Let’s see if I can work some numbers out. Let’s pretend that the stroke each way is one inch, and that somebody can do 16th notes with the metronome set to 200bpm. Hopefully this is right:
- Time per half trip is 75ms (a round-trip would be 150ms, for two string hits)
- Average speed over half trip is 1 inch / 75 ms or 0.758mph (we walk at around 3mph)
So the pick moves amazingly slowly, in some sense—much slower than one walks. The problem, I see, is that the pick is constantly turning around; and that, I suspect, is where the difficulty lies, in getting speed.
I’m not sure if the pick basically launches (in the sense of a twitch) to go in one direction, and then does that again, 75ms later, to flip around in a different direction… I don’t know.
Now, let’s prove that you’re right that the pick barely spends any time with the string: If it is in contact with the string for 2mm (say), then the amount of time to do that is around 2mm / 0.758 mph or around 6ms. So, it’s perhaps 8% of the time?
Now all of this is really approximate, but I think it points to the fact that turning around is where all the action is! (This assumes that the pick suddenly changes direction, but that’s not true, hence it will move somewhat faster.)
One interesting number that I would like to see from experiment (a Magnet video) would be “how long does it take the pick to turn around?” In fact, what I’d really love to see is the speed of the pick as a function of time.
In passing (I’m sorry that the post is getting so long!), a great Cracking the Code project would be to take a guitar pick (obviously a Jazz III) and glue a 3-axis accelerometer to it, and then all kinds of crazy interesting stuff would come out of it… however, I suspect that with the right software the Magnet already has all of this information ready to be extracted.