Hello Tony, You are correct in observing the slight drift to the left over the length of the video. I just measured the relevant distances and they are: drift from right edge of "Philips" label to center of label is 1/2" (0.5"). The distance from the spindle to the laser slash line is 13.5". That should change your calculations slightly.
The SME motor controller has incremental settings of, I believe, 0.1%. I can't remember where I read that though. The line does move slightly when lifting the stylus out of the groove. Halcros Victor TT video is very impressive in this regard. The SME belt drive table should be set for correct speed while playing an LP to account for stylus drag.
You are most observant to catch the slight drift to the left, which means the table is very slightly too slow. I can not hear this. A close examination of Halcro's DD Victor video also reveals a slight drift to the left of about the same 1/2". Note the laser dash is centered on the blue tack and by the end of the 4 minute video, it is just barely touching the blue tack as it has drifted to the left. This video has been held up as the gold standard reference for DD turntable speed accuracy. I have found no others on the net except the Sutherland website videos in which none of the turntables hold correct speed.
In the interest of collecting data on different turntable performance, and learning in the process, I borrowed Albert Porter's TimeLine and decided to share the results.
I'd be very interested in seeing your calculations and translating them into the actual speed, ie. 33.4 or so RPM. Thanks.