Perhaps it is time to put to bed the differences in basic training and using different techniques to control height/speed.
We are assuming that, as the serviceability of the SP seems to have been OK, that there has been some form of procedural/pilot error. So how did this happen?
We have 2 experienced pilots flying a relatively simple IFR approach, in IMC, as cloud was probably scattered at 200ft, broken at 300ft. They knew the MDH was 300ft so would be trying to fly as accurately as possible to achieve the required visual clues at, or just before, the MAP. Yet the airspeed was allowed to decay to such an extent that the a/c developed a RoD which was high enough for the a/c to enter the water?
The NHP would have his eyes clued to the instruments while the HP would certainly be monitoring them while looking up to see if he had the required visual parameters to land. His scan in/out would increasingly be out as he neared MDH and probably fully out by the 100ft to go call.
The AIB full report will make very interesting reading especially looking at the casual factors such as training/autopilot approach SOPs/management philosophy on the use of the autopilot. Then the human factors as to why 2 pilots didn't recover from a low speed U/P before they entered the water. Then any technical faults which may have confused or misled the crew into thinking they were in control.
HF