|
Herstory Begins Now posted:It makes people feel better, sure, but it's pretty immaterial because if you're good enough to do that, you'll win more games and if you tend to afk in spawn you're going to have a ton of losses. Def makes losses sting less, but elo systems are extraordinarily effective at getting people slotted in more or less exactly where they belong. People feeling better about the ranking system is the whole point, not that it's more accurate. As they're saying, having a more granular system than just pure win/loss ELO feels better for players while achieving the same result faster. Both in how much you win/lose by and individual metrics.
|
#
?
Apr 24, 2020 22:57
|
|
|
|
| # ? May 15, 2024 08:10 |
|
|
I don’t think people feel good about elo systems almost ever regardless of anything they do. As long as they keep it opaque they can tell people whatever they want to hear and it’ll still be more or less equally effective. If you take out individual performance and use pure win/loss you just take like 10% longer to get to the same confidence interval and you can track like 500x less data.
|
|
#
?
Apr 24, 2020 23:06
|
|
|
If a matchmaker has done its job correctly you're going to be pretty miserable so people are always going to be angry about it.
|
|
#
?
Apr 24, 2020 23:10
|
|
|
Herstory Begins Now posted:I don’t think people feel good about elo systems almost ever regardless of anything they do. As long as they keep it opaque they can tell people whatever they want to hear and it’ll still be more or less equally effective. If you take out individual performance and use pure win/loss you just take like 10% longer to get to the same confidence interval and you can track like 500x less data. yes but there's also no dumb playstyles to try and "game" the system. idk, the more I think about it the more I lean towards pure w/l
|
|
#
?
Apr 24, 2020 23:46
|
|
|
Herstory Begins Now posted:If you take out individual performance and use pure win/loss you just take like 10% longer to get to the same confidence interval and you can track like 500x less data. So it's less effective for managing balance and game development and takes longer. Sounds worse!
|
|
#
?
Apr 24, 2020 23:57
|
|
|
not sure if this was posted before but from what I've tested myself it seems accurate. maybe something for in the OP? credit to ClakeyD 
|
|
#
?
Apr 25, 2020 00:03
|
|
|
Only thing missing from it is that bunny hopping does give you a move speed penalty once you land for ~a half second. I've found a couple of bugs lately and posted on the reddit page but they're both pretty weird. Viper screen on Bind A from spawn will sometimes drop a portion of the screen in the sewers. Cypher cam on the same site can see enemies through spawn walls if placed high above sewer exit (they even show on minimap). It seems anywhere the cam can bug out through some geometry it can see enemies where it shouldn't be able to. There's a similar placement on B that let's you see through the big pillar in the middle.
|
|
#
?
Apr 25, 2020 00:07
|
|
|
Got my beta key last night, still can't play because of something to do with Vanguard. Went as far as uninstalling VirtualBox and a couple other services it supposedly flags, still can't play. GG Riot, good game you have here that requires the player to have a separate work computer.
|
|
#
?
Apr 25, 2020 00:37
|
|
|
From what I heard, performance based metrics will only be a thing for lower ranks and the higher your rank the less they will be taken into account until you hit Valorant where it will be pure win/loss. Seems ok, but I tend to agree with Herstory that if you're "hard carrying" you'll win more games and climb faster regardless so its really just feel good fluff. The Elo system is ruthlessly accurate and if you're "stuck" in a rank it's because you belong there.
|
|
#
?
Apr 25, 2020 00:42
|
|
|
ParadeRainer posted:From what I heard, performance based metrics will only be a thing for lower ranks and the higher your rank the less they will be taken into account until you hit Valorant where it will be pure win/loss. Except they explicitly said the margins of victory will also matter at the top rank.
|
|
#
?
Apr 25, 2020 01:30
|
|
|
If someone with a 4.5 KD carries my opponents to a 13-1 victory I want that person to gain as much rating as necessary to keep me from seeing them again. I was about to say "until I get that good" but let's be honest
|
|
#
?
Apr 25, 2020 01:52
|
|
|
I just want a system where I can be a middling player instead of a trash player. I do miss the days where you would just hop on to servers and maybe eat poo poo or maybe be a god.
|
|
#
?
Apr 25, 2020 02:29
|
|
|
Bottom Liner posted:So it's less effective for managing balance and game development and takes longer. Sounds worse! i'm more making a point about attitudes towards elo systems than what is the ~hypothetical optimal~ system that they should implement.
|
|
#
?
Apr 25, 2020 02:50
|
|
|
I want to go back to server lists thanks.
|
|
#
?
Apr 25, 2020 02:50
|
|
|
my hypothetical ~optimal elo ranking system~ involves tiers such as 'dogshit rank' 'garbage-tier' 'try to throw less' 'literally a trashcan' and so on and the game should insult you at the start of each match
|
|
#
?
Apr 25, 2020 02:53
|
|
|
prom candy posted:I just want a system where I can be a middling player instead of a trash player. I do miss the days where you would just hop on to servers and maybe eat poo poo or maybe be a god. Call of Duty already exists Groovelord Neato posted:I want to go back to server lists thanks. Same a thousand times
|
|
#
?
Apr 25, 2020 02:54
|
|
|
Bottom Liner posted:Except they explicitly said the margins of victory will also matter at the top rank. OK, true, I forgot about that part but that is still a team-wide metric vs an individual one, which is fine. flakeloaf posted:If someone with a 4.5 KD carries my opponents to a 13-1 victory I want that person to gain as much rating as necessary to keep me from seeing them again. I was about to say "until I get that good" but let's be honest Well that's how it's going to work until the highest rank, so you're fine 
|
|
#
?
Apr 25, 2020 02:58
|
|
|
usually main sova in this but had some people slampicking him so i decided to try out cypher. seems p loving good? went 32-15-4 in my last match.
|
|
#
?
Apr 25, 2020 04:39
|
|
|
also as an aside i have been playing online games since like 1996 and i have literally never experienced a more toxic player base than the people who play this game lmao
|
|
#
?
Apr 25, 2020 05:14
|
|
|
I find that hard to believe
|
|
#
?
Apr 25, 2020 05:18
|
|
|
A 50S RAYGUN posted:also as an aside i have been playing online games since like 1996 and i have literally never experienced a more toxic player base than the people who play this game lmao what games are you playing?
|
|
#
?
Apr 25, 2020 05:51
|
|
|
Groovelord Neato posted:I want to go back to server lists thanks.
|
|
#
?
Apr 25, 2020 06:03
|
|
|
A 50S RAYGUN posted:also as an aside i have been playing online games since like 1996 and i have literally never experienced a more toxic player base than the people who play this game lmao Liar
|
|
#
?
Apr 25, 2020 06:24
|
|
|
r6siege, destiny 2, overwatch, csgo, etc etc dota 2 comes fairly close but while people are assholes there i think i'm about 50-50 in valorant matches where someone just starts dropping slurs at the first round loss
|
|
#
?
Apr 25, 2020 06:51
|
|
|
Do you not play competitive modes or never are in chat or are deaf or something?
|
|
#
?
Apr 25, 2020 06:54
|
|
|
I've had one of maybe twenty matches where I actually had to mute the pubbies because they were being toxic in voice. I have no idea how some people have managed to get partied up with the depths of the brown sea. Also, play with goons.
|
|
#
?
Apr 25, 2020 06:57
|
|
|
A 50S RAYGUN posted:overwatch found the lie
|
|
#
?
Apr 25, 2020 07:09
|
|
|
If you only played certain modes I could see it.
|
|
#
?
Apr 25, 2020 07:10
|
|
|
Wait, there's a way to gain speed while bunnyhopping? What's the proper technique? Did this person actually time an improvement from getting to point B from point A? Is that just for a short-term over a few hops? Is that something you can do while traversing the map? I know how to bunnyhop in the old half-life engine and can do bunny-hopping in a similar way in Valorant, but my bhops are slower than running in a straight line if I time it over a medium-long distance.
|
|
#
?
Apr 25, 2020 07:38
|
|
|
You move 10% faster in the air as long as you perfectly maintain the bhopping, but again, you move slower (walk speed?) upon initial landing.
|
|
#
?
Apr 25, 2020 07:41
|
|
|
What is "perfectly maintain" the bhopping? I do what works in the goldsource/half-life 1 engine and it doesn't work here. Is it a 10% speed improvement total when moving from point A to point B or is it a 10% speed improvement of the momentum in the air while curving around a lot?
|
|
#
?
Apr 25, 2020 07:50
|
|
|
You can hold W while doing it instead of CS style air strafing, so yes it can get you to a point faster if you maintain. If you don't jump at the right time the slow effect from landing tags you and you lose momentum.
|
|
#
?
Apr 25, 2020 08:00
|
|
|
Is there a video of what correct bhopping over a long distance looks like? Do you have to hold W or does holding W make it more consistent? I want to know if maybe I'm arcing too wide or narrow or whatever. Is there actually acceleration from moving your mouse like in half-life 1 or is there some other engine mechanic in play to bhop? I remember in half-life 1 you get a lot of momentum from just swinging your mouse really fast into the direction you wanted to go to begin a bhop or other movement mechanic. comedyblissoption fucked around with this message at 08:44 on Apr 25, 2020 |
|
#
?
Apr 25, 2020 08:41
|
|
|
I'm fairly sure there's no ms bonus to strafing in valorant. Honestly I was a little surprised by that bit so I'm going to ask him. I'm kinda sorta expecting it has to do with maintaining momentum from other stuff, such as raze's blastpack or jumping down a ramp, but I'm probably wrong
|
|
#
?
Apr 25, 2020 08:48
|
|
|
Lex Neville posted:I'm fairly sure there's no ms bonus to strafing in valorant. Honestly I was a little surprised by that bit so I'm going to ask him. I'm kinda sorta expecting it has to do with maintaining momentum from other stuff, such as raze's blastpack or jumping down a ramp, but I'm probably wrong It used to let you maintain speed over sage's ice but they removed that in the last patch
|
|
#
?
Apr 25, 2020 09:23
|
|
|
I know but that's not what I'm talking about. Bunnyhopping indeed used to let you cross sage's slow field because there was no air speed penalty, but what I meant was that in other fpses you can gain speed by strafejumping - aka jumping left to right, which is why you see typically don't see bunnyhoppers go in a straight line; you absolutely don't want to press w in most of these games - but there's no such ms bonus in valorant so jumping left to right is pointless in this game afaik
|
|
#
?
Apr 25, 2020 10:26
|
|
|
Lex Neville posted:I know but that's not what I'm talking about. Bunnyhopping indeed used to let you cross sage's slow field because there was no air speed penalty, but what I meant was that in other fpses you can gain speed by strafejumping - aka jumping left to right, which is why you see typically don't see bunnyhoppers go in a straight line; you absolutely don't want to press w in most of these games - but there's no such ms bonus in valorant so jumping left to right is pointless in this game afaik You can still circle strafe mid air with A and D to corner peak and stuff, which you can't do with just W.
|
|
#
?
Apr 25, 2020 11:07
|
|
|
fair enough, let me rephrase: strafing does nothing for speed in valorant
|
|
#
?
Apr 25, 2020 11:13
|
|
|
The issue with individual stats affecting Elo gain is that once the exact metrics have been explained or otherwise deduced, it generally creates a much more toxic attitude within players and biases player behaviour to focus less on winning and more on their personal statistics. It encourages baiting, over-saving and so on. It creates a secondary goalpost where you can settle for just maximising your K/D if you think the game can't be won, even if it can be. Even more nuanced statistics measurements like, bombs planted/defused giving a slight bonus means that players fight over who gets to defuse, who gets to take the bomb and so on. In my view, although people crave it, entirely opaque ranking criteria is OK, pure win/loss is better. And transparently rewarding good stats is good for 1-2 people, and sucks for everybody else. Even displaying your KDA in-game publically only really serves to bring misery. In the few games I've played of UNRATED MM in this game, at least one person is flaming the bottom player.
|
|
#
?
Apr 25, 2020 11:53
|
|
|
|
| # ? May 15, 2024 08:10 |
|
|
Jeza posted:In my view, although people crave it, entirely opaque ranking criteria is OK, pure win/loss is better. And transparently rewarding good stats is good for 1-2 people, and sucks for everybody else. Even displaying your KDA in-game publically only really serves to bring misery. Overwatch might be more up your alley. It's pure teamwork with no scoreboard (it didn't even have a killfeed at launch!). I'm not saying this in a snarky way either. Players overwhelmingly want scoreboards and stats to see how they're doing. Yes the tradeoff is toxic players flaming someone not doing well but make liberal use of the mute buttons and your gaming will be much more enjoyable.
|
|
#
?
Apr 25, 2020 12:08
|
|