Welcome cosmicway,
I will definitely read the article on prediction science which you cited. So that you can add to your enquiry a discussion on how TrackMaster makes their variants (and speed ratings and class ratings), here is a link to a Youtube video which describes in some depth the methodology they employ to create their numbers. RDSS and some older Sartin programs employ these variants (Daily Track and Inter-Track) to equalize final and sectional times.
All variant making is imperfect, and we seek one which is not only the least imperfect, but perhaps more importantly -
good enough. I think it also pertinent to this discussion, and important, to cite a point-of-view of Dr Sartin's re variants and final times - namely that more often than not the
final time of a race is highly dependent on the matchup of horses' running styles, early and mid-race pressure - often much more so than the supposed track maintenance and weather related explanations for variances in final times. (Excerpt from 'Pace Makes the Race':
http://www.sartinmethodology.com/pub...inningPMTR.pdf - ending Q & A's re variants on pages 14 and 15).
It is for this reason that - for better or worse, though consistently - later Sartin software, including RDSS, frequently employs a percentage (e.g. 50%) of the Daily Variants to modify (i.e. seeking to equalize/normalize) final and segmental times - accepting that the inter-race matchup of running styles and pressure accounts for the significant portion of all final time variances.
Hope that is interesting addition fuel for the research fires. Good luck!
.
Ted