I think that's an excellent practice - not as a way of gaming the system, but for maintaining useful documentation about what happened and why.

A bug fix without an associated ticket is missing context: who spotted the bug? When? What were the steps to reproduce?

Even a ticket with no content is valuable - it gives me somewhere I can post additional comments and screenshots later, or link to other related tickets.

And an area that QA should be aware of, during testing. and maybe to update automated tests.

Uh yeah… that’s how it’s supposed to happen. Every code change should have a ticket/issue, a bit branch, and a pull request attached - no matter how small.

What’s not supposed to happen is have those analysed to be used as some dumb proxy for productivity/output.

Eh, are you allowed to just change code without opening a ticket? Every job I worked at every commit had to be against a ticket.

I do that (1 commit = 1 bug). Mostly to have a nice nested structure of bugs at the end of the quarter to help write my performance reviews. The bugs are linked in a parent-child relationship going from high-level 'user journey' bugs, to mid-level feature planning bugs, to individual commit bugs.