The acceptance ratios at any given law review are so law that the process has a significant dart-throwing quality about it already. On the long view, student-editor irrationality is only a serious problem if it keeps things from being published that ought to be published. It's our job as faculty to read scholarship seriously no matter where it comes out. Once the question becomes merely one of publication location, we can only complain about poor student editorial judgment to the extent that we're using it as a proxy for our own.
I always tear out my hair when I have to explain/limitedly defend/bemoan student-run law reviews and journals to friends in other disciplines. Scientists, social scientists, humanities scholars--every other discipline has more rigorous publishing standards.
"You mean you don't have peer review?" "How many law journals are there again?! More than there are law schools?!" "How do you know what is good?!"
It is true that there are too many journals out there, too many specialty topic journals (guilty as charged for being chief articles editor of one at Bourgie Metro Law School), too few standards, no systematic, rigorous process for article selection (I recognize that name is as close to a dominant heuristic as you can get), and not enough faculty advisor engagement/help. It's often the case that second year students (Hell, 1.5Ls) with no training in social science methodology, philosophy, critical theory, economics, history, technology--name your cognate discipline, highly specalized topic, or must-be-trained methodology--will be called upon to edit a piece that is out of their league. Basically, these students are editing for readability and grammar, and doing what they can actually do--enforce The Bluebook. But as to the ability of a student editor to remark on the substantive argument, particularly if it requires special knowledge--this is very spotty.
Of course, there are benefits of student-edited law reviews. There was a great deal of discussion on this on other law blogs (I will find links later). It does make for shorter windows of publication, which means that the gestation period is shorter, and so the articles can be churned out while they're still useful and timely--say, after a major case is relased or following publication of important data. Also, more venues to publish means that the insularity and heavy-hitter problems that occur at the top reviews will be mitigated by having more forums for more topics. Speciality journals have their place too, as clearinghouses for a particular subject--I of course am a fan of employment law journals.
But yes, there are many problems with student-edited law journals. The best journals from the best schools will always turn out good articles (because of the self-selection/insularity/heavy-hitter/name-recognition effect). Less certain in quality are the articles published by journals in the lower tiers of law schools. But there's a wide swath in the middle in which top work will be published but slip under the radar, or mediocre work will be published to readers' bemusement. Take the good with the bad. Read with a grain of salt, figure out who's doing interesting and important work in whatever field you're intersted in, and read them regardless of the placement of their publication. I often rely on Larry Solum's recommendations on Legal Theory Blog (his Downloads of the Week, his comments on articles that have been posted on SSRN) to get direction in areas in which I am not very knowledgeable.
There are tons of problems with the lack of peer review, cognate discipline training, expertise, and faculty advising with student-edited law reviews. I say this as a former student editor myself. But the problems aren't going to go away by just complaining about it, and professors are as much to blame as the hapless-but-ambitious 1.5L articles editors. Unless professors take a more active role in being faculty advisors to the journals they are formally associated with (it is often a figurehead/masthead position), unless they are willing to invest the time it takes to have a system of peer reviewed scholarship (talk to your friends in other departments, it is a significant time investment to be a reader or on an advisory board of a peer-reviewed journal) and unless professors do as James Grimmelman suggests and read as carefully as they want to be read--well, the problem won't just disappear magically by itself, and the problems will persist.
Don't complain (too much) about the problem if you are complicit in it.