Just for fun, here's another weird singular data point from this week's NYE gig—sound company supplied IEMs, and acoustic drums but otherwise everyone is DI-only w/ no stage amps or wedges.
The keyboard player is very experienced, but he's only on his second gig with this band. Normally, no problem at all, but in the first set we're all earning our pay—and keys most of all—with the just workable IEM mix. To launch the second set, the guitarist kicks into "Pride and Joy," and the keyboard player now can't hear *anything* in his IEMs. No biggie, thinks keys: he eases out one earpiece and has enough of the acoustic drums on stage to hear where the "one" is. But one chord in and now he's second guessing himself about whether we're playing the song in "Eb" (as recorded) or "E," and he's getting some of the FOH spill—but the stage sound is mostly sub and drum spill where keys are set up.
Keys thinks we're playing in "E" from the guitarist's hands. Hmm, that doesn't sound right. He shifts to "Eb." Still nope. I'm on the other side of the stage, wondering what's going on in the instrument bed under my IEM mix of very loud vocals and drums. Singer has a lot of keys in her mix; and while trying to stay smiling and not get pulled off pitch, she (bless her heart!) quickly mouths to keys, "F." He tries that. Still nope. Guitarist, when he can finally hear enough in his own IEM mix to figure out where the weirdness is coming from, lets keys know we're playing the song in "A" (b/c Etta James style female singer). Now keys has it; but we're well into the first chorus at this point.
Here's the thing: all through this trainwrecked first minute of the song, the crowd in the sold out ballroom is streaming onto the dance floor and they're looking euphoric. At the next set break, we're all just floored at how enthusiastic the room was about that song that we dished out in four different keys.
I'm not saying that crowds are stupid, or that they don't hear anything. But inexpert attention is weirdly limited, and in the right circumstances it tends to engage with what it *expects* to hear as the key foreground feature rather than attend to elements in the background or at the margins. (That's one of the basic lessons of the "invisible gorilla" selective-attention experiment.)
Got a guitar tune with happening vocals and guitar? Who knows what the crowd is able to *not* hear, especially around the edges and at least for a few moments?