Megan - Absolutely right. The point being that the aircraft was descending, but not necessarily accelerating towards the ground at 32 fps squared.
My interpretation of the control inputs was along the lines of;
Aircraft lifts off from the runway. Witnesses claim it was a longer ground run than usual. Did the pilots detect some slop in the controls and consider aborting the takeoff?
Once airborne, a right bank is initiated and a balanced turn commenced.
A left aileron input is made to adjust the turn and the control cable separates.
The bank angle increases and a left rudder input is made to counter the bank. The crew discover that the control column is not responding to aileron control inputs, but the aircraft is still in a climbing turn.
Progressively more left rudder inputs are made in an effort to control the bank angle and the control column turned fully to the left, to no effect.
As the bank angle increases through 30 degrees, the aircraft begins to lose height.
Bank angle approaches 40 degreees and the control column is pushed forwards to try and reduce the rate of turn. The elevators are now acting in place of the ailerons and rudder as a crude directional control. The rudder is now acting to keep the nose up.
As the bank angle continues to increase, the nose drops through the horizon and the aircraft descends clipping the embankment and shedding several feet of wing tip and engine four. The inertial kick from propeller strikes and the engine breaking away level the aircraft somewhat before the fuselage strikes the ground.
The aircraft was under some measure of control throughout the sequence and the crew performed the only actions that they could to mitigate the effects of the inevitable crash. Unfortunately, their efforts were in vain.
Had the aircraft not struck the railway embankment first, I would expect the fuselage to have made initial ground contact virtually fully inverted.
Given the speeds involved and the design of aircraft at that time, the accident would not have been survivable due to break up of the cabin interior and the physical stresses imposed on the passengers and crew during deceleration, even if the aircraft had simply flopped on its belly and slid to a stop.
If the aircraft had struck the embankment at a shallower angle of 30 degrees rather than the 90 degrees submitted in the official reports, I would expect both of the engines to break away from the wing and a much wider area of damage and debris crossing the railway embankment. There would also be two distinct parallel sets of propeller strike marks running over the embankment, which does not appear to have been noted at the time.