Jeff addressed this in the blog post Optimizing For Pearls, Not Sand.
Consider the question Does torture work well as an interrogation technique? on Skeptics. Is this a brilliant question? Is it even an original question? No, it’s just a mundane grain of sand question that could have been asked by anyone at any time. What makes it remarkable is the incredible answer on that question by Larian LeQuella with over 100 upvotes.
Sand, meet pearl.
(Emphasis and link original.)
It doesn't take much skill to have a problem. It does take some skill to communicate effectively about a problem.
It takes some skill to solve a problem. It takes quite a lot of skill to solve a problem well, write the solution out, describe how the solution works and explain why the solution is better than other solutions.
The weighting seems about right to me.
EDIT: adding a completely unscientific diagram to illustrate and support my comment. It's based on my own observations, not any real data.
[--Worthless junk---------------------------------------------Pearl of Lao Tzu--]
(---Typical question quality---)
(---Typical answer quality---)
Note also that the system awards a badge, Reversal, that honors great answers to terrible questions. Producing an answer like that is an achievement, after all. There's no such recognition for great questions that get only terrible answers. Those are just reminders that the system has failed.
* Never heard of the Pearl of Lao Tzu? Neither had I, until I was writing this answer. See Wikipedia.