The time of flight measurement does not appear to be between that of two discrete events, but rather between two distributions of events. As best I can make out, there is an underlying assumption that there is a linear correlation along the time durations between the two distributions. I think that is also what the author of this article is getting at in his critique: