Late to this thread, but a couple of notes I could add.
The velocity graph isn't useful much past burnout. For my products, I use the accelerometer for max speed at that point, but only as event detection after that. Certainly not for velocity on descent, or even apogee.
Once you appreciate how little pressure difference there is in a foot of altitude, or even 10 feet, you'd understand why buffetting on the way down leads to so much pressure noise during tumbling and on chute. Wind is evidence of a pressure gradient, and as you can imagine there's a lot of relative wind during a tumble. Kind of unavoidable, though you can filter it (as AltimeterThree does to calculate descent rates pre- and post-chute opening).
Terminal velocity of this flight was reached really quickly once the rocket arced over and began descending, probably by t = 20-22 seconds. That's when the slope of the baro altitude descent seems to have reached its final slope prior to chute release. As I'm sure some of you are tired of hearing me say, terminal velocity happens really quickly. The rocket is speeding up only for a brief moment at the start of the descent. After that it's just falling at a steady rate.
The relationships between rocket geometry and weight and descent rates are complicated. All of us have had a rocket that fell "flat" and rather slowly, and others that come down more vertically, and quite quickly. My own pet theory is that more attention should be paid to the BALANCE of the post-ejection parts and their Cg/Cp characteristics to promote whatever falling configuration you desire. For some folks, that's an organized, controlled vertical descent where the shock cord is stretched vertically using a drogue and nothing can get wrapped (though it may twist). For others, it's a more horizontal tumble. In any case, do what billdz did: measure!