Antiwhippy
the holder of the trombone
Against or with?
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Overwatch |OT4| You Want A Good Genji, But You Need The Bad Hanzo
- Thread starter GraveRobberX
- Start date
- Status
Against or with?
Or this as well.
Basically it's not giving you XP. It's adjusting your rank based on a multitude of factors.
If you didn't play significantly better than before you can't expect to be getting big changes to your rank just from a victory.
he's not a healer as far as i'm concerned.
68 is damn good.
But still, they already said they will revamp the system. I don't fucking care about it representing a bell curve, that just means that 10 levels are an ocean of competence. Also, there are 100 ranks. It should spread out the ranks properly, I mean effing Seagull is 79 and that's the first percentile. So how about making it flat precentile so his rank is 99. I am not about gains and losses, I feel it's just hard to judge how good or bad in respect to the rest of the community you are.
Meanwhile I played with friends and knocked to 74, I guess I'll see if I'm still ranking one a match, if I am I'm curious where that'll slow down finally.
This
Basically people need to stop looking at winning and losing as the cause of their rank ups. If you play at the top 1% of players at 67 and solo queue and even only win 40%, you'll still be ranking up, just a lot slower. Grabbing good friends just speeds the process up since your wins just get your rank ups per match and your losses are less frequent.
It's a fairly self explanatory system once you realize it's percentile based, people just need to get out of the mindset that winning/losing is what determines their rank.
You also don't want this though, because the top 1% is a huge difference, bigger than ever before. They'd need separate indicators for top ranks, similar to Rocket League having a variety of ranks, then 1% being very wide between .2-.4, .4-.8, etc.
Because a low top 1% vs a high 1% is basically a 50 vs a 70 atm
This
Basically people need to stop looking at winning and losing as the cause of their rank ups. If you play at the top 1% of players at 67 and solo queue and even only win 40%, you'll still be ranking up, just a lot slower. Grabbing good friends just speeds the process up since your wins just get your rank ups per match and your losses are less frequent.
It's a fairly self explanatory system once you realize it's percentile based, people just need to get out of the mindset that winning/losing is what determines their rank.
You also don't want this though, because the top 1% is a huge difference, bigger than ever before. They'd need separate indicators for top ranks, similar to Rocket League having a variety of ranks, then 1% being very wide between .2-.4, .4-.8, etc.
Because a low top 1% vs a high 1% is basically a 50 vs a 70 atm
Antiwhippy
the holder of the trombone
I've actually had fairly good luck with having ranked players going another healer with me when I go zen.
But if I'm the only healer yeah I'll go lucio.
You should try solo que some time and see what a fulfilling ranking system this is.
Antiwhippy
the holder of the trombone
Like maybe it's the rank I'm in but most of my rank gain has been solo queue honestly.
The same concept applies regardless of queue type. Theres a reason going 150-150 can end you in the 20s or 70s
Ghost is Bad
Banned
i think its fun dropping 40-50 bombs every comp game at the rank im at lol
everyones an idiot with 0 idea of how to play at 49 lol.
I need a citation for this I think.
Check top players winrates, not all of them are actually much above 50, showing it's possible
The 28 on reddit who acknowledges he's in the bottom 1% and wanted help has a 49% winrate.
There you go for your two people, once again backing this "performance into top percentile leads to 70+ trumps constant wins outside your skill level" concept. I just want it to be accepted or reasonably countered, except we have no other way to explain anything that goes on in a percentile based rank system, since it perfectly explains it.
edit: see an 80 with a 55% winrate http://masteroverwatch.com/profile/psn/global/Limo999
https://www.overbuff.com/rankings
Of the top 100 players, a scant 7 of them have win rates under 60%. Of the ones that do, 3 of them have over 600 games played. Of course, if you're in the top 1 percentile of players playing against other top 1 percentile of players for a long enough time, your win rate will inexorably creep towards 50% unless you're much higher than the system will concievably put you at.
Actually I'm not sure what is being discussed here. Are you arguing that the game measures your performance more than it does your wins/losses? Are you factoring in different the initial calibrations? I don't think it's impossible they're doing this but this kind of goes against the way MMR is usually applied to team-centric games.
Yes, because most players aren't in the top 100 if they're not able to crush everyone heading up there, was moreso using it as an example. I'm showing that it's possible, we've actually somewhat tested with my group with having large skill gaps and playing a whole night causing some people to end up a rank or two higher with others one down, even though we played the same matches on the same teams. W/L isn't the leading factor in the slightest.
MMR is percentile based on an individual level from what we can tell- it does this to wall you out from playing with people vastly out of our rank, even if it's team-centric, they'd rather have you play with a team your rank for comp instead of going with someone 20 below.
It's a weird system. But yes, that's why some of us went on crazy losing streaks with friends but then went and gained a full rank every match regardless. This is also why you can play a match with a friend and gain 75% of a rank and he only gains 20%.
I've done nothing but solo since the first week.
Well just try it out, and share your feelings after a week of it.
I'm not sure I like this. I thought it was strange that I was like a lot of other people around the 50s who were losing rank by a lot and winning by a little, but after the Zen buffs I played him a few times in Ranked and when I won I'd get a full rank, which was really surprising. If your theory is correct, it's because the matchmaker sees me as outperforming Zens of my level, but this is likely because Zen was kind of trash before the patch. Anyone who picked him in the 50th percentile bracket would've fed relentlessly and contributed nothing, which means my Zen performance is being inflated.
I'm not complaining about ez ranks but this is just rife for abuse if it's actually what's happening.
I'm not complaining about ez ranks but this is just rife for abuse if it's actually what's happening.
Why though
It's making the game harder
I do well with my team because I can trust them to heal me and they can trust me to teamwipe/give them pushes. This leads to all of us performing better, leading to full rank ups a match even in the 70s and under half a rank a loss. Why would I toss this out the window to intentionally play worse than I can?
*I also never practice OW/play it without them unless Lik wants to play some QP so I can practice my other stuff now. I play JRPGs.
I'm almost positive it's correct, I've personally shot down about a dozen theories, from streaks to under/overdog being the leading factor, to just w/l, to so on. Commonly you still hear 50s say "oh, as you get higher it intentionally shoots you lower and you need a lot more wins!", but I've seen the exact opposite, with my chart actually going exponentially up as I got better/higher. Stuff like that is innacurate and still said constantly.
C9 is a good control group for this experiment with each of them placing differently at the start and watching their ranks as they got up. There was an article on this rank system being weird that basically confirmed this train of thought and talked about the ways such a system could be flawed in theory regardless.
But yes this implies that being top 10% with 5 characters is probably not as good as being top 1% with 1, which makes sense, because if you're the best lucio in town, you'd never need to switch- you're helping your team win more than you'd ever be in the top 10%. Become a good zen if you want, zens are the best <3
You need to group up.
I thought it was always spectate
5-14 (and one of those wins was because someone quit on the other team!) in my last 2 competitive days, think I'm just done for the season. Just lost another game in SD with 4 gold medals, literally no clue how I could do any better than I'm doing in most games. I was mainly playing for points at this point for weapons, but holy fuck, I got one win each in two of my last sessions, complete waste of time, lol.
We had a game earlier on Route 66 where all but one person on the other team went Genji on attack; so 5 of us switched to Winston. One guy figured it out, but the rest of them stubbornly stuck to Genji then started talking shit after they lost. It was quite an odd match.
It's spectate by default, but if someone drops from a team, even the other team, it will fill the slot with a spectator since they're already in the game anyway.
yeah he joined before I could think to invite him to group. the kicker- the very next match it still puts him on opposite team. I thought it would auto put him on my team
Normally all my friends know to join the group, etc but this one didnt know
If you select Join Game without grouping up then it will put you in as a spectator who will then fill whichever space opens up next, regardless of which side it is.
On PC I think they can spectate, but as soon as a spot opens the person gets put on a team.
Been hanging out with some GAFfers and doing quick play with them's kinda been the way to go. Fairly low pressure, the company's jovial, and we've been running into more good teams and tenser matches in the past few days. Not even remotely thinking about comp at this point.
Feeling content
aaaaaaaaaand occasionally this.
Antiwhippy
the holder of the trombone
Also if you want to prolong OT tracer is way better and more annoying at it.
RIP double tracer+lucio+winston
I'm curious what the longest OT was, I'm sure on Lijiang Tower's garden zone you could do it for 10m easily on both sides, mainly because double lucio+tracer+winston will never all wipe togther.
Antiwhippy
the holder of the trombone
Pretty sure when 0HL and the old OT rules were used pro games could stretch OT to 10min+ just from tracer/dva spam.
Yeah, this is why the rules were implemented. I've heard crazy stories that a team could go infinite if they use lucio+tracer+winston doubles, I fully believe it.
Sometimes it feels like the game tosses me into a group as a the intended carry, but I mainly support (or tank which I also consider a support role) and I just can't compensate our inability to kill anything.
Tend to view every game like that as a loss and just work on improving my skills so maybe next time it'll be enough!
Tend to view every game like that as a loss and just work on improving my skills so maybe next time it'll be enough!
OdysseusVA
Banned
i enjoy getting teamed up with 3 unranked players and then getting steamrolled.
Fun times
Fun times
She's not an easy kill when she erects a wall constantly immediately after leaving ice blocks to escape and wait for teammates to come back. Bad Mei players always leave themselves open.
Antiwhippy
the holder of the trombone
Are they actually unranked or is it a bug? Because I've definitely been in games where the ranked number didn't "pop" until the end of the game
MarioHead3
Banned
Pretty sure that's the first night where my rank didn't change once. I just stayed at 61 the whole time, whether we won or lost.
Not complaining, just thought it was weird.
Not complaining, just thought it was weird.
Was hoping you kept playing Comp.
veloxStrix
Banned
I... don't know if I was garbage or my teams were, but I dropped 3 ranks today. Sitting at 61 now.
This is something I always struggle with: if someone picked Zenyatta, will my team be mad if I then went with Mercy instead of a damage dealer.
- Status