I have compiled the stats for the 45 most recently closed questions (because that's how far back the "Recently Closed" queue in the 10k tools goes).
Specifically, that's questions with the following IDs (ordered by timestamp of closure, from most recent to two days ago): 94472, 94491, 91950, 94486, 94517, 94514, 94519, 94523, 90877, 94507, 94494, 94529, 94533, 68520, 94525, 94566, 94598, 94552, 94615, 94615, 94593, 94542, 94619, 94577, 94655, 94675, 92704, 50917, 94699, 94720, 94668, 94698, 94568, 94660, 94739, 94719, 94710, 94742, 94729, 94676, 94677, 94750, 94754, 94662, 94790.
These 45 questions were closed for the following reasons:
- 11 dupe (incl. 1 merged)
- 10 NARQ
- 10 off-topic (incl. 1 migrated)
- 7 gen-ref
- 5 not constructive
- 2 too localized
The 10 questions closed as NARQ include questions from a banned user, who only posts them to mess with us, usually several a day.
Do note the top close reason. "Exact duplicate". That's not a fluke. It's perfectly in line with the long-term figures from the Data Explorer:
Exact duplicate 1575
General reference 967
Not constructive 371
Too localized 331
That bears repeating. 37% of our closed questions are closed not because they do not belong. They are closed because they do belong and as a matter of fact have been answered already. Four questions out of ten.
I am happy to hear SQL.SE does not get duplicate questions. Of course I could follow your example and reword that as "SQL.SE sucks balls when it comes to hunting down dupes. What an inefficient clique, obsessed with getting more and more reps from answering the same questions over and over again! WTF! Wake up sheeple!"
I will not do that, though. I also won't go over to SQL.SE and register two sock-puppet accounts a day,
to post inane questions, day after day, week after week, month after month.
What I will do instead is go back to looking at actual stats to address the rest of your question, starting with this bit:
the closure votes are drawn from at most ten high-rep members
A grave allegation to be taken very seriously. So let's have a look at who closed the 45 questions mentioned above.
- The 11 dupes were closed by: Andrew Leach (2), Cameron, Carlo_R., Cerberus, FumbleFingers (6),
Jim, KitFox, Mahnax, Marthaª, MετάEd (3),
Mitch (2), Mr. Shiny and New 安宇, RegDwighт (4), StoneyB (2), tchrist (5), waiwai933.
1 question was merged by RegDwighт; unclear (and irrelevant) who the closers were.
- The 10 NARQs were closed by: Andrew Leach (2), Armen Ծիրունյան, Carlo_R., František Stanko (2),
FumbleFingers (2), Hellion, Jasper Loy, J.R., John Lawler, jwpat7 (2), Kris (2), Mahnax,
Marthaª, Matt Эллен (2), MετάEd (4), Mitch (3), RegDwighт (2), Robusto (5), tchrist (6),
- 10 off-topic: Andrew Leach (3), Barrie England, František Stanko, FumbleFingers (7), Hellion (4),
Jasper Loy, Mahnax (3), Marthaª (2), MετάEd (5), Mitch (3), Mr. Shiny and New 安宇 (2), Robusto (3),
simchona, tchrist (6), waiwai933.
- 7 gen-ref: Andrew Leach (2), František Stanko, FumbleFingers (3), Hellion (2), Jasper Loy, Jim, John Lawler,
jwpat7, Kris, Mark Beadles, Marthaª, Matt Эллен, MετάEd, Mitch (3), RegDwighт (2), Rory Alsop, tchrist (6),
- 5 not constructive: coleopterist, Hellion, Henry, Kris, Lynn, Mark Beadles (2), MετάEd (3),
Mitch (3), Robusto, Rory Alsop, tchrist (3), waiwai933 (2).
- 2 too localized: Kris, MετάEd, Mitch, RegDwighт, Robusto, tchrist.
The grand total:
Andrew Leach (9) Armen Ծիրունյան
Barrie England Cameron
Carlo_R. (2) Cerberus
coleopterist František Stanko (4)
FumbleFingers (18) Hellion (8)
Henry Jasper Loy (3)
Jim (2) J.R.
John Lawler (2) jwpat7 (3)
KitFox Kris (5)
Lynn Mahnax (5)
Mark Beadles (3) Marthaª (5)
Matt Эллен (3) MετάEd (17)
Mitch (15) Mr. Shiny and New 安宇 (3)
RegDwighт (9) Robusto (10)
Rory Alsop (2) simchona
StoneyB (2) tchrist (27)
TimLymington (2) waiwai933 (5)
That's 34 people.
When you say that "the closure votes are drawn from at most ten high-rep members", you draw a line between Mahnax and František Stanko. Between helping close 5 vs. 4 questions out of 45 (11% vs. 9%). That strikes me as completely arbitrary.
And who are those people, anyway?
- Some of them are complete laymen; others have published books; others still have a degree in linguistics.
- Some are native speakers; others are not; for others still, English is not even the third language.
- Some hang out in chat for most of the day; others only visit the site every once in a while and never drop by in chat.
- Some are still in college; others have grandchildren.
- Some are from the US; some from the UK; some from Slovakia, Canada, France, Russia, Germany, Italy...
The only thing these people have in common is that each of them has 3000+ reputation on this site. Which they got by actively contributing to it: posting good questions, providing helpful answers, making useful edits. That reputation level enables them to vote to close (and reopen). That is not a "self-appointed patrol duty", as you call it. That is a duty appointed by the system, fully automatically, to everyone who earns it.
You seem to think that "closed" somehow means "go away with your stupid question".
You are wrong. "Closed" means, "there's a problem with this question, please fix it then it
can be reopened". In fact all the closed questions I checked have comments saying what made
people vote to close, and/or how they can be improved. More to the point, six questions did
get reopened, in the same time span we are talking about. Nine more have reopen votes.
The "fucked up" question actually has four reopen votes right now.
If you want to discuss why it got closed in the first place, I suggest you start from scratch
and post another meta question — using a calmer wording, sticking to the facts, and not
calling people names. I will be happy to answer it, too. There is a lot of backstory to be told about
questions dealing with fucking, poo, bitches, and niggers. We get a lot of pressure from above to close
them, and a lot of pressure from below to leave them up. We always had a fine line to walk there. There's
also something to be said about it being a translation question, and a lot to be said about it having hit
the MultiCollider. But this answer here is already way too long.