Welcome to the Forum Archive!

Years of conversation fill a ton of digital pages, and we've kept all of it accessible to browse or copy over. Whether you're looking for reveal articles for older champions, or the first time that Rammus rolled into an "OK" thread, or anything in between, you can find it here. When you're finished, check out the boards to join in the latest League of Legends discussions.

GO TO BOARDS


@Riot: 10% Crit chance is actually 8%

Comment below rating threshold, click here to show it.

Steven Mcburn

Senior Member

02-19-2013

Jesus people want 50k?

I'm pretty sure he's being damn accurate with anywhere over a thousand.


Comment below rating threshold, click here to show it.

nillthium

Senior Member

02-19-2013

bump


Comment below rating threshold, click here to show it.

Heyuu

Senior Member

02-19-2013

Quote:
Steven Mcburn:
Jesus people want 50k?

I'm pretty sure he's being damn accurate with anywhere over a thousand.


you obviously have no gasp of the scale required to find patterns in seemingly random events. . .


Comment below rating threshold, click here to show it.

Cameran

Senior Member

02-19-2013

Quote:
OBITWIZZLE:
If you had something closer to 50k data points, I'd be more inclined to jump on your bandwagon. In the grand scheme of things, 2-3k is fairly low.


^ Is awful at statistics. Anything over a few hundred is a sufficient sample size...The only legit complaint is if something in the crit-script varies on a game by game basis, and thus he would need to do it in more than one game.


Comment below rating threshold, click here to show it.

Heyuu

Senior Member

02-19-2013

Quote:
Cameran:
^ Is awful at statistics. Anything over a few hundred is a sufficient sample size...The only legit complaint is if something in the crit-script varies on a game by game basis, and thus he would need to do it in more than one game.


Bullsh*t

Having built seemingly random event scripts myself it is totally easy to provide enough variables in said scripts to require 2-3 MILLION tests to find patterns. . . ( this is how modern Cryptology works. )


Comment below rating threshold, click here to show it.

Bergtau

Senior Member

02-19-2013

Quote:
Cameran:
^ Is awful at statistics. Anything over a few hundred is a sufficient sample size...The only legit complaint is if something in the crit-script varies on a game by game basis, and thus he would need to do it in more than one game.


Nope. The amount of samples required depends on the degree of precision. To notice large trends, only a hundred or so is required. When a single deviation can throw off your results by a third of the amount of total deviation your sample size is insufficient.

Not that the OP's is insufficient.

Quote:
Heyuu:
Bullsh*t

Having built seemingly random event scripts myself it is totally easy to provide enough variables in said scripts to require 2-3 MILLION tests to find patterns. . . ( this is how modern Cryptology works. )


One variable is being tested.


Comment below rating threshold, click here to show it.

Heyuu

Senior Member

02-19-2013

Quote:
Bergtau:




One variable is being tested.



You are correct, and I stand corrected.


Comment below rating threshold, click here to show it.

TheGrandDuelist

Senior Member

02-19-2013

Wow, comon Riot get your crit together.


Comment below rating threshold, click here to show it.

Brohoof

Senior Member

02-19-2013

bamp from school


Comment below rating threshold, click here to show it.

LilyThai

Senior Member

02-19-2013

Bump

And just going to say I told you so, back when they were changing the crit chance in the first place, I said riot, this wont work, it's a bad idea. Now look.