[Archive]

Posted By: Eric Brehm

Actually the way to really estimate the number of ungraded cards of a particular type that exist (borrowing some reasoning from software reliability engineering, which is how I used to make my living), is to track changes in the rate at which new cards are being graded over time (i.e. the first derivative of the cumulative curve). If the rate drops steeply over time, that means that the underlying population of ungraded cards is relatively small, and that those remaining are getting graded fairly rapidly. Similarly, if the rate is relatively constant over time, or declines very slowly, we can deduce that the underlying population of ungraded cards is relatively large. (Of course I am only talking here about the population of cards that eventually will be graded; there is undoubtedly some size population of cards that never will be.) I don't suppose I will be undertaking this analysis myself however, at least not in this lifetime.