The minimum speed guarantee to a certain extent is irrelevant, the actual speed you get is almost entirely dependent on the distance you are from the cabinet, so assuming no defects on the line , the customer equipment and internal wiring etc , if the line were capable of (for example) 70Mb , that what your line syncs at , irrespective if the MSG being 60Mb or 30Mb , it’s not like the speed is restricted to the MSG , it’s simply a threshold at which a problem can be investigated.
The only issue would be is if your line was delivering 70 and then developed a problem dropping the speed to 50 , with a MSG of 30 , you may struggle to get a speed issue looked at by the provider as it’s still above the MSG , but if the MSG were 60, then 50 is obviously lower and triggers a response.
As far as the difference, if you had a line in service, it is generating real performance data , so the mathematical model /calculation of the line performance is superseded with real data , if the user of that line were ( for arguments sake ) using their router on the end of a cheap extension cord and that was negatively impacting the performance , or the line had an unreported problem that reduced the speed but it was unnoticeable ( so was never reported or fixed ) the MSG would be lowered reflecting the real data that indicates a slower sync rate , if that person stop using the defective extension, or the underlying fault repaired, the line speed improves and the MSG increases in line with the increased real world data.
