Skip to main content
ABC News
Are We Seeing The Start Of A Liberal Tea Party?

It’s hard not to compare the 2018 midterms to the 2010 midterms. A controversial president sits in the Oval Office. Resistance to his major policies spurs protests and grassroots activism. Special-election results portend a massive electoral wave that threatens to kneecap his ability to govern. So with the first primaries of 2018 taking place this month, will another 2010 phenomenon, the tea party, reappear? Quite possibly, only this time on the left.

A loosely defined mélange of grassroots conservative activists and hard-right political committees most prominent from 2009 to 2014, the tea party famously demanded ideological purity out of Republican candidates for elected office. In election after election during this period, tea party voters rejected moderate or establishment candidates in Republican primaries in favor of hardcore conservatives — costing the GOP more than one important race and pushing the party to the right in the process.

This Tuesday could mark the first time in 2018 that a moderate incumbent Democrat loses a primary bid to a more extreme challenger. Marie Newman, a progressive running for Illinois’s 3rd Congressional District, has argued there is no longer a place in the Democratic Party for Rep. Dan Lipinski’s anti-abortion, anti-same-sex-marriage views. Lipinski, however, has accused Newman of dismantling the party’s big tent and fomenting a “tea party of the left.”

But how accurate is that comparison really? One person’s “tea party rebellion” is another person’s justified excision of a Democrat or Republican “in name only.” To get a sense for whether Newman’s campaign against Lipinski looks more like part of a movement to pull the party leftward or simply an attempt to bring the district in line with the Democratic mainstream, we decided to look back at the original tea party — or, more accurately, since the term “tea party” is vague and sloppily applied, recent Republican primary challenges.

The following is a list of Republican incumbents in the U.S. Senate and House who have lost a primary election since 2010. To ensure that we’re capturing only Newman-style campaigns — that is, candidates who are challenging incumbents from the extreme wings of the party — we’re not including incumbents who lost to other incumbents as a result of redistricting throwing them together, nor are we counting two incumbents who were explicitly primaried from the center.1

Republican incumbents who were beaten from the right

DW-Nominate ideology scores of incumbent Republicans who lost primary challenges from the right and the scores that the challengers later earned in Congress (if they were elected), in races since 2010

District Defeated Member New Nominee
year Name Partisan Lean Name Score Name Score
’10 AK-SEN R+27 Lisa Murkowski 0.208 Joe Miller
’10 AL-05 R+27 Parker Griffith* 0.385 Mo Brooks 0.600
’10 SC-04 R+29 Bob Inglis 0.518 Trey Gowdy 0.663
’10 UT-SEN R+37 Bob Bennett 0.331 Mike Lee 0.919
’12 FL-03 R+28 Cliff Stearns 0.554 Ted Yoho 0.703
’12 IN-SEN R+12 Richard Lugar 0.304 Richard Mourdock
’12 OH-02 R+15 Jean Schmidt 0.467 Brad Wenstrup 0.577
’12 OK-01 R+36 John Sullivan 0.513 Jim Bridenstine 0.690
’14 TX-04 R+52 Ralph Hall 0.424 John Ratcliffe 0.746
’14 VA-07 R+19 Eric Cantor 0.518 David Brat 0.838
’16 VA-02 R+6 Randy Forbes 0.407 Scott Taylor 0.474
’17 AL-SEN R+29 Luther Strange 0.560 Roy Moore

* During his time in the House, Parker Griffith switched parties from Democrat to Republican. The DW-Nominate score listed is for his time as a Republican.

Partisan lean is the average difference between how the constituency voted and how the country voted overall in the two most recent presidential elections, with the more recent election weighted 75 percent and the less recent one weighted 25 percent.

DW-Nominate scores are a measurement of how liberal or conservative members of Congress are on a scale from 1 (most conservative) to -1 (most liberal). Because DW-Nominate is based on congressional voting records, it is unavailable for politicians who never served in Congress.

Sources: Greg Giroux, Ballotpedia, Daily Kos Elections, Swing State Project, Dave Leip’s Atlas of U.S. Presidential Elections, Voteview

This isn’t an exhaustive list of tea-party-style challengers — for example, it doesn’t include Republicans shooting for open seats, such as Christine O’Donnell in Delaware in 2010, or hard-liners who won GOP primaries to take on Democratic incumbents, such as Sharron Angle in Nevada (also) in 2010 — but it’s a good cross section. The reason it’s useful to look only at Republican incumbents who went down to defeat is that it allows us to use DW-Nominate, a data set that quantifies how liberal or conservative members of Congress are on a scale from 1 (most conservative) to -1 (most liberal). Since the score is based on congressional voting records, our focus on races in which incumbents lost means that both the losing candidate and, in most cases, the winning candidate have a DW-Nominate score.2

When you use DW-Nominate to try to quantify this slice of the tea party movement, what you quickly see is that there’s barely a pattern to it at all. “Tea party” candidates primaried plenty of moderate Republicans, such as Sen. Lisa Murkowski (a DW-Nominate score of .208) and party-switching former Rep. Parker Griffith (.385 in his time as a Republican). But they also toppled plenty of solid conservatives, including former Reps. Eric Cantor and Bob Inglis (both .518). Insurgents found success in moderate jurisdictions like Indiana (12 percentage points more Republican than the nation as a whole at the time of the election in question3) as well as dark-red districts like Texas’s 4th (R+52).

If, as this data suggests, the only prerequisite for being called a tea partier is to attack your Republican opponent from the right, then, sure, Newman is waging the mirror image of a tea party challenge. But that’s a fairly lazy conclusion; it lumps together all the primary challenges listed above when the data shows there are clear differences between them. For example, now-Sen. Mike Lee’s 2010 primary challenge to then-Sen. Bob Bennett in Utah made a lot of sense because of Bennett’s moderate DW-Nominate score (.331), the high contrast with Lee’s positions (he went on to earn a score of .919) and the fact that Utah was a very conservative state (37 points more Republican than the nation as a whole). By contrast, now-Rep. Brad Wenstrup’s defeat of then-Rep. Jean Schmidt in Ohio’s 2nd District in 2012 produced only a slightly more conservative representative (.577 to .467), and in a relatively moderate district to boot (R+15).

Lipinski’s DW-Nominate score is -.234, making him the 17th-most-moderate Democrat in the House. Lipinski is also a relatively strong ally of President Trump, at least as far as Democrats go. His Trump score (FiveThirtyEight’s measure of how often each member of Congress votes with the president) is 35.3 percent, 11th-highest among House Democrats. In other words, he is indeed notably more centrist than most members of his party. Going by DW-Nominate, this would place Newman’s challenge of Lipinski in the same ballpark as Joe Miller’s challenge of Murkowski in the Republican primary for Alaska’s 2010 U.S. Senate race. That’s one of the tea party cases in which it’s easier to see the ideological origins.

But the Illinois 3rd District is also only 12 points more Democratic than the nation as a whole. Just 39 Democratic-held districts are more conservative. If you subscribe to the notion that districts toward the center should have more centrist representatives, then Lipinski is still on the moderate side, but he’s not that moderate. For instance, Lipinski’s predicted Trump score is 33.7 percent, meaning that he votes with Trump almost exactly as much as we would expect given the political lean back home. According to DW-Nominate, several other Democratic House members are more moderate than he is4 and represent more liberal districts, including fellow Illinois Reps. Bill Foster and Brad Schneider. Neither of them is facing a primary challenger on Tuesday.

If you consider the district’s partisanship, then maybe Newman’s campaign is more like Richard Mourdock’s Republican primary challenge to incumbent Sen. Richard Lugar in Indiana’s 2012 U.S. Senate race: According to our partisan lean calculations, Indiana was as Republican then as the Illinois 3rd is Democratic now. Lugar was indeed moderate enough (a .304 DW-Nominate score) that a primary challenge made some sense on its own, but Mourdock infamously ended up losing that general election.

All things considered, it’s debatable whether Newman’s challenge of Lipinski is within reason, according to these ideological scores, or out of line. Lipinski is indeed a Democratic nonconformist who can’t reliably be counted on to vote against the Trump agenda — but he’s not wildly out of sync with his district either. The voters will have to decide how much heterodoxy they can tolerate on Tuesday.


  1. Kerry Bentivolio in 2014 and Tim Huelskamp in 2016 were both primaried from the center, as indicated by news reports and DW-Nominate.

  2. For this story, we’re using the first dimension of DW-Nominate.

  3. FiveThirtyEight defines a constituency’s partisan lean as the average difference between how the constituency voted and how the country voted overall in the two most recent presidential elections, with the more recent election weighted 75 percent and the less recent one weighted 25 percent.

  4. It’s fair to critique this measurement, however, on the grounds that it doesn’t weight issues by importance. Certainly, abortion and gay rights, two issues on which Lipinski is conservative, are centrally important to much of the Democratic base.

Nathaniel Rakich is a senior editor and senior elections analyst at FiveThirtyEight.