ImBigW

[ImBigW @ImBig_W]

@BunnyBuggs30595 @mnmcsofgp But after looking at the graph more it's only 1 standard error confidence intervals rather than the typical 95% confidence intervals that would be expected in graphs like this, so it's absolutely misleading and understating the sample size problem.

[ImBigW @ImBig_W]

@BunnyBuggs30595 @mnmcsofgp The point of what they were getting at is we have concepts in statistics that account for sample size. They're called confidence intervals and you can clearly see them on the graph. The takeaways that remain are cis women and trans men under perform trans women and cis men.

[ImBigW @ImBig_W]

@TheMightyBaloon @zac_gaming__ A deep neural network is literally a perfect application for something that is "dynamic in terms of variables"

[ImBigW @ImBig_W]

@TheMightyBaloon @zac_gaming__ Why? Taking an existing concept of WPA used in other sports like baseball and applying it to a novel universe (league of legends) is absolutely a valid math and operations thesis.

[ImBigW @ImBig_W]

@nzsonleo @lvstations You're completely ignoring the time value of money in your assumption

[ImBigW @ImBig_W]

@nttnnk1 @nxwill_ You realize the quote tweet is about mundo e passive right

[ImBigW @ImBig_W]

@CoachlessGG @Manenkos How do you guys go about determint if a sample size is good enough? Do you do bootstrapping to develop confidence intervals for the model, or is it just a set number of games you consider "good enough"

[ImBigW @ImBig_W]

@SpiderFaras @humzhlol @MakkroLoL Brand alone almost has same pickrate as anivia. If you add up all mage bots in masters+ it's about a 21.1% pickrate.

[ImBigW @ImBig_W]

@zizoufan000 @LSXYZ9 @ADAorabble He's said before support is terrible for trying to improve at the game. Improving at any other role will allow you to be at that same level when playing support at a minimum, if not higher.

[ImBigW @ImBig_W]

@notastan2424 @Emocean___ @cocabobi Correct if you are dealing more than 166% of an opponent's hp bar because of healing and shielding new mandate can out damage a single proc of the old one. New mandate damage amp will also be affected by armor (worse!) and mr since they reduce the base damage being amped.

[ImBigW @ImBig_W]

@notastan2424 @Emocean___ @cocabobi Please listen to what I'm saying. If the opponent has 2000 hp and you deal 2000 damage to them to kill them, 6% of 2000 would be 120 damage, or 6% of the enemy's current hp. Old mandate would do 200 damage. It's really not that hard to comprehend.

[ImBigW @ImBig_W]

@notastan2424 @Emocean___ @cocabobi Okay. Now tell me what percent of the enemy's health will the item do if you deal the enemy's entire health bar? 6% of their current health. But in reality the damage amp would contribute to killing them so it would only be 5.66%.

[ImBigW @ImBig_W]

@Emocean___ @cocabobi No, it's terrible. A single proc of old mandate did 10% current health damage and was way easier to proc, the max this can ever do (without massive healing and shieldning) assuming you kill the opponent is 5.66% current health. It sucks.

[ImBigW @ImBig_W]

@FabtonVT @zac_gaming__ Depends entirely on your past game experience. If you're a terran player in SC2 you're gonna have pretty good stutter step micro, potentially better than most league players depending on your SC2 rank.

[ImBigW @ImBig_W]

@Sebastrzian @ExcaIiburrito I'm not sure what you mean. This is standard macro rotations around objectives in high elo and pro play. Teams fight for mid prio and river control and the team that wins the space will maintain banana brush control to prevent entry into the river.

[ImBigW @ImBig_W]

@Sebastrzian @ExcaIiburrito It's not about the teamfight it's about the taking of mid prio and space prior to the teamfight. It's definitely possible to contest mid wave, take poke damage, rotate through river, maintain banana brush control without taking magic damage for 15 seconds to get value.

[ImBigW @ImBig_W]

@Sebastrzian @ExcaIiburrito Depends on team comp but generally if you're laning vs ap as tank you would go hollow radiance and jak'sho is the tank capstone item. For shield regen I'm talking about the ability to take poke damage, exit combat, and get the full 15% hp shield back for free hp.

[ImBigW @ImBig_W]

@Sebastrzian @ExcaIiburrito 1. Correct, that's why I said if you already have hallow radiance or jaksho
2. A difference is a difference and making an intentionally inferior decision because the difference is small makes no sense.
3. True but you're ignoring the fact that rookern shield can regen for value

[ImBigW @ImBig_W]

@ExcaIiburrito It doesn't get worse in absolute terms, it gets worse relative to the eHP of kaenic shield.

Kaenic EHP = 1.15H*(1+(MR+80)/100)
FoN EHP = H*(1+(M+125)/100)
Breakeven M = 120

[ImBigW @ImBig_W]

@ExcaIiburrito But I just told you if you have >= 120 MR kaenic rookern is mathematically better than force of nature, not the other way around. If you already have MR then FoN gets worse.