I don't have platform specific numbers with me. I know from looking at charts of oxygen analysis being done on some test jet aircraft (OBOGS) that the numbers on the ground are usually max performance (95-100%). For some aircraft, there is an OBOGS source change that happens after Weight Off Wheels that immediately shows a drastic decrease in O2% by vol, then increases back up to about 60-80%. The F-16 OBOGS O2 concentration to the face is very similar to the F-15 LOX, except it has a steeper increase at low atitudes.
A lot of pilots let their mask hang until they absolutely have to put it on, but then they are getting that mass influx of oxygen to the lungs. When it is absorbed by the blood stream, and the pilot is still breathing a very high percentage of oxygen, the alveoli start to collapse since there is not a high enough concentration of Nitrogen to retain the shape (or so the theory goes).
Luckily, with all of the OBOGS issues, they are implementing sensors in a lot of places in a lot of aircraft and this theory will be proven or disproven soon.
As for the T-6 IP hypoxia issues, I would have to look at the debrief from the flights/ground events. I don't know if I buy off on lack of SA because a lot of the bad cases of hypoxia end up being "Oh, I remember being cleared to land and I didn't really feel good. I remember taxiing off the runway, but don't remember actually landing the plane." If that's the case, then no amount of (lost) SA is going to help you keep 69 seconds of flight time that would have otherwise been lost due to oxygen deprivation. Personally, I'd look for 1 or 2 aircraft that have the most amount of hypoxic events associated with them and start digging deeper. In addition, it would be a good idea to do hose leak check, mask fit check, mask leak check, concentrator and regulator replacements, check for water in the system (obviously), etc for all aircrew/aircraft that had an incident. I'm sure whichever team has been tasked to deal with this issue is already knocking out that easy stuff.