Everyone is capable of making poor decisions and straight up logic errors. Senior devs sometimes more-so because we tend to get entrenched in a particular issue solo for longer periods.
My thoughts are also on actual senior engineers, not 2 years of experience etc.
A common fallacy is to assume authors of incomprehensible code will somehow be able to express themselves lucidly and clearly in comments.
Often, the "right abstraction" for a problem is not call/return based. So you get to choose between having the right abstraction, and code that is simple when viewed with that abstraction in mind, but "weird". Or alternatively choose the wrong but better-supported abstraction and have code that is needlessly complex but "straightforward".
I hate to agree with you, but "picking the right abstraction for the problem" elegantly expresses what I've been trying to tell people for the last few years now, but far less succinctly (I usually ramble on about "underlying data models" and "what this actually is in the real world, not just how we view it in our application")
The reason I hate to agree with you is that I just became very disappointed that this is a difficult skill to master. "Just use the right model and the code is easy, duh. Why aren't you doing this?" Well. Now I just feel like a jerk. I thought they just gave me the "senior" title because I was old.
It's unfortunate that one of the most important skills in the industry is so intangible and difficult to quantify. Even more difficult to teach.
You nailed it really, senior engineers code is the most simple looking as generally they've picked the right abstraction for the problem.