You have to use caution in determining rpm from video, aliasing can produce strange (even static) perceived rpm, as we see often on videos of rotors or props. It may be valid to say theres an rpm change, but whether it's positive or negative or how much is not really possible.
For me the accident sequence is:
- Sudden collective pitch change.
- 1 blade strikes fuselage & departs
- Out of balance rotor over-stresses the shaft and fractures