Originally Posted by sabino56
The reason is the distance from where the tire touches the pavement at the front straight back to where it leaves the pavement is a little shorter than arc of an undeflected tire (draw an arc and connect ends with straight line - the straight line is shorter than the arc).
I'm afraid you're still missing the point. I wouldn't expect the manufacturer's circumference and wheel revs per mile information to correlate EXACTLY. But, this has nothing to do with that which you're describing. You're describing a radius change which will, most definitely, affect the thrust delivered by the tire patch, but has nothing to do with the tire circumference which must pass under the axle with every revolution.
I suppose you could check this out by comparison rollouts with 10 psi and 50 psi. Again, they might not come out EXACTLY the same, but they'll be a whole lot closer than road surface to axle centerline differences would dictate.
In short, give this a lot of thought before you decide to disagree with the way the industry handles its performance calculations.