This article presents the first empirical analysis of patent value by examining renewal rate data for nearly 100,000 patents. Finding that 53.7% of all patentees allow their patents to expire for failure to pay maintenance fees confirm common perceptions of patent issuance being a poor measure of innovation value. Even more interesting is the finding that patents which expire for failure to pay maintenance fees share some common identifiable characteristics. In particular, we found that renewed patents had more claims, cited more prior art, received more citations, had more related applications, had more inventors, and spent longer in prosecution. We also found that renewal rates varied both by assignee (individual versus corporation) and (foreign versus domestic) and by technology. By providing a means of systematically identifying worthless patents and their ex ante characteristics, this article compliments the author's earlier work on identifying valuable patents by comparing litigated and issued patents. Renewal rate data, however, seems a better predictor of value than litigation data as renewal rate data captures the many ways a patent may be of private value to its owner such as revenue generation via licensing or litigation, defensively, or for signaling purposes. Hence, rather than analyzing a subset of really valuable patents (those that are litigated) which may or may not be representative of all valuable patents, analysis of renewal rate data captures the population of valuable patents.