An interesting recent study looked at the question of what patterns emerge from U.S. Supreme Court citations of its own opinions, using insights from the mathematics of network theory.
It turns out that from 1885 to 2005 the U.S. Supreme Court has made 298,556 decision, including even minor ones, like denials of writs of certiorari. Of those, 47,869 have been cited in, at least, one other U.S. Supreme Court case. Roughly 28,000 of those cases have been cited in the U.S. Supreme Court only once. "Only two percent of the total number of decided cases receives fifty-six percent of citations in the network." Thus, there are only about 1,000 decisions of the U.S. Supreme Court which are important to future U.S. Supreme Court precedent.
The study goes on to make some important observations about how reliance on U.S. Supreme Court precedent has declined dramatically at the Court since Rehnquist and Scalia took office. But, the raw data above is interesting for another purpose, which is to quantify the elusive concept of important v. unimportant cases, which is at the heart of both the debate over allowing citations to unpublished opinions in appellate courts, and to the question of the cumbersomeness of the case law method. The perception that it is hard to use case law is one of the reasons that almost every country in the world except those who were formerly members of the British commonwealth have instead chosen to adopt legal systems modeled on the civil law systems of France and Germany.