Wow. I never knew this is how patent law worked. It seems like this might be a big 'patent integrity' bottleneck. Not that the Judge is not objective (I have no way of knowing), it just seems like there is too much room for opinion and influence to seep in when it is so centralized.

As the article points out, ths centralization is a byproduct of Judge Ward, of necessity, having "built a better mousetrap" in terms of court rules for processing cases. As is customary in federal courts, when Judge Ward first became a judge, his initial case load was built by the court's other judges transferring the cases they didn't especially want to deal with; this included most of their patent cases. So Judge Ward found himself inundated with patent cases, and needed to figure out a way to keep up. Looking around for ideas, he adapted court rules that had been developed in the Northern District of California, which encompasses San Francisco and Silicon Valley. Those court rules, plus the low numbers of federal-crime cases in that rural division (by law, criminal cases get priority), meant that Judge Ward got his patent cases to trial quickly. This attracted plaintiffs wanting just such speedy "service."

At first. But as the article points out, later on the east Texas district was no longer the fastest, but it still attracted a large number of plaintiffs, probably because its procedures and its decisions tended to favor plaintiffs.

Any time that one side gets to decide who runs the system and makes (and enforces) the rules, there are natural forces of evolution that will tend to favor that side. One might also say this is the same thing that is wrong with the binding arbitration system, where something like 95% of all decisions between consumers and a business are made in favor of the business (who selects the arbitrator).

This seems similar to the "traffic jams on new roads" phenomenon, by which adding more lanes or roads in an urban area actually increases traffic rather than alleviating it as highway-construction fans always expect. Providing public resources like courts or roads is never a neutral action, so political corruption is always a possibility.

From what I have seen in life, of how everybody including myself behaves, and how I think/feel, it is not just a possibility. The question is not is there any political corruption at all, the question is how much more than zero is it? When we fail to have checks and balances at all, or that actually try to work, then we're kinda doomed.

Can we do something about the clickbait title? This is Judge Rodney Gilstrap, of the East Texast Federal District Court, in Marshall, TX.

The story is good. In particular, it highlights how a small cascade of incentives can lead to a grossly perverted and corrupt result, from the USPTO's issueing of crap patents (with full impunity), the Eastern District's procedures, Rodney's actions, and a home-grown trolling industry.

There must be a way of doing a statistical analysis of the court's judgements to determine how fair they are. Anyone have any ideas how this could be done?

Incidentally, anyone who provides an answer that is patented will automatically get an upvote from me :-)

This 2010 article [1] by (the highly-regarded) Stanford law professor Mark Lemley [2] is old, but might be helpful. Excerpt: "Notably, the Eastern District of Texas, while it has a higher than average plaintiff win rate, is not in the top five districts.

This 2015 PwC study [3] notes that from 1995 to 2014, "Interestingly, patent holder success rates for the top seven judges also tend to exceed the overall success rate, particularly in the Eastern District of Texas." The three E.D. Tex. judges listed in the chart (at page 20 of the PDF) had patent-owner win rates of 63%, 73%, and 64% respectively.

[1] http://amlawdaily.typepad.com/Lemley.pdf

[2] https://law.stanford.edu/directory/mark-a-lemley/

[3] https://www.pwc.com/us/en/forensic-services/publications/ass...

It's not a matter of win and loss rates. The vast majority of patent cases are settled without trial. Especially in EDTX, patent defendants know they cannot get a fair trial and settle to avoid the outrageous verdicts that come out of Gilstrap's courtroom.

And it costs millions to even go to court in EDTX because of the extra special discovery and expert costs and timing restraints Gilstrap imposes on defendants.

Which would tend to indicate that the defendants believe that they will lose badly, no?

Thus the win and loss rates become interesting, because then it can be seen if the number of judgements against the defendants deviate significantly from the national mean.

Actually the multi-million dollar cost of litigating in EDTX for defendants (it's cheap for plaintiffs) is usually more than the plaintiffs are asking for to settle. So defendants that expect to win big still have to settle and pay or they'll ruin their business.

Most EDTX patent defendants have under $10MM in revenue. Trolls target small companies in these suits more often than large ones.

I am afraid the biggest problem will not be the analysis, but it will be getting the right people to understand and appreciate the results.