As an applicant, you should consider it a good thing. You work in one of the few fields where it is possible to easily and quickly test whether someone is incompetent. If it weren’t for the tests you would be in direct competition with legions of bullshitters.
Consider if you're trying to see if someone plays basketball well. But instead of asking them to show basketball skills, you ask them to jump hurdles and juggle. Sure, the people who can do those well probably have good agility and coordination that's necessary for basketball, but you're probably filtering out a lot of great basketball players.
You also get pretty high resolution data about how your applicant performed (like their intermediate submissions, editing history etc) and how their solution performs compared to other applicants.
So all of that comparative info on a less directly relevant test can actually be more useful than n=2 or 3 on a direct proxy of your actual job.
Also I disagree with the premise that these exercises don’t carry over. The ability to puzzle through an algorithm you haven’t seen before in under an hour is directly useful and a common task in any domain that is algorithmically complex and not just gluing CRUD frameworks together (and even that gets complex once the scale is big enough)
You're arguing that standardized testing is a good indicator of candidate outcome. If this was the case, Universities would not be ditching the SATs en masse.
When you test for programming puzzles, you're just filtering out people who have grinded leetcode, just like how the SATs filtered out people who took SAT prep classes. If that's your goal, great. However, we shouldn't pretend programming puzzles are a good way to filter for great engineers. I can guarantee you that leetcode is filtering out a lot of great engineers.
My goal in running a team is performance rather than political signalling, so I prefer to use the best performance metrics available to me.