[Cold Heat]

The shuttle was going too fast and was executing, (or trying to execute) the roll program in order to slow down.

If the control surfaces were not working on the left wing at that time, it would have had a hell of a time keeping trimmed out and would be fighting to make the turns that it was being asked to do. The computers would likely begin to over correct and gradually loose control. (which it did)

[chimera]

At that altitude, its hard to imagine the control surfaces being all that effective. It'd have to be thruster activity. Failure of control thrusters is not unheard of. Gemini 8 had to be brought down early because of a stuck-open thruster. In May of 1968, LLRV no. 1 (LLRV was the predecessor to the more advanced LLTV) crashed while being piloted by Neil Armstrong. The failure was in a bank of control thrusters. Armstrong bailed out and lived to fly another day, this time a bit further and with more success.

When this happened my initial thought was a control system problem. It doesn't take much at those speeds to lose control. Loss of yaw control would put it into a spin. From that one video there seemed to be a violent yaw 90 deg. to the flight path. Whether that was a sympton or cause, we don't know.

[SauronOfMordor]

If the control surfaces were not working on the left wing at that time, it would have had a hell of a time keeping trimmed out and would be fighting to make the turns that it was being asked to do. The computers would likely begin to over correct and gradually loose control. (which it did)

The problem with software is that if you have code to deal with unusual situations (like unusual drag on one wing), that code might not ever get exercised in real life until something goes somewhat wrong.

And if there's a bug in that section of code, it may turn a "somewhat wrong" situation into a "catastrophicly wrong" situation. I would take a good hard look at the possibility that unusual drag (or loss/corruption of sensor input) may have caused the computer to overcorrect at Mach 20 (with disasterous results).

[r9etb]

The shuttle was going too fast and was executing, (or trying to execute) the roll program in order to slow down.

This is standard, and designed into the entry trajectory months prior to launch. What you're referring to is "energy management," which is a technique used to match up the trip from the runway to entry interface (400 kft), and the trip from deorbit burn to entry interface.

(They start with a known spot -- the runway -- and work backwards from there.)