I've been following this since the accident happened. I maintained, fitted and removed, and occasionally flew in a number of Ejection Seat types whilst serving in the Royal Air Force: 4PB - Jet Provost 5A; 4H series - Hunter T-birds; 6MSB - Buccaneer. I also had knowledge of all Ejection Seat types through my role as Spares Forecasting Officer - Conventional Armament Systems - at CSDE. I also did the 10A - Tornado F3 Ejection Seat as part of the Tornado AD Wpns Q course as part of the FI prep; in the end I didn't go down south, but that's not relevant here.
B Word - thanks for posting that schematic; very useful.
Something that I don't believe has been mentioned is the phase of flight, i.e., was the aircraft on the ASP prior to engine start; post engine start, taxy for take-off, etc. This could have a bearing - fully admitted speculation on my part.
On the question of inadvertent initiation of the Rocket Pack, I asked the question regarding whether or not an Ejection Seat could go up the rails without the normal Primary/Secondary Cartridge initiation sequence having taken place. The consensus from Bay/Line Armourers, CSDE Wpn Spec Flt and a number of M-B engineers I spoke with was: maybe.
If the Seat Pan were to be positioned in its highest position - this being the most beneficial for maintenance access to the cockpit floor areas - there could be sufficient space available for the blast and reaction forces to be dissipated. Were the Seat Pan to be in its lowest position there could be sufficient reactive force from the rocket efflux acting on the base of the Seat Pan to overcome the Ejection Seat Top Latch - please note: the Top Latch is not "released" - thereby pushing the Ejection Seat up the Ejection Gun and Guide Rail Assembly. It would depend on the position of the Seat Pan, and whether or not the Ejection Seat was occupied at the time.
My thoughts are with Sean's family and friends, and the Red Arrows, especially the Armourers. RIP