This is a great post.
I think what haraldo failed to emphasize is the reason Dunlavy, Vandersteen, Thiel, etc disputed the measurement techniques of Stereophile is because that measurement style is "inaccurate" for that type of design (first order crossover, sloped baffle for time and phase accuracy), or so they say.
I don't know if it's the first order crossover or the time and phase part, but all of those manufacturers confidently claim that 1.5 meters is too close and doesn't allow the drivers to sum properly. Whether that is true or not I honestly can't say, but it would make sense based on the thousands of owner and listener comments.
In the end you really need to listen for yourself to decide if you can hear the claimed FR deviations from the Stereophile measurements or the benefits of time and phase accuracy. I'd bet many people couldn't hear either (assuming the FR deviations are legit), as most people haven't been trained or trained themselves to do so.