President Trump summarized his value to Republican candidates in a tweet this month.

“As long as I campaign and/or support Senate and House candidates (within reason),” he wrote, “they will win! I LOVE the people, & they certainly seem to like the job I’m doing. If I find the time, in between China, Iran, the Economy and much more, which I must, we will have a giant Red Wave!”

He had previously told Rush Limbaugh that he might campaign “six or seven days” a week for Republican candidates this fall, bringing that magic touch to his party nationally. After all, he wrote right before the tweet above, “Republicans have now won 8 out of 9 House Seats” in special elections since the 2016 presidential race.

He said that, too, in an interview with the Wall Street Journal on Wednesday. It’s not correct; Republicans have won eight of 10 races. Trump likes to ignore a special election in California in which the sole Republican candidate got about 1,400 votes.

Trump’s political director, Bill Stepien — once an ally of former New Jersey governor Chris Christie — was on hand for part of the Journal interview, confirming Trump’s unblemished record in his Republican primary endorsements. Although that, too, is incorrect. Trump famously endorsed Luther Strange in his bid to hold the Senate seat to which he had been appointed in Alabama. Strange went on to lose the primary to Roy Moore. (Whom Trump endorsed before Moore lost in spectacular fashion. In Senate special elections, Trump is 0 for 2 — in the same election.)

It’s what Trump said to the Journal next that really ought to worry Republicans, though.

“Asked if his campaign appearances might also mobilize Democratic voters, Mr. Trump said, ‘It may — but it energizes my people much more than it energizes them.’”
“‘I think the Democrats give up when I turn out,’ he said. ‘If you want to know the truth, I don’t think it energizes them. I think it de-energizes them. I think they give up when I turn out.’”

A lot of this that appears to be wish fulfillment. As in that tweet at the top, in which Trump asserts that voters “certainly seem to like the job I’m doing” — his approval rating is well below 50 percent — Trump’s view that his showing up at campaign events depresses Democratic turnout is far from the mark.

We can visualize how this is incorrect by comparing the number of votes cast in each district for the Democrat and the Republican in special elections after 2016 with those cast in the presidential election. In each case, the number of votes received by the Democrat in the special election was lower than the number received by Hillary Clinton, and the number received for the Republican was lower than what Trump earned in the same district.

But with two exceptions, the Democrats got a larger percentage relative to Clinton than the Republicans did relative to Trump.

On average, the Democrats got 61 percent as many votes as Clinton, while the Republicans got 46 percent of Trump’s totals. The two exceptions (marked in gray) are Georgia’s 6th District, where Karen Handel (R) got more votes relative to Trump than her opponent did relative to Clinton, and Utah’s 3rd District. Trump endorsed Handel, but he didn’t attend a rally for her.

Where he did hold a rally, the Democrats did better relative to 2016 than did the Republicans. In Alabama’s Senate race, the Democrat got 92 percent of Clinton’s vote to Moore’s half of what Trump got. In Pennsylvania’s 18th District, one of those Democratic victories, the Democrat got 80 percent of Clinton’s vote total to the Republican’s 53 percent. In the recent special election in Ohio, the Republican got about 49 percent of Trump’s vote total — and the Democrat got 62 percent of Clinton’s.

Part of this may be that Clinton was not very popular in these particular districts. After all, Trump won all of them. If we compare the special election results with the 2014 House contests in each place, though, the same pattern holds.

Several of these races didn’t have an actual contest, given the incumbency of the candidates on the ballot. On average, the Republicans in special elections got 65 percent of the tally earned by Republicans in 2014. The Democrats, on average, overperformed relative to 2014 vote totals.

In Ohio’s 12th District, where Trump went earlier this month to rally voters, the Democratic vote total was up 163 percent over the vote total in 2014. The Republican total was only two-thirds what was earned then.

But there’s a good reason for it: The race in 2014 was between a longtime incumbent and a Democratic challenger in a red district. That has been the case with nearly all of the districts where there have been special elections: With the exception of the seat in California, they were all Trump districts, many of which had unchallenged incumbents or Republicans who cruised to victory in 2014. Many were vacant because Trump had tapped those Republicans for administration jobs — a choice he made in part because they were from safe Republican districts. Presidents don’t generally appoint Republicans who hold swing districts or seats where the other party might do well. They try to ensure that those vacancies are filled by members of their own party. Yes, Trump has won eight of nine House seats in special elections, excluding California, but before those elections, the Republicans held nine of nine — easily. Before the special elections, they also had two Republican senators in Alabama.

Back to Trump’s point. There’s no sign in these special elections that his showing up to rally voters is depressing Democratic turnout; in each case, Democrats saw better turnout relative to 2016 than did the Republicans (as measured by votes cast). Sure, Trump’s primary endorsements have largely been successful, but that’s because he’s appealing to voters who do “like the job [he’s] doing.”

In federal Democrat-vs.-Republican contests in which Trump has endorsed before the fact since 2016, he’s 2 for 4 after that contest in Ohio. In districts that he generally won by a wide margin two years ago.

This is not the good news that Trump wants us to think it is.