I'm not sure if it matters whether it was a wet or dry system in this case. I think what might have happened is that the N20 solenoid leaked. As someone else in this thread mentioned, it will seek the low point and fill whichever cylinder has an open intake valve. N20 is approximately 60% more dense than atmosphere, considering that the atmosphere is only 20% O2 and N2O contains one oxygen atom vs 2 for O2, the piston would only need to be 1/4 of the way down for an equivalent 100% volumetric efficiency, yet highly volatile charge.
The injectors dump a ton of fuel in the startup algorithm, so what most likely happened is that he ended up with a cylinder full of N20 and plenty of fuel to ignite it, probably before the piston hit TDC. The end result would be a broken crankshaft