What Patent Litigants Need Most and Why 2020 Will Provide Crucial Data
What Should I Be Looking for in a Court?
Submitted for your approval--patent litigants (with non-frivolous cases) should seek one thing above all else from the courts: substantive findings that materially decrease litigation uncertainty. The costs of litigation are sunk, and we all know uncertainty is an anathema to most businesses. Substantive findings from the courts decrease both; they provide fixed data-points that either resolve cases outright or provide objective criteria necessary to elicit settlement.
Statistically speaking, a given case is unlikely to be resolved once-and-for-all by a final, contested ruling of the court. Thus, when analyzing which courts best minimize the disruption (cost and uncertainty) to businesses, I suggest looking more to the prevalence of substantive findings, not just win/loss versus settlement numbers.
My hypothesis is that some courts are better at driving cases toward substantive findings than others. As discussed below, this is not an easy hypothesis to test. But observational data can be helpful--and it's the best we have.
How Do I Find What I'm Looking for in a Court?
A number of factors control how court cases progress. Some examples include case-load per judge, local practice, judicial attention, judicial expertise with patent cases, judges' ability/willingness to adjust discovery burdens based upon the nature/value of the case, and the amount of live interaction permitted between judges and litigants. Unfortunately these criteria are subjective and/or difficult to track.
That said, among the death-and-taxes certainties of life, cases will inevitably resolve, as time goes on, either via trial or by one side backing down prematurely to avoid mounting litigation expense. Sometimes courts provide assistance (in the form of a substantive finding) in doing so; sometimes not. Thus, we can look at aggregate data (the collective universe of resolved cases) and determine what percentage are resolved with the help of substantive findings by the court.
Various federal court databases track case outcomes (sometimes more precisely as "accusation" outcomes). But they often treat settlement (e.g., via consent judgment or stipulated dismissal) and wins as mutually exclusive. That forced separation ignores that substantive findings are often an essential part of the settlement process. Additionally, many databases treat default judgments as wins--even though default judgments are based upon uncontested findings and thus not very illuminating.
One of the databases (Lex Machina) tracks what it calls "Patent Findings." These include affirmative findings (such as infringement or invalidity)--but exclude the large volume of denials of requests for findings (e.g., denial of a motion for summary judgment on infringement). The "Patent Findings" metric is helpful, even if it is imperfect.
The imperfection arises from the fact that the term "Finding" is somewhat subjective. For example, I am familiar with cases in which I would consider a patent finding to have been issued, even though Lex Machina tags it as a purely procedural resolution. (And I chose not to personally check all 17,000+ cases in the data set.) But, for aggregate purposes, Lex Machina's metric will suffice, albeit with some refinements.
I filtered Lex Machina's results for cases that have been resolved, have a "Patent Finding," and were not rendered as part of a default judgment, consolidation, or transfer (the "Substantive Cases"). I then calculated the percentage of Substantive Cases that included a "Patent Finding." I did so for three districts over the time period from 2012-2019:
- Illinois (Northern District or "ND")
- Texas (Eastern District or "ED")
- Delaware
Results
The plot below shows the percentage of Substantive Cases, for each year in which they were filed, that included a "Patent Finding." It also depicts (in red) the period during which the Supreme Court took up the TC Heartland case, which upended venue in Texas and Delaware: between late 2016 (where the circumstances in which certiorari was granted indicated what was to come) and the actual ruling in May 2017.
Conclusions
First, the overall trend appears to be downward. But the median case pendency in these jurisdictions is between 1.8 and 2.5 years. Thus the sample size from 2017-2019 is lacking, making it unclear how much inertia the trend has.
Second, Illinois (ND) is in the middle of a downward trend. Anecdotally, I expect this to level off. The numbers in 2016 and 2017 are skewed by one extremely high-volume patent entity (Sportbrain Holdings), which filed 139 of the 402 Analyzed Cases in those years. The short pendency (median of 150 days) and absence of any "Patent Findings" indicate to me that the Sportbrain subset constitutes an unusual slew of nuisance value cases, which artificially drove down the Illinois numbers. Time will tell.
Third, I predict the relative prevalence of "Patent Findings" between Delaware and Texas (ED) to continue to reverse course in the coming years. The TC Heartland decision diverted a large number of case filings out of Texas and into Delaware. I predict that the change in case loads will lead to a greater prevalence of "Patent Findings" in Texas and fewer in Delaware, respectively.
Fourth, 2020 data will be crucial in analyzing these trends, as a large portion of the cases filed after TC Heartland (in 2017 and 2018) will conclude.