|
I just always assumed they filtered for that, and looked for games where even the winners gave negative ratings or the losers gave positive ones. Guess that still didn't turn up anything usable.
|
#
?
Jun 1, 2017 20:42
|
|
|
|
| # ? May 22, 2024 08:46 |
|
|
PantsBandit posted:1v1 is still a bad game-mode because 9/10 times the character with the self-heal is the one who's going to win. Also do they really need the intro slowly panning across both characters before every round? If your mechanical skills are good, it's a quick way to get loot boxes. I've lost 1 out of my last 12 1v1 games. And yes, if Hog, Mei, or 76 are a choice, pick one of them. You will almost always win if they pick a hero without self-heal.
|
|
#
?
Jun 1, 2017 20:51
|
|
|
I'm continually amazed that Blizzard, the company that's been in the online gaming biz for literal decades, can still be surprised about stuff like the "rate this match" function winding up nothing more than a way for players to bitch when they lose. You'd think at some point they'd learn something about human psychology even by accident.
|
|
#
?
Jun 1, 2017 20:54
|
|
|
I spent months rating 2CP maps down and KOTH maps up I guess this thought wasn't widespread enough
|
|
#
?
Jun 1, 2017 21:00
|
|
|
Kai Tave posted:I'm continually amazed that Blizzard, the company that's been in the online gaming biz for literal decades, can still be surprised about stuff like the "rate this match" function winding up nothing more than a way for players to bitch when they lose. You'd think at some point they'd learn something about human psychology even by accident. I rarely use it and only to give high marks to a match that was close and hard fought, win or lose. Other than that I ignore it 
|
|
#
?
Jun 1, 2017 21:00
|
|
|
I mean it doesn't help that there's literally no tangible sense of feedback whenever you use one of their systems like this. I've heard people conjure up rumors for months about what the "rate this match" function does, all of which were made up in an effort to assume that it did anything at all instead of throwing a meaningless datapoint into the void. That and "rate this match" without any more depth to the question than that is pretty meaningless from the get-go. Am I rating it down because I lost? Or because the match had someone who spent all game walling people into spawn/teleporting people off cliffs/etc? Or because my team had an attack Bastion that refused to switch and got killed 20 times? Or because someone picked Genji before I did and refused to let me play him? Or etc.
|
|
#
?
Jun 1, 2017 21:24
|
|
|
yeah dota 2 replaced the generic rate question with more specific ones, like bringing up a specific player and going is this fucker totally poo poo y/n
|
|
#
?
Jun 1, 2017 21:45
|
|
|
I mean... yeah it's obvious most people are going to vote down losses and vote up wins. It'd still be interesting to look at the data to observe outliers (those who upvote on a loss, or those or downvote on a win) and try to draw some conclusions based on that. Alas as Blizzard confirms it's just not worth the trouble.
|
|
#
?
Jun 1, 2017 21:45
|
|
|
every time i remembered it was there i usually just voted the middle one as a way to tell them it was a useless feature
|
|
#
?
Jun 1, 2017 21:47
|
|
|
Tiler Kiwi posted:yeah dota 2 replaced the generic rate question with more specific ones, like bringing up a specific player and going is this fucker totally poo poo y/n They also tell you when your report results in a ban which is much more satisfying and better feedback than nothing at all
|
|
#
?
Jun 1, 2017 21:50
|
|
|
klockwerk posted:I mean... yeah it's obvious most people are going to vote down losses and vote up wins. I mean it's not worth the trouble because their methodology is lazy, it's like asking people to take an internet poll to gauge something and the being surprised when only a dozen people respond, 11 of the voting for the jokey answer you put at the end.
|
|
#
?
Jun 1, 2017 21:53
|
|
|
He elaborated on their original plans for the feature https://us.battle.net/forums/en/overwatch/topic/20755445338?page=4#post-73 We were hoping that we could correlate match enjoyment with other data over a prolonged period of time. For example, did people enjoy matches more or less after we instituted the 1-hero limit, or after we rebalanced a certain hero or added a new feature? But ultimately no meaningful analysis came out of the feature beyond what I listed earlier in this post.
|
|
#
?
Jun 1, 2017 22:07
|
|
|
Honestly? If they wanted it to do that they needed to make it more than 1/2/3 stars. To be, 1 is "poo poo match no re plz", 2 is "eh, it was okay i guess" and 3, being the max, is "holy gently caress that was loving AMAZING". With such little gradient between them, how the gently caress were they expecting to get meaningful data on if a change resulted in a better time? If I really enjoy a match and rate it 3, then they change something and I enjoy another match just that little bit more because of it, I literally cannot express this via that rating system. tl:dr it didn't fail just because players are whiny shits who downvote when they lose. It failed because it just couldn't provide meaningful data no matter what, even when the player was rating it in good faith.
|
|
#
?
Jun 1, 2017 22:11
|
|
|
turtlecrunch posted:He elaborated on their original plans for the feature I mean they're also perfectly content to ignore broad swathes of feedback like "hey maybe 35m falloff McCree isn't a great idea" and "hey uh this Bastion rework seems pretty dumb?" before they shove it live and have to go back and undo poo poo so I'm highly skeptical that they would be rushing to usefully heed player-generated feedback curated by clicking a 1-3 star rating.
|
|
#
?
Jun 1, 2017 22:12
|
|
|
blizzard seems very excited about gathering data but not so much about using the data
|
|
#
?
Jun 1, 2017 22:14
|
|
|
Snazzy Frocks posted:blizzard seems very excited about gathering data but not so much about using the data they're selling the data to the government now trump knows you're a hanzo main and he's sending the death squad to your door
|
|
#
?
Jun 1, 2017 22:15
|
|
|
Lol what the gently caress? Maybe have some indication as to the purpose or function of a feature if you want people to use it. Doesn't seem complicated.
|
|
#
?
Jun 1, 2017 22:17
|
|
|
Snazzy Frocks posted:blizzard seems very excited about gathering data but not so much about using the data That's a pretty good way of putting it, yeah. They also do an exceedingly poor job of explaining what prompts them to do the things they do and allowing other people in on the data they're using for it.
|
|
#
?
Jun 1, 2017 22:24
|
|
|
Snazzy Frocks posted:blizzard seems very excited about gathering data but not so much about using the data Companies in general are very bad about making useful or accurate conclusions based on the data they collect, cause you need a very specific kind of nerd to properly collect and interpret it
|
|
#
?
Jun 1, 2017 23:00
|
|
|
I always rated down matches that were stomps, and rated up matches that were well-balanced and went on for a significant amount of time, the exception being 2CP maps since smashing my head against the enemy team for 8 minutes on the second point is really dull. Speaking of which, I wish they'd cap the amount of time you can gain or something. Defending (or even sieging) Anubis B for nine minutes is dumb as hell. Cap it at like 7 minutes or something.
|
|
#
?
Jun 1, 2017 23:08
|
|
|
Kassoon posted:Companies in general are very bad about making useful or accurate conclusions based on the data they collect, cause you need a very specific kind of nerd to properly collect and interpret it Pop that poo poo into a computer program that can run an ANOVA and then show a room full of people the pretty graphs it made so they'll make duplicates from loot boxes 75% of the value rather than 20%.
|
|
#
?
Jun 1, 2017 23:09
|
|
|
SKULL.GIF posted:Speaking of which, I wish they'd cap the amount of time you can gain or something. Defending (or even sieging) Anubis B for nine minutes is dumb as hell. Cap it at like 7 minutes or something. They can't do this or it would create disincentives to win quickly, which would be terrible.
|
|
#
?
Jun 1, 2017 23:13
|
|
|
I don't think there is a significant reason to change lot box rates given the propensity of whales to buy boxes. I don't even think the same team that cares about balance is on the same planet as the team that monetizes the playerbase
|
|
#
?
Jun 1, 2017 23:25
|
|
|
HATECUBE posted:I don't even think the same team that cares about balance is on the same planet as the team that monetizes the playerbase They both enjoy ignoring feedback in favor of their "vision."
|
|
#
?
Jun 1, 2017 23:34
|
|
|
Slime posted:they're selling the data to the government Well, yeah; you slam picked before he could. What'd you expect?
|
|
#
?
Jun 1, 2017 23:36
|
|
|
Snazzy Frocks posted:blizzard seems very excited about gathering data but not so much about using the data
|
|
#
?
Jun 1, 2017 23:57
|
|
|
Tuxedo Catfish posted:They can't do this or it would create disincentives to win quickly, which would be terrible. I can't envision a scenario where a team that's able to capture the first point quickly would go "Well, we'll back off and let them respawn and regroup" and it actually being advantageous even if the max time was capped. Like, how does this work? My team clears them off Anubis A, there's 4 minutes remaining and if we cap now the time remaining is 7:30 (5 minutes gain being capped to the lower maximum) -- or we can clear them off, then hang around waiting for no reason, then cap at 2:30 remaining and then get to assault Anubis B with 7:30? That's advantageous?
|
|
#
?
Jun 2, 2017 00:08
|
|
|
ANGRYGREEK posted:Hahaa, get hosed Roadhog. that won't work anymore after ptr. new hog does max 300 damage assuming a perfect headshot, so getting another hog down in 2 shots is gonna be really rare. really good roadhogs will still be good, it's not like you can't hit gm as junkrat or other trash, but they'll definitely be worse and get a lot less kills off of the hook. yes, even if they shoot the guy first, there's a decent chance that those shots won't kill the target
|
|
#
?
Jun 2, 2017 00:15
|
|
|
Attackers are crushing defenders on Anubis- using your example. They can then spawn camp until Max time-4 minutes (5 minutes in Quickplay/arcade), until they cap and not lose any effective time. Then they proceed to cap point B quickly. This creates a negative play experience for the defenders. Worse perhaps, than having to try and defend an extra minute. E: It also inherently discourages aggressive play, and favourably encourages poke at the choke to build up a bunch of ults. After all, you're not losing any time for that two minutes of game. berenzen fucked around with this message at 00:26 on Jun 2, 2017 |
|
#
?
Jun 2, 2017 00:22
|
|
|
SKULL.GIF posted:I can't envision a scenario where a team that's able to capture the first point quickly would go "Well, we'll back off and let them respawn and regroup" and it actually being advantageous even if the max time was capped. No, but you should be rewarded for a fast capture. You don't want to spend more than seven minutes defending point B, then defend point A properly. berenzen posted:E: It also inherently discourages aggressive play, and favourably encourages poke at the choke to build up a bunch of ults. After all, you're not losing any time for that two minutes of game. Pretty much. Tuxedo Catfish fucked around with this message at 00:27 on Jun 2, 2017 |
|
#
?
Jun 2, 2017 00:24
|
|
|
I could see attackers clearing the point but then pussyfooting off and on the point to try and bait the attackers back to point A or go for picks or build ult or something.
|
|
#
?
Jun 2, 2017 00:26
|
|
|
Keeshhound posted:Well, yeah; you slam picked before he could. What'd you expect? Trump is the ultimate Hanjo?
|
|
#
?
Jun 2, 2017 00:35
|
|
|
Kuros posted:Trump is the ultimate Hanjo? Nah, he's throwing the match really hard, but Hanjo was never obnoxious. Just a simple "I Hanjo." every so often. What I'm saying is Hanjo would have won.
|
|
#
?
Jun 2, 2017 00:48
|
|
|
Kuros posted:Trump is the ultimate Hanjo? Once upon a time the election was just for humor purposes https://www.businessinsider.com/donald-trump-mains-hanzo-billboard-overwatch-2016-10 http://www.trumpisnotateamplayer.com
|
|
#
?
Jun 2, 2017 00:50
|
|
|
resistentialism posted:I could see attackers clearing the point but then pussyfooting off and on the point to try and bait the attackers back to point A or go for picks or build ult or something. they can do this right now (they don't)
|
|
#
?
Jun 2, 2017 01:13
|
|
|
You can, but without that change there's one extra major disincentive to doing so: you're wasting time.
|
|
#
?
Jun 2, 2017 01:18
|
|
|
It's mostly that it'd feel bad as an attacker if they changed it now. You feel like you earned that time and taking it away would be as popular as adding sudden death back in imo
|
|
#
?
Jun 2, 2017 01:23
|
|
|
How does Overwatch calculate SR? Completed my matches today, won 8 games, lost 2, had at least 1 gold medal every round, and it put me at 1322.
|
|
#
?
Jun 2, 2017 02:16
|
|
|
10 Beers posted:How does Overwatch calculate SR? Completed my matches today, won 8 games, lost 2, had at least 1 gold medal every round, and it put me at 1322. According to resident conspiracy theorists, somewhat on your average performance across all game modes, but mostly on your last season's performance and your performance within the ranked game itself. It typically places you lower than you "belong" at first so you can climb back up. If the game "thinks" you need to be higher, you'll typically lose less and gain more from each loss and win respectively.
|
|
#
?
Jun 2, 2017 02:21
|
|
|
|
| # ? May 22, 2024 08:46 |
|
|
Your performance is compared to other people playing the same character in competitive.
|
|
#
?
Jun 2, 2017 02:21
|
|