Hoffman- that plot appears to show a lot of reversion at the valve on both IO and IC? Also, how do you correlate those plots to your goal of 300fps + or - as stated?
That is a the valve seat. As you can see, moving up into the throat then just a little over an inch into the port, it drops off.
I think one thing to keep in mind is what good low-lift flow can and can't do. Low lift flow will make the engine think it has more duration than it does (remember, seat-to-seat duration is what the engine really sees).
Another way of looking at it is this, especially as it relates to a 2 valve engine:
To get a certain amount of lift you need a certain amount of duration. The velocity and acceleration of the lifter is going to be dictated by type and diameter. There is a fixed value. Yo can only move the lifter so fast. Since we only have a certain amount of time (duration) to create lift, this will ultimately be the limiting factor. Spintron development and spring technology have allowed us to do more and be less abusive to the valvetrain, but ultimately we're still limited.
Take for example the NASCAR flat tappet change to roller lifters. I asked Jay Wiles of Hendricks if they would gain any more power. His response was no because they were already at the acceleration limit of what a spring could do. The flat tappet cams used high rocker ratios (2:1+). The rollers, while allowing for more aggressive profiles will need less rocker ratio to keep the valve acceleration equal to that. The net is there was really no change in the actual valve lift profile as the engine sees it. Anyway that is a tangent.
So getting back to velocity. You could convert it back mathematically, but the point to illustrate is what is important in the valve lift equation. There is a lot to be said about velocity - just look at what Larry Meaux, Darin Morgan, Chad Speier, Larry Widmer, Curtis Boggs, and many, many others have written over the years. Just to rant, it cracks me up when people talk about drag racers hogging the ports out. Nope, not the good ones. The only difference is that a drag race engine will likely rev a little higher, so the port cross section will be a little bigger, but the same physics apply. This is why when you see a cylinder head that is sized properly and produces peak power at say 7500rpm on a 1.8L engine, when put on a 1.0L engine will want to peak at 9500rpm. Do the math and you'll see why. Piston cfm demand drives the whole engine equation and everything else is sized from there.