I’m reasonably convinced that a 90deg head tube with reversed forks is just as stable and rideable as the conventional setup, it’s a biomechanical reason that normal bikes don’t use it.
Stable yes, rideable is debatable.
According to the calpoly uni 2wheeled vehicle chassis design course atleast … the steeper the headangle (for a given trail and COG) the less force will be exerted through the handlebar to allow you to feel for what the bike is doing. Holding the bike in a steady state turn, straightening the bike and turning harder all exert different forces at the handlebar. These forces are important for the rider’s ability to control the bike.
a 90degree headangle will result in a bike with no “feel”.
I can only assume this means a negative headangle will result in the forces which allow you to feel the state of the bike will be reversed.
I messed about making a bike with a 75degree headangle and 15mm rake fork, it was a fairly wicked thing to ride at times. Turned on a dime, which is what I wanted it for, but the steering didn’t give very much feedback and it was a difficult to ride. Other bikes with similar trail and more wheelflop handled a lot more intuitively.
The issue with a lot of the analyses above are that none of them are attempting to put the human into the equation. Hands-free stability theories will agree with what you are saying, Ben. But when you have a human rider as the control of a closed loop system the best bike examples have feedback that comes from wheelflop from a non-vertical headtube.