesko ahead by “high single digits or double digits.” Considering that House special election polls come with very wide margins of error, any of the outcomes shown by any of these polls — from a narrow Tipirneni win to a Lesko rout — is plausible.
3. The players
Unlike in Pennsylvania, the candidates in Arizona’s 8th District are about evenly matched in terms of quality and fundraising. Lesko represented much of Arizona’s 8th in the Arizona House of Representatives from 2009 to 2015 and in the state Senate from 2015 until earlier this year, earning the gratitude of her retiree-heavy constituents when she pushed a bill to allow people to drive golf carts alongside regular traffic. Tipirneni is a political neophyte, but she has a good story to tell as an Indian immigrant who became a doctor and cancer researcher.
Neither candidate is an ideological outlier; both champion their party’s usual positions on issues like health care, taxes and guns — although embracing the usual Democratic positions may make Tipirneni an outlier in such a conservative district, which would work to Lesko’s advantage. Both have also weathered mini-scandals: In January, Lesko transferred $50,000 from her state campaign account to a federal PAC working to elect her to Congress, which may violate rules forbidding candidates from coordinating with PACs. Tipirneni has had to deny that a malpractice lawsuit she settled in 2006 was the reason she ended her career as an ER doctor just months later.
On the rubber-chicken circuit, the Democrat has raised the most money, but Republican outside groups are negating that advantage. As of early April, Tipirneni had raised a total of $741,000 and spent $616,000 of it; Lesko had raised $564,000 and spent $511,000. But, starting shortly after the party’s loss in Pennsylvania, GOP outside groups began directing a steady stream of campaign spending to Arizona’s 8th District, adding up to more than $1 million. That may seem like a lot for a supposedly uncompetitive race, but it’s a fraction of the $10.7 million in soft money Republicans anxiously poured into Pennsylvania. An ounce of prevention, they likely reason, is worth a pound of cure. For their part, liberal groups are getting involved in the race only indirectly, if at all.
4. The actual votes
Tuesday may be Election Day, but even as you read this preview, most — some estimates say as many as 80 percent — of the votes in the contest have already been cast. Arizona has a generous early-voting period (and it is well used — we estimate that more than 60 percent of votes in the 2016 election were cast early3), and no excuse is required to cast an absentee ballot (in fact, Arizona maintains a Permanent Early Voting List, and enrollees are automatically mailed a ballot in advance of each election). As of Monday morning, about 150,000 people had already voted, according to the Arizona secretary of state.
Although those ballots haven’t been counted yet, the data on who submitted them — their age, gender, party, etc. — is public information. Unfortunately for Tipirneni, 49 percent of those who have voted so far are registered Republicans; only 28 percent are registered Democrats. Nearly 60 percent of the electorate so far is age 65 or older.
Normally, we put very little stock in early-voting-based predictions. There are too many places for your train of logic to go off the rails: Registered party members may not be voting for their party’s candidate; we have no idea how registered independents are voting; the people who vote on Election Day might behave completely differently from early voters and drown them out; and so on. However … a 21-point party gap is a huge difference, and there probably aren’t going to be that many more votes cast on Election Day itself. By no means do the early-voting numbers suggest that Tipirneni is doomed, but they’re definitely good for Lesko.
The surplus of early votes will also be a boon to those who follow the results live on Tuesday night. Expect the early vote to be released about an hour after polls close at 7 p.m. local time (10 p.m. Eastern Time); since those initial results are expected to constitute the vast majority of ballots, we might know who the winner is right then and there. If not, the truly dedicated can stay up late to watch the Election Day results trickle in on the secretary of state website. Since only one county (Maricopa) lies within the boundaries of the 8th District, we can’t provide our usual county-by-county benchmarks to watch for, but analyst Ryan Matsumoto has put together precinct benchmarks that you can use to get a sense for whether Lesko or Tipirneni is hitting the vote goals she needs.
5. The bottom line
Even if, as expected, Democrats do lose this race, it could still be good news for their prospects in November. The key question to ask is this: How much did Tipirneni outperform partisan-lean-based expectations by? Even if Lesko ends up winning Arizona’s 8th District comfortably, beware of a media narrative that portrays it as good news for Republicans, or at least a reprieve from the recent run of close elections. It makes no sense to compare elections based on the moving target of how close they are; you need a fixed reference point like improvement over base partisanship. Based on FiveThirtyEight partisan lean, an 8-point Lesko win would be exactly in line4 with past special-election results that have pointed to a Democratic wave. If, however, Lesko wins by a margin in the teens — thus holding Democratic overperformance to 12 points or fewer — then perhaps special-election results are beginning to come into agreement with the tightening generic ballot polls; maybe Democrats’ position truly is eroding. Or maybe not — it’s just one data point.
The margin may also influence both parties’ thinking about November’s big U.S. Senate race in Arizona, which could be decisive in the battle for control of the upper chamber.5 Again, this special election will be just one data point, but it will be the first one we have for the Grand Canyon State, which is important because Democratic special-election overperformance has varied widely from state to state. Arizona as a whole has a FiveThirtyEight partisan lean of R+7.5, so theoretically, any Tipirneni overperformance of 8 points or more would bode well for Democrats’ chances to flip Sen. Jeff Flake’s open seat.