It definitely can be difficult, and it's one of the reasons that for the official critiques, we're so adamant about our restrictions. When students follow them to the letter, they avoid a lot of the pitfalls that ultimately can make critiquing their work more time consuming.

That said, there are still plenty of cases where a student may be weaker with certain underlying concepts - for example, if they perhaps didn't focus as much on their freely rotated boxes with line extensions in their warmups, they may have gotten rusty when it comes to judging those convergences. Lesson 6 and 7's permission to use a ruler alleviates this somewhat, if the student actually leverages it to its full potential by using the ruler itself as a way to see how the line is shooting off into the distance before committing to it (in order to judge those convergences). So if they're running into trouble with that when drawing their objects' initial bounding boxes and such, that's something I'd remind them of. Keep up with the box challenge boxes in your warmups, and here's how you can use a ruler to avoid this (as long as you're putting some thought into what you're doing).

Beyond that, I really just try to focus on what the core focus of the lesson was, and attempt to look past whatever other issues there may be. I'll comment on them, but my thinking is that unless they're specifically asking questions about issues they run into with them, then it's a matter of the student simply forgetting to practice that area, and getting rusty. That's something they can handle themselves. It is of course not easy to look past all issues, because there are definitely cases where you can't be sure whether the student is fumbling because of an underlying technical issue, but they do understand the concepts of the lesson - but if it gets to that point, I'll probably just tell them that we can't really move forward with the critique until those issues are addressed. At the end of the day, a critique is about providing the student with their next steps, and sometimes those next steps involve going back and addressing core issues before we can really go any further.

At the end of the day, you do get to choose which submissions you critique, so if one stands out as being rough enough that it's going to take more than you are able to provide right now, then you don't need to critique it. That said, as you noted yourself, there's value in critiquing for your own improvement and learning. I myself have improved my understanding of these concepts immensely since I started this course, purely by explaining them over and over, and having to reflect upon what I think I know. It does take some of the wind out of the sails if we can't even be sure that the student's themselves going to read the feedback (or apply it - I've got plenty of situations where students for whatever reason may end up forgetting half of the feedback they received previously), but it is what it is.