Let's add some stats to this discussion:
- 348 posts match
- Of those, 218 are questions
- 31 are closed (9 of which are dupes), 187 are open
- 35 have no answer, 116 have at least one answer, but no accepted answer, and 67 have an accepted answer
- 68 have a net score of 0, 102 have a net positive score, meaning 50 have a net negative score
- 7 of the closed questions have a positive score, 9 have a 0 score, so 15 (almost half) have a negative score.
Some things to also realize is that there are a number of questions that, according to our relatively new criteria, should be closed (my guess would be more of the negative score questions, but not necessarily all of them), and some of these questions don't even belong in this discussion. Also, this search might not catch every Minecraft Crashes question. With that said, the numbers say that this class of question isn't a big problem, even if it's perceived to be.
As to why It's perceived as a big problem, well personally, I think it's because most of the questions are untestable (because it's nearly impossible to reproduce the problem), and therefore the proposed solution lacks verifiability. This means we're basically guessing, and we don't like to guess. It's because of this, the common practice of not getting all of the requested information out of the gate, and the incredible similarity between all the questions despite not really being able to close them as dupe that really frustrates us.
Let's not forget that we still get lots of good Minecraft questions. In particular, I've seen a pleasing surge in good minecraft-commands questions.