Two more, or more detailed hypothesis:
a) We are looking at something lighter than steel, perhaps aluminium cladding, that suddenly changes trajectory as it sails down. Chandler doesn't account for changes parallax, if he ever accounted for it in the first place
b) We are looking at two objects that pass each other closely
Also, I wouln't be surprised if Chandler underestimates the distance of the object from the tower wall and thus overestimates velocities and accelerations.
Same overestimate of v and a happens if he underestimates the the lateral component of the fall - the velocity at which the object moves away from the tower.
Unfortunately, we don't have Chandlers work to assess, only the results.
Truthers ought to be asking Chandler to release his work for scrutiny. Some folks at the board that RedIbis linked to basically did so already. Not sure if Chandler responded.