The 45HP hyperbole wasn't lost on me. I get your point.
A restrictor would have to be small enough so that what is both downstream and upstream of said restrictor does not have much influence on the end result. When the restrictor isn't sized as such, then what is up and down stream of said restrictor influences how well that restrictor flows.
Just like a wing. That element isn't always going to produce x amount of downforce at a given speed and angle of attack. What the air is doing in front of and behind that wing will influence the flow and efficiency of that wing.
As was stated above valve lift is not averaged, unlike deck height and manifold measurements.
If you set the max lift at MAX and go run it for 40 minutes at race speed (say 5-10% of it TBO) you will have wear affect that reading by maybe by .0005", maybe by .001"? I don't know exactly, but it won't be the same. The issue is that the cam nose, the pushrod ends, and both ends of the rocker, the tip of the valve stem, the valve seat and the valve are all some amount different. When you now set the valve lash at ZERO, the geometry is some amount different than it was before the wear and you will now end up with a different lift than you had before.
How close you want to cut it is up to you. I wouldn't be giving up .002-.003 thousandths on valve lift either. Heck, I don't want to give up 10# at the scales as a safety margin. But that's my choice. As it is to all those building these engines as to how close they and their customers want to push it. I don't fault any of them.