Message from @MOS MAIORVM (AKA Sola)

Discord ID: 311378303389007874


2017-05-09 05:43:26 UTC  

More is not better

2017-05-09 05:43:30 UTC  

My whole argument

2017-05-09 05:43:55 UTC  

You believe it's impossible to get 7 heads in a coin toss of 10 times?

2017-05-09 05:44:04 UTC  

I think most people would disagree

2017-05-09 05:44:10 UTC  

even if its not likely

2017-05-09 05:44:17 UTC  

it's still quite possible

2017-05-09 05:44:42 UTC  

Again. I said enough. Is 10 flips enough in principle to establish a certain premise about coinflips? No.

2017-05-09 05:44:46 UTC  

But 100k is

2017-05-09 05:44:51 UTC  

And more wont help

2017-05-09 05:45:01 UTC  

100k is more than 10

2017-05-09 05:45:06 UTC  

100k helped compared to 10

2017-05-09 05:45:11 UTC  

hence proving my point

2017-05-09 05:45:18 UTC  

Come on man

2017-05-09 05:45:22 UTC  

What?

2017-05-09 05:45:25 UTC  

This is really basic stuff

2017-05-09 05:45:25 UTC  

My point is what?

2017-05-09 05:45:30 UTC  

I'm not trying to knock you

2017-05-09 05:45:36 UTC  

Yes. Basic and irrelevant

2017-05-09 05:45:45 UTC  

It's not irrelevant

2017-05-09 05:45:48 UTC  

It's quite important

2017-05-09 05:46:12 UTC  

You're acting as if sample size doesn't matter

2017-05-09 05:46:15 UTC  

that's clearly not the case

2017-05-09 05:46:21 UTC  

It matters to a point

2017-05-09 05:46:27 UTC  

It matters period

2017-05-09 05:46:29 UTC  

Depends on what you study

2017-05-09 05:46:36 UTC  

No. It doesnt matter period.

2017-05-09 05:46:36 UTC  

Yes, we only have so many resources to expend on it

2017-05-09 05:46:43 UTC  

Per the 100k vs. 10 mil example

2017-05-09 05:46:49 UTC  

But it's important to get a large sample size

2017-05-09 05:47:16 UTC  

Maybe. It certainly isnt some inviolable principle. Maybe it is useful.

2017-05-09 05:47:22 UTC  

10 mil will still give you a more accurate result than 100k

2017-05-09 05:47:25 UTC  

It doesnt make it MORE TRUE

2017-05-09 05:47:47 UTC  

10 mil vs. 100k in coinflips

2017-05-09 05:47:56 UTC  

It's not about being "more true" but more accurate

2017-05-09 05:48:00 UTC  

Will not give you a more accurate result

2017-05-09 05:48:06 UTC  

it will

2017-05-09 05:48:19 UTC  

More accurate what though

2017-05-09 05:48:38 UTC  

Than what you'd likely otherwise get just doing 100k

2017-05-09 05:48:41 UTC  

in other words

2017-05-09 05:48:53 UTC  

Hang on

2017-05-09 05:49:00 UTC  

sure thing