The statisticians in the audience will certainly quibble with the title of this article. The sample size is too small for starters. But the fact is that of the first 20 patents taken to a final decision by the Patent Trial and Appeal Board (PTAB), 17 went down in flames. And as the reader will learn below, it’s actually worse than that.

Proponents of the new inter partes review (IPR) and covered business method review (CBM) often touted the belief that the patent office could better judge validity than district courts, and that the new system would improve the quality of patents. But so far improving patent quality seems to mean invalidating the incremental inventions that have been the life-blood of the patent system for many decades.

A year-and-a-half into IPR and CBM implementation, the PTAB’s first 20 final decisions have been brutal on patent owners. In the 20 final decisions, the Board has considered the patentability of 357 claims. Only 13 claims survived the process, yielding a survival rate of 3.6 percent. It gets even worse when considering motions to amend. The first 20 completed trials also included 12 motions to substitute a total of 113 additional claims. All 12 motions to amend were denied. Accounting for those failed claims yields a survival rate of 13 out of 469, or 2.8 percent. On statistics alone, a patent scrutinized by the PTAB is almost guaranteed an inglorious death.

Indeed, members of the patent community have already raised the question of whether IPR and CBM are too anti-patent, pro-challenger. For example, at the AIPLA conference last fall, Chief Judge Randall Rader of the U.S. Court of Appeals for the Federal Circuit likened the roughly 300 PTAB judges to “death squads killing property rights” that 7,000 trained patent examiners worked to grant. And the original patent examiners are one thing — they might not have had all of the pertinent information. But the PTAB has also shown throughout its existence that it shows little deference to the results of reexaminations, including contested inter partes reexaminations. Combine that with the Board’s use of the broadest reasonable claim construction, an expansive view of the obviousness doctrine, and a dim view of most evidence of secondary considerations, and you have the lopsided results we have seen so far.

But still, 13 claims have survived right? Not really. Three claims survived CBM2012-0003 filed by Liberty Mutual against Progressive because the Board determined that the primary prior art reference is not actually prior art. But the Board joined that CBM with a subsequent one filed by Liberty Mutual, and then crushed those three remaining claims. One claim survived in Microsoft Corp. v. Proxyconn, Inc. because the Board could not find one of the claim limitations anywhere in the prior art.

The best result by far is the final decision in Synopsys, Inc. v. Mentor Graphics Corp., the source for nine of the claims that have run the PTAB’s gauntlet. These claims, covering simulation and prototyping of integrated circuits, appear to have been saved by persuasive expert testimony. The Board concluded:

For claims 1-4, 6, 7, 11, 28, and 29, however, we give significant weight to the testimony of Mentor Graphics’s expert, Dr. Sarrafzadeh, who persuasively explains that Gregory does not disclose each and every element of the claims.

Later in the decision, the Board repeated this sentiment, again giving the expert’s testimony “substantial weight.” This is consistent with many decisions to institute review, which credit the expert testimony of one party or the other to support the Board’s decision. Expert testimony needs to be specific and on point, because conclusory statements are afforded no weight. But when it comes down to the key limitations, and central issues in dispute, the Board looks closely for persuasive expert testimony.

The scary initial conclusion is, however, that if the Board can find the limitations of your claim anywhere in the prior art, they will put it all together and invalidate the claim. That was certainly true in Garmin International, Inc. v. Cuozzo Speed Technologies, Inc., where the Board canceled the claims based upon combinations of three, and even four, references. The notion that the claims are obvious if the elements exists somewhere in the prior art has not been the law of the Federal Circuit. The patent community has long taken it as a given that most inventions are combinations of known elements. Federal Circuit review of these decisions is sure to be interesting.