the hydrodynamic loads on the rudder are a squared function of speed. When we went through these calculations on my boat, the NA, Edson, and Jefa all used as the load criteria the rudder at stall at "hull speed". But any boat that is relatively light (and certainly that Schumacher is light) can exceed "hull speed" by a wide margin. In the narrative there is talk of surfing at speeds in the high teens. Hull speed for that boat is around 9 knots. 18 knots will put 4 times the hinge moment (and other loads) on the rudder compared to what it will see at hull speed. On my boat the NA and vendors seemed happy with an ultimate of 2x over the hull speed number. That figure becomes working load at just 1.4 x hull speed. That is around 12 knots on my boat, a figure we have seen a few times, in conditions in which the autopilot was steering very actively! They planned for around 800 ft lbs torque at the rudderpost. I built for 3500 ft-lbs torque, and I don't think it is vastly overdone. The autopilot is rated at 1600 ft lbs at the rudderhead, equivalent to about 120 lbs applied at the rim of the wheel. Had I built per the standard calcs, the autopilot would be capable of breaking the steering gear, and that is just wrong. The rudder and steering gear need to be designed to handle ANY speed the boat is every going to see.

The bracket it their photo is a bent SS plate, shaped like a cardboard box with no top or bottom. This shape collapses easily. I'll bet that thing has been flexing like mad for years. There is a gusset at the foot that has pulled loose which is just going to stiffen it locally. That causes the bending moment in the plate due to the lozenging to be resisted by the bolts in the foot in tension, until they pull loose. It needed diagonal bracing to stiffen it up, or preferably a shear plate underneath the ram filling the "box". If that had been done then it would have survived until the fasteners sheared off.

One other factor: hydraulic ram pilots are considered to be more robust than electric drive units. While this might be generally true, when the valves on the ram system close, the steering is locked until something breaks, unless there are pressure relief valves in the system. A electric actuator will blow a fuse and get back driven when overloaded. Which is the better outcome might depend on the situation, but they do fail in different ways.