Let's pretend that you were the commander of a sub below an aircraft breaking up. Would you idle your prop and watch the action or try to get out of there in the most expeditious manner. One thing which happens when the thrust line drops relative to mass centroid is that the nose should pick up while the stern drops. As the stern drops more of the prop would be exposed to the water. Maybe a balls out condition gives a higher speed than simple cruise. I don't know sub specs but can imagine that the physics change. Force vectors act similarly in all systems. When my employer was testing ramjet powered target drones, they used a booster which fell away when the main engine took over. Major problem developed when the flight control transients switched and we had many drones dig holes in the desert. I suggested that the booster motor be installed inline with the cruise ramjet so that the flight control would have very little change after boost. Marquardt went with this suggestion and created the Low Volume Ramjet and had a new market niche. As far as I can recall, I never received credit for this suggestion. I also came up with a new chaff type of corner reflector which could jam receiving radar stations but it did not go into production. I feel that much of my own experience supports my TWA800 pitch. With thanks to Elmer Fudd for replaying my earliest thoughts on the crash. My education over the past few years has changed my theory a great deal.
"Let's pretend that you were the commander of a sub below an aircraft breaking up. Would you idle your prop and watch the action or try to get out of there in the most expeditious manner. "
In the real world, even if you want something to happen, sometimes it just isn't possible. I can give my wife's mini-van all the gas in the world, but it will never go from 0 to 60 in 5 seconds. Likewise, the commander of a submarine on the surface may want to leave an area at 30 knots, but it just isn't going to happen.