I was Nano'd as Lucio the other day and boy was I confused.
Nano'd lucio is a time honored tradition and your opportunity to go maximum reddit lucio. Healing is for cowards, boops for the boop god!
A trap is for fish: when you've got the fish, you can forget the trap. A snare is for rabbits: when you've got the rabbit, you can forget the snare. Words are for meaning: when you've got the meaning, you can forget the words.
+2
Options
miscellaneousinsanitygrass grows, birds fly, sun shines,and brother, i hurt peopleRegistered Userregular
miscellaneousinsanitygrass grows, birds fly, sun shines,and brother, i hurt peopleRegistered Userregular
the base hero sprays have only gotten consistently better and better with the release of every new character, i wish i had like 10 spray slots for hammond
While Dallas looked bad, SF also looked legit excellent, especially on goats. I think Dallas will be ok, though it might take them a couple of weeks.
A trap is for fish: when you've got the fish, you can forget the trap. A snare is for rabbits: when you've got the rabbit, you can forget the snare. Words are for meaning: when you've got the meaning, you can forget the words.
0
Options
TexiKenDammit!That fish really got me!Registered Userregular
Overwatch League has been hype as fuck all day so far.
I really don't feel that with this mirror goats comp on nearly everything. It's not fun and results in way too much just stomps and resets and you can tell the audience doesn't care for it either, it's why they get so excited when Bastion or Torbjorn show up (and Houston today did an old school Widow Pharmercy attack on Volskaya A today which got a good response). Linkzr on Widow or McCree, yes. Linkzr on Bridget, come on.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Overwatch League has been hype as fuck all day so far.
I really don't feel that with this mirror goats comp on nearly everything. It's not fun and results in way too much just stomps and resets and you can tell the audience doesn't care for it either, it's why they get so excited when Bastion or Torbjorn show up (and Houston today did an old school Widow Pharmercy attack on Volskaya A today which got a good response). Linkzr on Widow or McCree, yes. Linkzr on Bridget, come on.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Yeah Goats is still prevalent, but there are a good few dps-Goats variants and total other comps hiding in the bushes. It's not ideal but I want to be hype! Also, I think we all realise now that Goats is really the symptom of a few ingrained design features within the game. I used to be like "the problem is DVa is a dps! No, Brig is the problem! No, it's aoe healing!"
It's really a tricky combination of all that and more. And I'm glad Blizz is addressing it at a slow jogging pace, because fixing ingrained design issues in full-on panic mode is pretty dangerous for your videogame andyour spectator sport. For now, some more variety than last season's Goatsathon!
I love that biggoose is the one lucio who's able to make reddit-lucio'ing work in OWL
He's just bouncing all the fuck over the place, constant boops
Kana on
A trap is for fish: when you've got the fish, you can forget the trap. A snare is for rabbits: when you've got the rabbit, you can forget the snare. Words are for meaning: when you've got the meaning, you can forget the words.
Overwatch League has been hype as fuck all day so far.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
Overwatch League has been hype as fuck all day so far.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
Edit: Just realised you might have been asking for that specific statement by Jeff, which @TexiKen linked below. So, uh, here's a wall of text in a spoiler box?
It's never been outright proved or admitted by Blizard. It's more the end in a chain of reasoning beginning with statements Blizz has made about their systems. Each step in the reasoning is probably true, but we can't be sure. I dunno, but here it is:
1) Performance Based SR grants you more SR - or takes less SR away - based on certain unknown metrics that Blizzard tracks, comparing those numbers and metrics to other players of the same hero at your rank. This is known and true.
2) Therefore, if you post bigger numbers on a hero, you will gain more SR for a win and lose less for a loss. This is basically true, though we don't know for sure what numbers Blizzard tracks. Though I saw an interesting mathematical breakdown on Winston's Lab IIRC that showed which stats had greater correlation to PBSR bonuses. tl;dr: it's probably at least somewhat hero specific: it's not just pure damage and elims and healing output.
3) The best way to maximise your numbers output is probably to one-trick or main a particular hero. Learn all their mechanics, playstyle and little tricks completely, and you'll do slightly more damage, take slightly less damage, heal more, block more, kill more, etc, than casual flexers of the hero that you main. This means that when you win you'll get 26 SR where you might have gotten 24, and if you lost you lose only 22 instead of 24, f'rex.
4) Therefore, by making the value of winning versus losing less a 1:1 and more like, say, 1.1:0.9, you will (probably?) rise higher and stay higher than your pure winrate deserves.
5) Also, switching off one hero to another that you might be less competent on and might only play for a couple of minutes for a specific purpose...well that means you'll post lower numbers per minute, so if you win you'll gain less SR than if you stayed on your main, and if you win you'll lose more SR than if you stayed on your main. This apparently causes people to stay on their main even when switching would increase the likelihood of victory, because they plan to mitigate the SR loss and then make it back and then some with a victory later.
This was most evident with the old mass Rez Mercy back when PBSR existed even up to the highest ranks: various Mercy one-tricks would play for big team rezzes rather than possibly more valuable single rezzes, and rose to the highest ranks - where they just didn't belong skill-wise - through PBSR valuing mass Rezzes over winning the game. Personally, I've never not swapped because of PBSR concerns, or tried to maximise my numbers for their own sake. Not because I'm some saint, it just never occurs to me. Maybe others do? I don't know.
IMO, PBSR probably does disrupt matches somewhat, by fooling the system into creating matches where the likelihood of victory isn't 50/50: ELO systems are meant to match players as if their SR represents a 50% winrate at that skill level, but when their SR actually represents, say, a 48% winrate, then they are a net drag on their team - and they are rewarded for it with a higher SR. That's a pretty ass-backward incentive.
That said, Blizzard hasn't opened the hood on their SR and matchmaking systems so much of this is conjecture and theory. We don't know if and how they account for any of this. But something is going on, and I know this because I recently saw a chart of winrates in different skill brackets. Winrates without exception rise as the player's SR rises, except at the line between Plat and Diamond, where it dips about a percent, then returns to rising steadily again. That's the line where both PBSR and SR decay cut off. And anecdotally, plenty of players have noted that high Plat / low Diamond is the biggest shitshow in the whole SR range, including Jayne who noted that rising his smurf through that range took longer than all of the rest of Diamond and Masters. Something is going on. But we don't know what it is.
Overwatch League has been hype as fuck all day so far.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
It was in this video (around 5:10 is when it's brought up but it's in response to Jeff saying at 3:28 he wasn't expecting people to main characters or one trick in this game):
and this is the most recent blizzard forum thing I can find on it on a quick search, it seems to ultimately come down to being judged by other players using that hero on a whole match as opposed to doing say a round as Junkrat and then a round as Mei, as well as how you use the ult per hero too.
Overwatch League has been hype as fuck all day so far.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
It's never been outright stated, it's more the end in a chain of reasoning that while probably true at each step, might not be true in total. I dunno, but here it is:
1) Performance Based SR grants you more SR - or takes less SR away - based on certain unknown metrics that Blizzard tracks, comparing those numbers and metrics to other players of the same hero at your rank. This is known and true.
Overwatch League has been hype as fuck all day so far.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
...it seems to ultimately come down to being judged by other players using that hero on a whole match as opposed to doing say a round as Junkrat and then a round as Mei, as well as how you use the ult per hero too.
Yep.
'Chance, you are the best kind of whore.' -Henroid
0
Options
DragkoniasThat Guy Who Does StuffYou Know, There. Registered Userregular
Yeah...I play a variety of different characters cause I get bored one-tricking.
But even then I don't really do attack/defend switches its more "I want to play this character this game".
Well then no, you don't get less for switching characters. You get less if you switch to a character you're not as good at, and then perform worse on. And at least the way I understand it, your stats are measured as a per-minute number, not based on the entire match (the same way winrates are calculated, in other words).
Like if I'm doing bad on my main and switch to someone else and start doing better, I'm not going to get less SR than if I had stayed on my main and continued doing bad (unless my "bad" on character 1 is still better than average and my "good" on character 2 is still worse, which still makes sense from an SR perspective). It's not just a straight "oh you changed your character so you're gonna get less SR now no matter what". The SR system definitely has issues, but I don't believe this is one of them.
The last week Hamster just... kind of clicked with me? I had been playing him too similarly to D.Va, but he’s not really a dive/distraction/peeler tank so much as a disruption one. I think where he really excels is at knocking the enemy team out of position and opening them up to elimination by teammates, rather than eliminating them himself. In that regard he’s similar to Winston, where if your team sucks you aren’t doing shit with him.
The last week Hamster just... kind of clicked with me? I had been playing him too similarly to D.Va, but he’s not really a dive/distraction/peeler tank so much as a disruption one. I think where he really excels is at knocking the enemy team out of position and opening them up to elimination by teammates, rather than eliminating them himself. In that regard he’s similar to Winston, where if your team sucks you aren’t doing shit with him.
yes!, or maybe as a single hunter in certain instances. Like i can chase down a single hanzo or widow if no one on my team can get the job done, but i'm much better at opening up holes and disrupting healers so the rest of my team can rack up the kills.
Overwatch League has been hype as fuck all day so far.
i hope dallas get their shit together enough for their series tonight, last season's fuel v dynasty games were all very entertaining
Dallas looked like a completely different team against Seoul. AKM popped off, the support line coordinated well, OGE did OGE things, and Zacharee didn’t feed the entire time.
Also Effect smiled. There are very few bad teams this year. It feels like most teams have a fighting chance in any given match. Plus Atlanta might be actually really good; they pushed Philly to the absolute brink and might have won had Carpe not gone completely insane.
+1
Options
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
edited February 2019
Man I thought the only tradeoff for swapping heroes mid-game was losing ult charge. It seems really weird they would take hero swapping into account in calculating SR. Shouldn't it just be if you won or not? Why complicate things?
Edit: Though now this somewhat explains behaviors I've seen in silver comp. For example, the ludicrous degree to which people prioritize getting kills over the actual objective / winning.
Monkey Ball Warrior on
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
Edit: Though now this somewhat explain behaviors I've seen. For example, the ludicrous degree to which people prioritize getting kills over the actual objective / winning.
No that's just the way people are, I guarantee you they are not thinking about the marginal effects of an opaque system
Man I thought the only tradeoff for swapping heroes mid-game was losing ult charge. It seems really weird they would take hero swapping into account in calculating SR. Shouldn't it just be if you won or not? Why complicate things?
Edit: Though now this somewhat explains behaviors I've seen in silver comp. For example, the ludicrous degree to which people prioritize getting kills over the actual objective / winning.
Well that might also explain why those people are still in silver.
+3
Options
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
I've seriously been chased half way across the map. By tanks. Tanks that really should have let me go and stayed with the cart. I mean, congrats, you just wasted 20 seconds chasing me around the map to force me into a 10 second respawn, meanwhile the entire rest of my team is pushing the cart into the checkpoint, and your team can do nothing to stop them because they are short a tank. Thanks!
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
Man I thought the only tradeoff for swapping heroes mid-game was losing ult charge. It seems really weird they would take hero swapping into account in calculating SR. Shouldn't it just be if you won or not? Why complicate things?
Again, you're not losing SR just for switching heroes. You're losing SR if you swap to a hero and underperform on that hero. Also, once you hit diamond, the performance aspect is removed so it does only matter if you won or not. They use performance stats at lower levels presumably so that if you really are better than your rank and deserve to be higher, your personal performance will help raise you up out of that rank quicker.
0
Options
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
They use performance stats at lower levels presumably so that if you really are better than your rank and deserve to be higher, your personal performance will help raise you up out of that rank quicker.
I don't think that's a reasonable policy. Law of Large Numbers means if you play enough games, you will eventually move to where you should be. Anything that motivates anyone to do anything other than win has the potential to mess things up.
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
They use performance stats at lower levels presumably so that if you really are better than your rank and deserve to be higher, your personal performance will help raise you up out of that rank quicker.
I don't think that's a reasonable policy. Law of Large Numbers means if you play enough games, you will eventually move to where you should be. Anything that motivates anyone to do anything other than win has the potential to mess things up.
Yeah that law has me bounced between 2.9k all the way down to 2.2k, it's a principle with vast room for improvement
Man I thought the only tradeoff for swapping heroes mid-game was losing ult charge. It seems really weird they would take hero swapping into account in calculating SR. Shouldn't it just be if you won or not? Why complicate things?
Again, you're not losing SR just for switching heroes. You're losing SR if you swap to a hero and underperform on that hero. Also, once you hit diamond, the performance aspect is removed so it does only matter if you won or not. They use performance stats at lower levels presumably so that if you really are better than your rank and deserve to be higher, your personal performance will help raise you up out of that rank quicker.
It's a nice theory, but in reality I think they made the cutoff at Diamond because the hardcore players were bitching the most about PBSR, but Blizz couldn't even admit to themselves that the system was fundamentally flawed so they catered to the vocal minority only. Then the issue died down because you can't expect the minority to get furious about all those OTHER people stuck with a shitty system.
It's a strong strategy, used often by many political actors.
Posts
https://youtu.be/oVbZsu80QpY
i have no idea where that zen even was, apparently my brain did tho
boo-oooostioooooo
CANNOT STOP
WILL NOT STOP
GAS GAS GAS, I'm gonna step on the gas!
Nano'd lucio is a time honored tradition and your opportunity to go maximum reddit lucio. Healing is for cowards, boops for the boop god!
i hope dallas get their shit together enough for their series tonight, last season's fuel v dynasty games were all very entertaining
I really don't feel that with this mirror goats comp on nearly everything. It's not fun and results in way too much just stomps and resets and you can tell the audience doesn't care for it either, it's why they get so excited when Bastion or Torbjorn show up (and Houston today did an old school Widow Pharmercy attack on Volskaya A today which got a good response). Linkzr on Widow or McCree, yes. Linkzr on Bridget, come on.
Also TIL that if you switch heroes in game you lose more SR with a loss and get less SR if you win, way to talk out of both sides of your mouth Blizzard. And Jeff is puzzled by people one tricking.
Yeah Goats is still prevalent, but there are a good few dps-Goats variants and total other comps hiding in the bushes. It's not ideal but I want to be hype! Also, I think we all realise now that Goats is really the symptom of a few ingrained design features within the game. I used to be like "the problem is DVa is a dps! No, Brig is the problem! No, it's aoe healing!"
It's really a tricky combination of all that and more. And I'm glad Blizz is addressing it at a slow jogging pace, because fixing ingrained design issues in full-on panic mode is pretty dangerous for your videogame andyour spectator sport. For now, some more variety than last season's Goatsathon!
hAmmONd IsnT A mAin TAnk
I'm also going to get dinner. You can't get dinner with me.
I bought that spray the day she came out, with actual credit money
No regrets, best spray
He's just bouncing all the fuck over the place, constant boops
Interested where you heard this because I've never heard anything like that and it definitely doesn't match up with my experience.
Edit: Just realised you might have been asking for that specific statement by Jeff, which @TexiKen linked below. So, uh, here's a wall of text in a spoiler box?
1) Performance Based SR grants you more SR - or takes less SR away - based on certain unknown metrics that Blizzard tracks, comparing those numbers and metrics to other players of the same hero at your rank. This is known and true.
2) Therefore, if you post bigger numbers on a hero, you will gain more SR for a win and lose less for a loss. This is basically true, though we don't know for sure what numbers Blizzard tracks. Though I saw an interesting mathematical breakdown on Winston's Lab IIRC that showed which stats had greater correlation to PBSR bonuses. tl;dr: it's probably at least somewhat hero specific: it's not just pure damage and elims and healing output.
3) The best way to maximise your numbers output is probably to one-trick or main a particular hero. Learn all their mechanics, playstyle and little tricks completely, and you'll do slightly more damage, take slightly less damage, heal more, block more, kill more, etc, than casual flexers of the hero that you main. This means that when you win you'll get 26 SR where you might have gotten 24, and if you lost you lose only 22 instead of 24, f'rex.
4) Therefore, by making the value of winning versus losing less a 1:1 and more like, say, 1.1:0.9, you will (probably?) rise higher and stay higher than your pure winrate deserves.
5) Also, switching off one hero to another that you might be less competent on and might only play for a couple of minutes for a specific purpose...well that means you'll post lower numbers per minute, so if you win you'll gain less SR than if you stayed on your main, and if you win you'll lose more SR than if you stayed on your main. This apparently causes people to stay on their main even when switching would increase the likelihood of victory, because they plan to mitigate the SR loss and then make it back and then some with a victory later.
This was most evident with the old mass Rez Mercy back when PBSR existed even up to the highest ranks: various Mercy one-tricks would play for big team rezzes rather than possibly more valuable single rezzes, and rose to the highest ranks - where they just didn't belong skill-wise - through PBSR valuing mass Rezzes over winning the game. Personally, I've never not swapped because of PBSR concerns, or tried to maximise my numbers for their own sake. Not because I'm some saint, it just never occurs to me. Maybe others do? I don't know.
IMO, PBSR probably does disrupt matches somewhat, by fooling the system into creating matches where the likelihood of victory isn't 50/50: ELO systems are meant to match players as if their SR represents a 50% winrate at that skill level, but when their SR actually represents, say, a 48% winrate, then they are a net drag on their team - and they are rewarded for it with a higher SR. That's a pretty ass-backward incentive.
That said, Blizzard hasn't opened the hood on their SR and matchmaking systems so much of this is conjecture and theory. We don't know if and how they account for any of this. But something is going on, and I know this because I recently saw a chart of winrates in different skill brackets. Winrates without exception rise as the player's SR rises, except at the line between Plat and Diamond, where it dips about a percent, then returns to rising steadily again. That's the line where both PBSR and SR decay cut off. And anecdotally, plenty of players have noted that high Plat / low Diamond is the biggest shitshow in the whole SR range, including Jayne who noted that rising his smurf through that range took longer than all of the rest of Diamond and Masters. Something is going on. But we don't know what it is.
hAmmONd IsnT A mAin TAnk
It was in this video (around 5:10 is when it's brought up but it's in response to Jeff saying at 3:28 he wasn't expecting people to main characters or one trick in this game):
https://www.youtube.com/watch?v=6O-eN-umIsw
and this is the most recent blizzard forum thing I can find on it on a quick search, it seems to ultimately come down to being judged by other players using that hero on a whole match as opposed to doing say a round as Junkrat and then a round as Mei, as well as how you use the ult per hero too.
Yep.
Yep.
But even then I don't really do attack/defend switches its more "I want to play this character this game".
Like if I'm doing bad on my main and switch to someone else and start doing better, I'm not going to get less SR than if I had stayed on my main and continued doing bad (unless my "bad" on character 1 is still better than average and my "good" on character 2 is still worse, which still makes sense from an SR perspective). It's not just a straight "oh you changed your character so you're gonna get less SR now no matter what". The SR system definitely has issues, but I don't believe this is one of them.
yes!, or maybe as a single hunter in certain instances. Like i can chase down a single hanzo or widow if no one on my team can get the job done, but i'm much better at opening up holes and disrupting healers so the rest of my team can rack up the kills.
Blizzard: Pailryder#1101
GoG: https://www.gog.com/u/pailryder
Dallas looked like a completely different team against Seoul. AKM popped off, the support line coordinated well, OGE did OGE things, and Zacharee didn’t feed the entire time.
Also Effect smiled. There are very few bad teams this year. It feels like most teams have a fighting chance in any given match. Plus Atlanta might be actually really good; they pushed Philly to the absolute brink and might have won had Carpe not gone completely insane.
Edit: Though now this somewhat explains behaviors I've seen in silver comp. For example, the ludicrous degree to which people prioritize getting kills over the actual objective / winning.
Well that might also explain why those people are still in silver.
Again, you're not losing SR just for switching heroes. You're losing SR if you swap to a hero and underperform on that hero. Also, once you hit diamond, the performance aspect is removed so it does only matter if you won or not. They use performance stats at lower levels presumably so that if you really are better than your rank and deserve to be higher, your personal performance will help raise you up out of that rank quicker.
I don't think that's a reasonable policy. Law of Large Numbers means if you play enough games, you will eventually move to where you should be. Anything that motivates anyone to do anything other than win has the potential to mess things up.
i've got the credits for two, but i'll put those towards the archives event, those skins are always good
i still need zarya's from last year but i'm content to wait for the next go round
Yeah that law has me bounced between 2.9k all the way down to 2.2k, it's a principle with vast room for improvement
It's a nice theory, but in reality I think they made the cutoff at Diamond because the hardcore players were bitching the most about PBSR, but Blizz couldn't even admit to themselves that the system was fundamentally flawed so they catered to the vocal minority only. Then the issue died down because you can't expect the minority to get furious about all those OTHER people stuck with a shitty system.
It's a strong strategy, used often by many political actors.
hAmmONd IsnT A mAin TAnk