As promised, final thoughts on the OP.
His argument might have held more water if the LSAT were the only factor in law school admissions, and more so if games were the only part of the LSAT, and more still if getting all the questions on the games were vital to getting a high enough score to get into law school. None of these are actually true. To wit:
You can get a 170 on the LSAT by getting about half the games wrong if you get everything else right. More realistically, if you're good at the other stuff, a 160-165 score is not terribly unlikely. And a 160-165 is certainly good enough for law schools in the lower end of Tier 1 or in Tier 2, especially if you have high grades, regardless of other factors.
Moreover, if you have a good "overcoming adversity" story, combined with decent grades, you can overcome a lower LSAT score and get into some GREAT schools. Yale accepts somebody in the 150's every year. By definition, a quarter of the people that law schools accept have LSAT scores below their 25th percentiles. Some of those people are reverse-splitters (low LSAT, high GPA) designed to drive up other numbers. But a lot of those people are people with good stories, like those that the OP described.
Finally, you don't need to go to a prestigious law school to become a worthwhile lawyer. If you have very particular things that you want to do and are a hard worker and, yes, get a little bit lucky, you can get a good law-related job even if you come out of Tier 3 or, yes, even Tier 4. It's (much) harder to do, but if you play your cards right, you can do it, if you want it badly enough.
So there's hope for those widows and orphans yet. They might become compassionate
Public Interest Lawyers who help the general welfare of society despite the deck being stacked against them, despite those darned logic games.
And for the rest of the applicant pool, the 20-somethings who don't really know what they want to do with their lives, haven't taken the idea of going to law school very seriously, and (frankly) often aren't very hard-working, and it puts a serious roadblock in their way. Here's something that no one knows how to do. Show us you're serious by learning it, and we'll let you in. But you have to show us you're serious, first.
It
is true, however, that over-reliance on a single metric (even the LSAT) is pernicious, and USNWR rankings make it always a tempting proposition. Criticism of standardized tests has been heavy since at least the 1980's, and I'm proud to say that Princeton Review has been one of the leaders in this criticism (under John Katzman, anyway). However, in response to this heavy criticism of the SAT, etc., the tests have changed. The SAT today is a better test than it was 30 years ago. (Antonyms? Analogies? Quantitative comparisons? No calculator?) The LSAT and MCAT are much better tests than they were 30 years ago. While you can't ever rely on any single number to measure educational success or potential, it
is legitimate to include standardized tests as
part of your process.
The proper number to consider, in other words, is not how well the LSAT correlates with first year law school performance. The proper numbers to compare are correlations between GPA and first year law school performance vs. correlations between GPA + LSAT and first year law school performance. That is, does the LSAT together with GPA predict first year law school performance better than GPA alone does? If so, then the LSAT provides additional useful information that you can't get from GPA alone. And at that point, it's a useful test to have. Could it be better? Surely. But using a standardized test -- and one that you can and should study for -- as an admissions component (component only, not the sole criterion) for law school seems reasonable.