While it was the paradigm back in LF film days, it was only "close enough" for general purposes and not really a precise way to handle it. Even back then, dedicated landscape photographers would spend the time to critically focus to the corners of their GG's with high-power loupes to insure maximum resolution for large prints.
A more correct methodology is to focus at your the pivot axis of whichever standard you were using for tilt -- in the case of your M-2, this will be the front and so your tilt axis "moves" whenever rise or fall is imparted. (With rear tilt, your axis does not change relative to the image on the GG itself.) Then as you tilted, foreground elements -- seen inverted at the top of your GG above the axis -- would pop into focus while background elements also crisped up below the axis similarly.
Finally, all of this becomes significantly more critical with MFDB due to the small sensor size and associated hyper-critical focus zones. So IMHO you really need to spend the extra amount of time insuring critical focus at all of the important ares of your composition, which means focus will become a sort of iterative process. Fortunately, once you get the hang of it, you zero in on the "best compromise" iterations pretty quickly.
Cheers,