the same reason that averaging out everyones guess for a count of jelly beans in a jar is always pretty close. the voices of a lot of the people in the crowd are off, but some are off below where they should be, and some people are off above where they should be, and it averages out to around the correct notes
Lots of the answers here are wrong. When you’re at a concert and listening to the crowd, they aren’t amplified. You are only hearing the people near you. If the people near you are singing well, you’ll think the crowd is singing well.
Also, when you’re at a concert, you’re “in” to the performance and don’t notice everything. I remember going to a concert and thinking that the crowd sounded awesome when they sang along to a sound. A few days later, I went to youtube to see if anyone recorded it. Found it. And the crowd sounded like crap.
There’s this thing called a “resonate frequency” and what a frequency is, is air molecules or air atoms smashing together at a faster or slower pace, we measure this in something called “hertz” which is measured by “cycles-per-second” these cycles determine whether something sounds “high” or “low”.
The thing about sound, is that sound is 3 dimensional, meaning like your chair or anything you can see its not flat like a piece of paper.
So things like the shape that the air atoms/molecules make when they smash up against each other effect how the frequency sounds regardless of what the frequency is.
This is called “Timbre” and it refers to every other aspect of sound beside frequency.
When human beings speak their voices (because they are 3D) produce multiple frequencies because your mouth and your throat which pushes air molecules/atoms around pushes a lot of atoms around, so when it leaves your throat you not only have one frequency that appears, but multiple, so you might be wondering? What makes crowds able to sing on key then?
Well here’s the interesting thing, when your brain hear’s a frequency, it has to understand it, otherwise it would not be able to know the frequency and all sound would sound like pots and pans banging around (or think a crowded mall.)
(And interesting fact there are people like this who cannot tell the difference between keys and they describe drums as sounding like banging pots and pans and music as screaming).
And so your brain takes the loudest frequencies of that sound and majorly cuts out the rest of the other frequencies, and the way it does that is it takes the highest frequency, and the lowest frequency, and finds the middle (the mean) and that is usually what you hear.
(This is also how some people can sing “two notes” at the same time. By messing with their resonate frequency.)
So with a crowd of people although some people are “off key” if enough people are *fairly* on key what usually happens is it averages out the sounds and you hear what the notes are.
And there is also a bit of correction that goes on as well, for instance since you know the song you anticipate the notes and so notes that wouldn’t match up with the sound but get close enough in terms of frequency lets your brain usually auto-tunes it on key (but your brain, much like auto-tune can only do so much) there’s a similar phenomenon where if you are hearing a concert you haven’t gone to from like 2 or 4 blocks away you’re likely to just hear a bunch of random frequencies and not be able to determine the lyrics or the notes.
This is also *coincidentally* why when you hear someone speak in a foreign language that you aren’t familiar with hearing often (even if you *lets say* study the language) you have to ask them to speak slowly because your brain doesn’t recognize the pattern in their speech so it literally just sounds like gibberish until your brain starts making the nueral pathway which connects the words and phrases in your mind to actual speech.
There is a lot more on this topic that I could go into but this is really all you need.
Latest Answers