When businesswoman Lynda Bennett lost the GOP runoff for North Carolina’s 11th Congressional District on June 23, it ended the political equivalent of Joe DiMaggio’s hitting streak: It was the first time this year that a candidate endorsed by President Trump did not win his or her primary.
As long as Trump has been in office, he has been the central issue in Republican primaries: Candidates talk up their own support for the president and commonly attack their opponents by saying they were once a “never Trumper” or won’t support Trump’s agenda. Unsurprisingly, Trump’s endorsement has quickly become the most coveted prize in Republican primaries. In 2018, we found that Trump-supported candidates went 15-for-17 in GOP primaries for the U.S. Senate, U.S. House and governorships in which no incumbent ran — a higher win rate than the other 11 endorsers we tracked.
In 2020, FiveThirtyEight has once again been tracking endorsements in every Senate, House and governor primary so far (i.e., those decided by Aug. 25). Trump has endorsed 111 candidates in these primaries, and 109 of them won the Republican nomination or advanced to the general election.1
That 98 percent win rate is nothing short of extraordinary — and Trump has not been shy about bragging about it on Twitter. But a closer look at the circumstances of those endorsements reveals that Trump’s win-loss record isn’t solely due to his influence on Republican voters. It’s also the product of a strategy of endorsing candidates who are already in a good position to win — perhaps with the explicit goal of inflating his top-line win rate, and therefore his reputation.
Most obviously, Trump endorsed 21 candidates who were unopposed in their primaries, including 20 incumbents and Army veteran Sean Parnell, a Trump favorite who is running in a swing district in Pennsylvania. Among his 90 endorsements in contested races were also 67 incumbents, who almost always win renomination. Rep. Scott Tipton was the only incumbent Trump endorsed this year who lost his primary.2
|Type of Endorsee||Endorsements||Wins||Win rate|
|Challengers to incumbents||0||0||—|
That leaves only 23 non-incumbent candidates whom Trump has endorsed in contested primaries. And, to Trump’s credit, he still has an excellent record in these races: 22 of his endorsees advanced to November. (The aforementioned Bennett was the only exception.) And some of these were truly competitive races where Trump’s involvement probably did make the difference. For example, former Auburn University football coach Tommy Tuberville led former U.S. Attorney General Jeff Sessions by less than 2 percentage points in the first round of Alabama’s Senate primary — but then Trump endorsed Tuberville, and Tuberville beat Sessions in the runoff by 21 points. Likewise, Texas congressional candidates Ronny Jackson and Tony Gonzales once looked like at least mild underdogs, with their opponents securing valuable local endorsements — but then Trump endorsed them and they eked out wins.
|Tommy Tuberville||AL Sen.||Won||+21|
|Jason Lewis||MN Sen.||Won||+71|
|Bill Hagerty||TN Sen.||Won||+11|
|Beth Van Duyne||TX-24||Won||+44|
|Cynthia Lummis||WY Sen.||Won||+47|
On the other hand, some of these primaries were contested in name only, such as the one for Senate in Minnesota (Trump endorsee and former Rep. Jason Lewis defeated his closest challenger 78 percent to 7 percent) and Georgia’s 6th Congressional District (former Rep. Karen Handel beat her closest competition 74 percent to 15 percent). And in some races, Trump endorsed extremely late in the race — sometimes even the day before the primary — which could be more about trying to run up his win rate than making a major impact on the race.
It’s easy for Trump to make last-minute endorsements given his preferred method for bestowing his support: tweeting. Twitter allows Trump to make many more endorsements,3 including last-minute endorsements that don’t require much (if any) planning. In some cases, the planning is so minimal that the endorsement comes as a surprise to the candidates or the party. So it seems like Trump is endorsing at least some of these candidates on a whim.
In other cases, the rollout of endorsements was much more organized. Take Pennsylvania. On May 22, a little more than a week before the Pennsylvania primary, Trump sent a slew of tweets endorsing candidates and linking to each candidate’s page on WinRed, the Republican fundraising platform hoping to counter the Democrats’ ActBlue. All told, Trump endorsed candidates in 10 out of 18 races in Pennsylvania; nine of those candidates were running unopposed.
That said, the same level of organization was not evident in every primary: Trump let the early-August primaries in Wisconsin, Kansas and Michigan pass unmentioned on Twitter. Indeed, there have been dozens of Republican primaries in which Trump has not endorsed — primaries of all ilks, from uncontested Senate races in Michigan and Georgia to bitter intraparty battles like those in Tennessee’s 1st Congressional District or New Mexico’s 2nd. There was also at least one race, for Senate in Kansas, where Trump was urged to weigh in but didn’t, possibly because he didn’t want to come down on the wrong side.
Even though Trump’s endorsements come in fits and starts, it seems to work for him: His high win rate provides evidence both that he is picking his battles smartly and that he has at least some pull with GOP primary voters. Even if you look only at contested, incumbent-less primaries in which Trump endorsed well before Election Day, he’d still have a nearly perfect record — since only two Trump-endorsed candidates have lost primaries so far this year, his record is heavily weighted to the win column no matter how you cut the data.
But Trump’s endorsement may not hold the same power over the electorate as a whole. In a recent academic paper looking at 2018 congressional races, political scientists Andrew Ballard, Hans Hassell and Michael Heseltine found that, while Trump’s public endorsement provided a financial windfall for Republican Senate candidates, it also boosted fundraising for their Democratic opponent, so it was ultimately detrimental to the endorsees’ chances. Given this finding, Trump should be cautious about assuming his golden win rate can extend to the general election.