The Halifax issue was the stalling of the fin and rudder, caused if it rudder was applied excessively. Test found that with any trim applied the rudder overbalanced at speeds below 150mph, and became completely unrecoverable if one or more engines was lost. Losing an engine wasn't the cause, but it didn't help.
My point about Schräge Musik was that the aircraft often didn't suffer a structural failure and break up - they were set on fire and blew up when the bombload went up (as happened to two friends of mine on different crews) or the fuel level was such that the fuel tank exploded. Taking figures for losses of this nature as a 'mid air break up' doesn't really prove the strength of one airframe against another as the loss is caused by damage and enemy action rather than an inherent weakness.
As for the survival rates, I'd be interested to see the survival rates if similar numbers of Halifaxes and Lancasters were counted, as opposed to a large number of Lancasters against a small number of Halifax as was the case during the Bomber offensive. To illustrate the point and how it can be easily swayed - what were the Lancaster survivability against Halifax on Coastal Command missions? Does that mean it was a better aircraft for Coastal missions? No..
Statistics are just numbers on a page. You can make them do as you please.