So under the assumption that children are randomly assigned to classrooms, 70% of 20-kid classrooms will have at least one student drawn from the 6% pool, and 30% won't. That's a "guarantee"?
Right, half is not "pretty much". Its meaning is more like "mostly", that's why it fits the above. It's something like, that situation is common enough it should be your base assumption when talking theoreticals.
For example, if you have 9 boys and 1 girl in a room, and pick one at random, you're not guaranteed to get a boy, but you're pretty much guaranteed to get a boy. It's a wishy-washy way of saying the most likely outcome by a decent margin.
The post you replied to said “they're not both girls”. That’s 75% not half.
Weirdly though, “it's pretty much guaranteed that they're not both girls” sounds unacceptable to me, but “it's pretty much guaranteed that at least one of them is a boy” sounds ok-ish, even though they’re equivalent.
“Big” and “small” shift their meaning in context in ways we are aware of, and not aware of.
Things like “almost always”, “nearly guaranteed”, “effectively never” also adjust their meaning relative to context in ways we are not aware of.
There isn’t a solution to different people having different instincts about these numbers, except being aware of the issue, and not trying to project one person’s intuitions as being the “right” intuitions.
I strongly suspect that what's going on here is not that a bunch of commenters are happy to agree with statements like "in craps, you're pretty much guaranteed not to roll 7" or "when you hit in blackjack, you're pretty much guaranteed not to get a face card".
Rather, they're making the statements they do because they feel that 17 one-in-seventeen chances must add up to one 100% chance, or because they can't tell the difference between 50% and 75%, and they don't feel like checking to see whether the things they're saying make any sense.
Comments
6% is about 1 in 17. Classrooms where I grew up had 20-30 kids.
So under the assumption that children are randomly assigned to classrooms, 70% of 20-kid classrooms will have at least one student drawn from the 6% pool, and 30% won't. That's a "guarantee"?
They said "pretty much guarantees". With those numbers, yes, that idiom is pretty much accurate. It's not the same thing as "guarantees" alone.
With those numbers, yes, that idiom is pretty much accurate.
About as accurate as saying that "if you see two children, it's pretty much guaranteed that they're not both girls".
That's not something that anyone would accept as accurate.
Right, half is not "pretty much". Its meaning is more like "mostly", that's why it fits the above. It's something like, that situation is common enough it should be your base assumption when talking theoreticals.
For example, if you have 9 boys and 1 girl in a room, and pick one at random, you're not guaranteed to get a boy, but you're pretty much guaranteed to get a boy. It's a wishy-washy way of saying the most likely outcome by a decent margin.
The post you replied to said “they're not both girls”. That’s 75% not half.
Weirdly though, “it's pretty much guaranteed that they're not both girls” sounds unacceptable to me, but “it's pretty much guaranteed that at least one of them is a boy” sounds ok-ish, even though they’re equivalent.
“Big” and “small” shift their meaning in context in ways we are aware of, and not aware of.
Things like “almost always”, “nearly guaranteed”, “effectively never” also adjust their meaning relative to context in ways we are not aware of.
There isn’t a solution to different people having different instincts about these numbers, except being aware of the issue, and not trying to project one person’s intuitions as being the “right” intuitions.
I strongly suspect that what's going on here is not that a bunch of commenters are happy to agree with statements like "in craps, you're pretty much guaranteed not to roll 7" or "when you hit in blackjack, you're pretty much guaranteed not to get a face card".
Rather, they're making the statements they do because they feel that 17 one-in-seventeen chances must add up to one 100% chance, or because they can't tell the difference between 50% and 75%, and they don't feel like checking to see whether the things they're saying make any sense.
Yea, pretty much.