Whatever works best in your system.
B744, 757 767 777 787, VNAV does an adequate job but it is only as good as the wind profile that is entered into the box. WInd change is not linear in most instances, so any assumption of that condition will result in less than optimal path establishment. It is interesting to watch crews fly a VNAV path in an altering gradient, as the energy state of the aircraft is paradoxical to the crews actions; e.g., in path, an headwind that is greater than planned will result in a higher speed than anticipated, and heads off towards a "drag required" message, and putting out the boards, whereas the energy state of the aircraft is now low... resetting the path by going direct to the current waypoint, or by reentering descent speed etc will give the opposite result, waking up the throttles..
Airbus VNAV is disappointing in it's limited scope.
Using a multiple times your height sounds great, if you know your distance, and you are going to do the maths. Most nav systems will provide time to a WPT and as the desired outcome is what is the correct vertical path, which is dy/dx, but is flown as a rate of descent, dy/dt, then time seems a logical measure to apply. 5 minutes to a waypoint, x feet to lose... An assumption of X times distance is only valid when you have the correct value for X which changes every wind case (and different weight if a constant speed profile is intended to be flown).
In extreme cases, time works nicely; In one sporty jet, I known I can get down from FL450 to 10,000' in 2:00, anything else is a choice... (don't recommend that one at night... 30 degrees nose down at MMO/VMO is quite aggressive even when the blue is kept at the top. For a large vertical offset, such as cutting 20 miles off a distance to touchdown leaving 10nm to touch, at 300kts, FL100, well, for that particular type plonking out flaps 40, full brake and gear gives a rate of descent that can be applied to the time to threshold etc... telling the driver if it is going to work or not well before the ground rush comes up. time to target tells you something meaningful and immediate. That is also what you are seeing when you look out the window, your aiming point is the result of your dy/dt, for any given GS, dx/dt, so the basis of the analysis is time.
An efficient descent based on idle from TOD to configuration complete and established at stable condition at an appropriate altitude on final, nowdays is mainly wishful thinking. It doesn't happen in ZGGG, RJAA, UEEE, KLAX, and only the first arrival of the day may do it into EGLL..., it doesn't happen many other places, our airspace is a victim of the success of the industry. Mainly descent efficiency is compromised by STAR design constraints and traffic needs. As such, being off your original desired speed, means that the target path really is already mussed up, and what you have left to work with is time to target... giving needed sink rate.
Whatever works for you is arguably the best method.