Statistics on how often BPAI finds swear-behind declaration sufficient

Yesterday I mentioned a new post on PatentlyO discussing some pro's and cons of swear-behind practice. More relevant stuff at PatentlyO: a new post (here) with statistics showing how often BPAI finds swear-behind declaration sufficient. You should read the entire post, but here's the highlight:

BPAI rejected the attempt to antedate in 77% (56) of these decisions, accepted the attempt to antedate in 22% (16), and partially accepted and partially rejected the attempt to antedate in 1% (1).

With such long odds, should any practitioner even try a swear-behind?

I'll agree this looks discouraging at first glance. But I read quite a few decisions while doing research for my blog, and I found that a lot of swear-behinds were not just insufficient, they came nowhere close.

When I say "nowhere close", I'm talking about declarations that said "here's an invention disclosure form showing we invented before the reference date and here's a few emails between inventors and attorneys showing diligence". If you read the first post in my swear-behind series (here), you'll see that what's really required is an explanation of how the facts show reduction to practice and diligence, not just conclusory statements.

So until I know how many of the 77% were declarations were full of conclusory statements, I won't interpret this statistic as "it's impossible to get a swear-behind accepted at the Board." Instead, I'll say "don't take your case to the Board unless your swear-behind meets the requirements set out by the case law."