soudle_noop

soudle_noop



17 Jan

Comment

Originally posted by Will-the-game-guy

I've found the bot wont let you do a manual report if you already reported someone in game.

Am I doing it wrong?

I double checked with some folks and apparently what you are observing is correct! I think my previous suggestion was in fact, wrong.

Sincere apologies.


16 Jan

Comment

Originally posted by Will-the-game-guy

Anyone that runs into accounts like this, dont report them in game.

Screenshot the name (if you can) and manually report them after the game via the website.

Itll get reviewed by a human and theyll get the boot.

Feel free to submit player tickets if you feel strongly about the name.

That being said, you SHOULD also report them in game if you think it's inappropriate.


10 Nov

Comment

I'm very curious to see the prediction accuracy of human agents vs a machine learning model.


26 Oct

Comment

The use of additional statistics to compliment the model is a pretty smart idea, especially when dealing with this kind of problem which has very few number of games available.

On top of my head, I think another interesting thing to try is to use a model that can propagate through time. Say this one (warning: I've never personally used it), or any other kinds of TTT (Trueskill Through Time) implementations that are out there. It might lead to more accurate estimates.


17 Mar

Comment

Originally posted by Bird-The-Word

Seraphine and Skarner...nsfw?


05 Dec

Comment

Originally posted by canPaykek

if u want ill post this every monday night <--

Thanks for posting this :)


17 Sep

Comment

I'm only got to Platinum last year while working full time. Is my job holding me back from my TRUE POTENTIAL (tm)?


30 Jul

Comment

Originally posted by _Trixrforkids_

Rioters are asleep! Quick play Aurelion Sol!

:eyes:


24 May

Comment

Originally posted by mundus108

Out of curiosity, is R used at Riot? Or is it more of a "whichever tool prefered" culture?

We do use R sometimes. For example we want to run a statistical model but the model only exists in R.

Comment

Cool visualizations using R and ggplot!

One thing I'd suggest is a URL library such as this so you don't need piece things together manually.


24 Apr

Comment

Originally posted by alanliang

For my (fellow) nerds, I wanted to address two main things.

1. Defining Plat+ (Classification Accuracy Shrinking Sample Size)

TL;DR 1: Outlined below, we currently use a classification system (A) which we believe prioritizes accuracy, but does lower the sample size pool (ex: Plat+) by a small amount. The old system (B), which most sites use, inflates win rates and sample sizes - plain and simple, it's not acceptable to use a system that says a class (Plat+) has a 55% winrate on average. That would incorrectly mean that most builds have a positive win rate in Plat+. That skews the data way more than sample size. I'd try to remember that size of sample is not the be-all-end-all of accuracy.

Let's use this sample game:

Team 1 (W): Plat 4, Plat 4, Gold 1, Gold 1, Gold 1 | Team 2 (L): Gold 1, Gold 1, Gold 1, Gold 1, Gold 1

As a simple example, here are 2 classification strategies (among many). We currently use Cla...

Read more

Kudos for using Class A instead of Class B. IMO it's the right decision. I've always worried about third party websites not doing their homework and use Class B instead.


25 Mar

Comment

Originally posted by JAZEYEN

Again, I really hate how they've made a habbit of releasing teasers after skins are hitting PBE.

Like these are going to be live in a really soon and we've already seen them fully on PBE. Why do they show the teasers now?

This builds no hype because we already know, it's wasted resources.

I love Riot's hype team!(If that even is a thing, I just assume a portion of their marketing goes to a secret hype team lol) They've just had some bad habbits recently.

Perhaps you are a highly engaged player who actively seeks out new information from the PBE. Maybe you have even played her using the skin on PBE. For you, this teaser probably feels redundant.

That doesn't automatically mean it's wasted resources. Marketing initiatives work, and we have evidence to back it up.


23 Feb

Comment

Originally posted by Blood_Raptor

Maybe if you had the skin, you'd play him more often

I haven't played Amumu for a very long time, but after seeing Inferno Amumu from Your Shop I decided to pull him out... and now I look like an Amumu one trick.


21 Feb

Comment

Originally posted by Mafros99

On another note, can you tell Blau to drop Yasuo? It'd greatly enhance his mmr


21 Jan

Comment

Originally posted by stopfeedingplz

Wait i thought every1 had this. Am i part of a small minority? Why me though?

Yes, only a small proportion of players from selected regions got this new store.

You got it because you were randomly chosen.