I am pushing an inquiry as to the LSAT's predictor of lawyer performance. Ninja had the best post on this so far. The problem that I'm finding is that "lawyer performance" is ill-defined, at best dependant on too many variables that the LSAT doesn't take into account.
Well in terms of the LSAT and lawyer performance there'd probably be a strong correlation in some fields and a weak one in others. For example:
Law articles/academia: There'd be a strong correlation between the people who perform well in this arena and the people who performed well on the LSAT. The people who generally excel in this arena are law school profs. who almost universally went to top schools with top grades which correlates very strongly with LSAT scores.
Law M&A type work: Probably a strong correlation. Maybe not as strong as academia but the M&A types are generally from top schools with very good academic records which correlates with LSAT scores.
Law appellate type work: Same as above.
Law general litigation: More of a mix. Some fields like antitrust, tax, you'd probably see a pretty strong correlation. Product liability, personal injury a much weaker correlation.
Oh and re speed and the practice of law. Speed is a significant issue with at large areas of legal practice. Not just from a billable time perspective but from a legal deadline perspective. Deadlines can be absolutely brutal at times and how fast you can perform tasks is a critical component of being able to practice law at least in many litigation areas of law. Probably different in transactional areas.