1) AR-style rifles eject their expended brass upward and forward.
2) You'd think this "100,000 round shooter" would have noticed that ejection trajectory while he was getting pummeled by hot brass bouncing back from that sheet of plywood he was shooting through...
3) When shooting through a window, the best-concealed position is well back inside the room -- but, anyone who isn't masochistically enamored with being beaten on the head with a hammer will soon learn to stick at least the muzzle outside, to minimize the reverberating muzzle blast.
4) With the muzzle outside the window, where will most of the brass go? Out the window!
5) Gee -- one might wonder if the brass wound up in roughly the same place the broken window glass probably did: on the flat roof(s) of the 1-2 story building extensions 30-odd floors below his shooting position...
~~~~~~~~~~
The whole article is an exercise in conspiracy-wacko dumb@$$ery...
Generally speaking, I think you're right, especially for the most popular AR15 configuration (16" barrel, chambered .223/5.56, with carbine-length gas system). However, I've read (and seen it confirmed on occasion ;>) that ARs with weaker-than-normal ejection will eject empty cases more to the side and rear (from roughly 4 o'clock to 5 o'clock, with the muzzle at 12 o'clock). Weak ejection can obviously result from any number of causes, including under-powered ammunition, worn/out-of-spec parts, and (I assume) lack of lubrication. Your suggestion that many/most of the cases went out the window(s) is probably "spot on"...