I realize and accept that "exact" = "impossible". When all 3 scenarios fail to reproduce the only clearly photographed observable, and when NIST admits that this observable can be created in the simulation with a minor adjustment, I have to question their not doing so!
Well, yes, but asked and answered: Because it would not significantly change the results of the simulation.
Short Version: There is a random element to the simulation, and the conclusion they reached was based on the most likely results of the randomness, allowing for the fact that ANY given run of the simulation would result in anomolies.
Longer version: the report outlines the anomolies in the simulation; Simulated events that differ from what they could observe happened during 9/11. We don't know, however, how exact the rest of the simulation was.
There is essentially a randomness to the simulation, due to many unknown values and Margins of error. Even using the same known values and running the simulation several times will yield different results. Suppose they ran the simulation with every measureable value exctly as it was on 9/11, but didn't know it. And still the wheel only passes through 1 time out of 10 runs of the simulation with those numbers. If the Wheel is the ONLY measure of how correct the sim is, then you'd throw it out. Therefore, there was more to their decision to accept the simulations that they did accept besides just the ejected debris.
Trif