You seem to have a naive concept of how some people will act at 200 ft on air. I don't know which is worse, but both scenarios both present significant mortality risks.
You seem to have misunderstood my post.
I was describing comparative, not binary, risk in response to
@LI-er's comment "I can top my own example."
Both dives are trust me dives and involve significant violations of every training/community standard there is. Both are risky. However, one type is pretty obviously riskier than the other based on the fact sets of the respective dives.
If the dive described by
@LI-er is in fact the San Francisco, it's 200' to the sand. 170' to the deck. Most recreational divers will on a sightseeing tour in the 160' area. Moreover it's conducted as an open water (non-penetration) dive in warm water and good viz with no chance of silting out or getting lost. TBT is 10-15 minutes. This is a tag the wreck bounce dive. The guide is right there and can easily see the whole group. If it's the operator I'm thinking about this dive is the last one in a week long series of progressive recreational dives and would not be offered to someone who screwed (gave the guide concern) up on a previous (shallower) dive. Contrast that with the type of engine room penetration
@LI-er described where the wreck may be upright or on it's side swimming down multiple ladders/multiple buklheads/cat walks three decks down with little to no ambient light, possible leaking oil, lots of silt and a conga line of untrained unskilled ill-equipped (poor lights/no redundancy) divers following a single guide who's leading from the front. The chances of making a wrong turn and getting separated, getting entangled, losing situational awareness (direction, gas managenment, etc) and otherwise getting lost/freaking out or doing something stupid is material. Narc is influenced by many factors (not just PPN2). Many of the local guides do not use a long hose or doubles so exiting OOG, while keeping the whole group together, would be a nightmare.