Over the past 5 years or so, there has been an increased focus on the empirical outcome of patent litigation, primarily because the number of such lawsuits have increased noticeably. Jay Kesan & Gwendolyn Ball from the University of Illinois have published a recent study titled "How Are Patent Cases Resolved? An Empirical Examination of the Adjudication and Settlement of Patent Disputes." The study builds on previous findings that detail many of the pathologies of the litigation process:

"divergent expectations" - for trials to occur, the parties must be either behaving irrationally, or there must be some uncertainty about the probability of a victory in the courtroom.

"asymmetric stakes" - if the defendant's loss does not equal the plaintiff's gain, there is little incentive for the defendant to avoid trial and no point in bargaining to a settlement.

The "free rider" problem - defendants having an eye towards settlement have no compulsion to spend money to invalidate "bad" patents, since others (i.e., competitors) will similarly benefit from the effort.

• On average, 10.7 patents per 1000 will be litigated (1%);

• The probability of a patent appearing in litigation increases with the number of later patents citing it, but decreases with the number of backward citations;

• Litigation probability increases with the number of "self citations" (patents cited by other patents owned by the same party);

• Litigation probability increases with the number of claims describing the invention;

• Patents in "crowded" fields are subject to an increased probability that disputes will arise.

• The overall decision rate in favor of patentee is 58%.

• In jury trials, 68% of the cases come out in favor of the patentee; 71% of jury trials find that the litigated patent is valid and infringed.

• In bench trials, only 51% of the cases come out in favor of the patentee. Also, 64% of bench trials find that the litigated patent is valid, but only 59% of bench trials find infringement.

The authors tracked the evolution of patent cases from 1995, 1997 and 2000 and found that most of the previous findings were largely correct. However, the study clarifies the notion that "95% of all cases settle," noting that previous studies did not necessarily consider other procedural terminations, such as summary judgment when determining this statistic. When this is taken into account, the number drops to almost 80%.