Very interesting video and findings. It lead me to a couple more questions:
1) Is the meter speed consistently the same for a ball?
2) Is the meter speed the same every game you play?
There was a theory a few years ago that the meter speed for a ball wasn't totally consistent for each swing. I think it was Andyson who did some testing and came up with the theory that there were 3 different speeds for any ball, each with just a very slight difference. The belief was that this was to try and combat auto-dingers. I remember Andyson also saying that he would always let the first swing run through, then hit the ball on the 2nd swing because firstly the speed was more consistent, and secondly it was less prone to meter spikes. I don't know how true this is, but I do remember once trying exactly this while playing an 18-hole round on Royal St Georges, might have been in the Virtual Open qualifier in 2011 or 2012, and had my best ever 18-hole score at the time. Never did it after that.
With regards to the second question, i always feel my meter is smoother and slower in the mornings. I've always wondered if my mind is just playing tricks on me or is it true.
So, having seen Yiannis's video, I decided to do some testing myself. I played 10 swings with a level 81 TM ball and then 10 swings with a level 47 TM ball. I did this last night - late evening of 1st Feb. In a similar way to Yiannis, I slowed down the video in order to do the timing. Unfortunately I kept making mistakes during the timing of the 10 swings, sometimes forgetting to start the timer or clicking a fraction too late as the meter starts the downswing, occasionally clicking the RESET button by mistake instead of the STOP button, and then almost every time I would think I've got my 10 timings and there was another swing! (There was definitely only 10, I counted them afterwards at normal playback speed) So, I abandoned it for the evening (had been doing it unsuccessfully for about an hour!) and decided I would finish it today. This morning I did another video with 10 swings on each of the two balls so that I can compare timings for different parts of the day. Unfortunately once again, when I was doing the timings I kept messing up, but I will get it done, just bear with me.
One thing I did learn from last night is that the timings for the level 81 TM ball were mostly between 10.7-10.8, and for the level 47 TM ball between 10.0-10.2. This was with the playback running at 25% of its original speed so that I could try click more accurately. Now obviously 1/10th and 2/10ths of a second aren't much, and could be accounted for by my hand/eye coordination when clicking the START button at beginning of the downswing. However, as you well know, the meter accelerates as it reaches the ding, so any difference in the time would definitely result in missing the ding, which would definitely affect any auto-dingers. And it wouldn't be difficult to program - you just put in an adjustment factor, random within a range, probably similar to what they do for precision.
Later today I'll try get timings to comapre morning and evening to see if there was any difference and I'll post my findings.