I'm not here to question Ethan expertise, but I would like this clarified before I go a head and start treating my mixing room. On another forum some members claim an air-gap behind bass traps is of little use. The following post is pasted from said forum.
"Regarding the data and simulations with air gap. Those results are with the perimieer sealed, meaning the sound has to go through the absorber material. It can't get around it. If your installation has a route to the air gap that does not require passing through the absorbent, sound with roughly the same wavelength (or more) as width of your panel will not get much benefit from the air gap.
A 2ft wave is about 500Hz. Since the primary benefit of the air gap is to get absorption at lower frequencies you can see that not sealing around the edge of the panel is going to take away the benefit of the air gap.
However if you just cant seal it off you can help by making the panels bigger. Butt your 2ft by 4ft panels together to create an 8ft by 8ft panel. That brings the frequency that can sneak around the panel down to 125Hz."
Another forum member concurred and linked to an old video of Ethan's (a would-be tour of his personal listening room) and pointed out, in his opinion, the flaws and bad sound of the room, calling Ethan's expertise into question.
To Ethan or any other knowledgeable people; Air gaps behind bass traps, yes or no?
Your idea of part of the sound wave going around the panel does not prevent the part of the sound wave impinging the panel directly from being absorbed. The part or section of the sound wave directly hitting the panel will vibrate the material which in turn dissipate the transformed energy in the form of heat. If you have more surface area (more panels), you'll absorb more sound because a greater section of the sound wave will hit the panel(s).
It's true that lower frequencies tend to "turn corners" but in the case of a wave going around a panel, think of the wave as wide as the room it's in and not like a Roomba or a radio controlled model car. The edge of the wave that goes around the panel will hit the wall behind the panel, bounce back off the wall and some of it will hit the back of the absorber. You can do an experiment to prove this in a wave tank.
You have to qualify the statement 'little benefit'. In my tests, gapping makes a significant improvement or can even be detrimental. For instance, gapping my cloud 4x vs 1x in this room created or worsened an upper bass null by over 12dB. Angling my side panels, thereby modifying the gap, I went from not achieving my target of +-10dB SPL and 20dB decay within 150ms. Likewise, panel location matters though that doesn't get quite as much coverage, pardon the pun, as things like material, quantity and shape.
To kind of specify on my statements in the linked thread - it seems like 1x tends to be the golden mean. I said from flush to 2x because that's how I have my side panels angled - and even some corner ones - and that's working well. When it comes down to it, you're going to have to experiment, measure and analyze to fully squeeze the value from your installation.