That makes no sense considering that his endorsements actually won 83% of the time in 2022. That 83% is the highest of Trump endorsement wins over the years, indicating that more people are voting for those that Trump endorses. So they're clearly not moving away from Trump.
Endorsements by Donald Trump
That number is rather misleading. I'm going to copy (though edit a little) a post I made at
Trump endorsed candidates did not lose at the level that lib media claims. Here we go.
The problem with this statistic you offer is that many of those elections were not competitive at all. Anyone can score a high percentage if they just endorse people who are certain to win anyway. Furthermore, the criticism of Trump is that he would back more fringe candidates in the primaries (causing them to win those primaries), but then they'd perform badly in the general election.
Thus this 83% is a useless number. One should only be examining cases where Trump endorsed a candidate in a primary that wasn't essentially certain to go to that candidate to begin with
and the general election was one that could have plausibly gone for either the Democrat or the Republican.
So, for example, Ron Johnson and Marco Rubio should not be counted, as they would have won their primaries regardless of what Trump did; Rubio had no primary challengers, and the only one who went up against Johnson was clearly doing so as a protest rather than a serious candidate.
Similarly, we shouldn't be counting people like John Boozman, who won in Arkansas; it's a state the Republican is going to win regardless of who gets nominated by either party. And, to be fair, we shouldn't be counting losses in non-competitive states either, like Leora Levy losing in Connecticut. Finally, only cases where he give an endorsement during the primary should be counted; giving one after the primary is irrelevant for this purpose.
So once we get rid of all of those cases, here are the ones we seem to be left with:
Ted Budd (North Carolina): Won
J.D. Vance (Ohio): Won
Blake Masters (Arizona): Lost
Adam Laxalt (Nevada): Lost
Mehmet Oz (Pennsylvania): Lost (note that Oz was Trump's second choice; his first endorsement was Sean Parnell, but he got hit with a bunch of scandals that caused him to withdraw during the primary, and Trump changed his endorsement to Oz.)
As a result, we end up with a
40% success rate in Senate elections. Not very impressive ,and certainly much less than 83%. Not sure how it compares for his endorsements in Governor or House races, which I didn't calculate (and for anyone who does calculate them, remember you should remove any endorsements on sure-to-win cases in either the primary or general election, as noted above)
Whether Herschel Walker should be counted is debatable. As I just noted, he was polling really strongly before the formal Trump endorsement. On the other hand, as Hans Blaster just pointed out, it was apparently Trump's suggestion that he run to begin with. I didn't count him in the 40% above, but if one does count him, then Trump's rate drops down to 33%.