Same thing happened to me when I was independently querying the NTSB database a few nights ago. Twice I read fatality reports involving pilots I knew were still around (e.g., KS)! Turns out there were two reports for the midairs, one for each aircraft. The reports, often written by investigators who know little about soaring, are imperfect albeit sobering. And they're searchable by various criteria. And downloadable.

Without mentioning types, I was struck by how often certain gliders seemed to show up. Maybe 9B could do some analysis leading to a "safest gliders to fly" or "safest gliders for lower-time pilots" or "safest gliders to crash" report. Seriously. Many of the data needed are available, not all from the same source, of course: accidents, registrations, hours and ratings of the affected pilots (usually), hours in type, context (contest participation is often mentioned when applicable), contest experience by pilot and by type, pilot ranking at the time, age, etc. I wish the digitized versions went back further (1982 IIRC) but there is still a lot of valuable information.

Chip Bearden