Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Papercut
Aug 24, 2005
First the Mei wall and then the pro Sombra tactic:
https://clips.twitch.tv/iddqdow/ShyOryxFunRun

Adbot
ADBOT LOVES YOU

Papercut
Aug 24, 2005

Jinh posted:

And this part was solid gold

The enemy Mercy in that game will always be able to say he 1v1ed iddqd with Mercy while iddqd was tryharding. That is amazing.

Papercut
Aug 24, 2005

Kamikaze Raider posted:

So, correct me if I am wrong, but it looks like his season high was ~2400, which means he had to deliberately throw a poo poo ton of games to tank his rating to get there, and now he's making GBS threads on people who are very obviously worse at the game than he is, and talking poo poo about them?

Isn't that, you know, what assholes do?

I don't know why he's doing it, but iddqd is one of the top professional DPS players in the world so yeah he's going to be playing people who are much worse at the game than him until he gets into the top-500, basically.

Papercut
Aug 24, 2005

Bust Rodd posted:

I thought of elo hell as more like if you are just slightly better than your SR and the collective dead weight of Genji/Hanzo slams would work like quicksand.

That's not how most people use it, and conceptually it doesn't really make sense. There's no way to tell if you're "slightly" better than your current SR. It's only obvious that you're playing much worse players when you're actually playing people who are much worse than you. If you're slightly below or above your actual skill level in SR, that's not Elo hell, that's as correct as your SR can reasonably be.

Papercut
Aug 24, 2005

Jeza posted:

iddqd carrying himself solo to GM doesn't really prove that a kind of Elo hell doesn't exist. It just means he has the skill to bruteforce the system and carry almost any team to victory. A player could theoretically be trapped for a much longer than expected period of time in an SR ranking below their level simply because they string together long streaks of losses due to leavers, trolling, simple bad luck etc. and then climbs back in a piecemeal but reasonable fashion, such as a at 55% win-rate, only to string losses together again.

Even if you maintain a winning average overall, you can still stay in roughly the same place depending on how those wins and losses clump together. This kind of thing is more likely in an unpredictable environment, for example a particular SR area where there are statistically more griefers, smurfs, leavers, cheaters - anything that can step in from outside the regular matchmaking algorithm and mess your game up.

iddqd is just a more extreme example of exactly what you're describing. He can easily have a 100% win-rate through low plat, likely. And then as he gets closer and closer to his actual skill, his win-rate will slow down until he basically stabilizes at his "true" skill. The fact that his "true" skill is 4500 and someone else's is 2500 doesn't mean the 2500 players' experience is fundamentally different, it just means they have a different ability level than him.

Seriously, anyone who thinks they're in Elo hell, try throwing games until you drop a few hundred points from the level you're "stuck" at. This is easy to do and it becomes immediately apparent just how big a difference 300-500 SR makes. Anyone can easily carry at 500SR below their "true" rating.

e: This,

ToastyPotato posted:

A GM carrying himself to GM rank from the bottom is equally as useful at disproving ELO Hell as a Diamond carrying themselves to Diamond. Being a GM, he will do it faster, but it still completely proves the notion that you will settle to where you belong.

Papercut
Aug 24, 2005

CharlieFoxtrot posted:

I tried some aim training drills today and when I actually focus on what I'm doing I realize how uncoordinated I really am. You'd think that your movement hand should be able to coordinate with your aiming hand since one brain is doing the work, but when doing movement drills I realized my aim was always playing catch-up.

The first quick play match I did already felt better though, I'm gonna stick with this and see how much it helps.

What training drills are you doing?

Papercut
Aug 24, 2005

Bleck posted:

And I'll still lose a big chunk of points for that, because the rating system thinks that I could have carried them, somehow, if I were good enough.

I don't know, it's just frustrating, you know?

You don't lose a big chunk of points for it unless it's your 5-10th loss in a row. And it doesn't even matter because if losing that big chunk takes you way below where you should be, you'll just easily win a few and get that chunk back.

Papercut
Aug 24, 2005

Cicadalek posted:

Is SR gain/loss definitely affected by performance? I thought it was all just based on the combined SR of your team vs the enemy team.

If it is, it's by a barely noticeable amount. I personally don't think it is, it introduces way too much unnecessary complication when over time just W/L will have the same effect without having to make any obtuse assessments of performance.

Papercut
Aug 24, 2005

Good Dog posted:

Yeah if it were as simple as wins/losses while weighing against your team's average SR lots of people would place with the exact same rank.

When I placed I had 5 straight wins, started getting placed against plat/diamond players, and then lost a few and ended up like 6/4 or 7/3 which gave me an SR of 2300. A friend won 9 of 10 and ended up below me. It is possible that he did less in his games to help the team win than I did in my games, but that is really hard to quantify across characters. Also I didn't even have the game when season 2 was out so season 3 is my first taste of competitive, whereas he placed very close to where he finished season 2.


It makes sense to me that if I'm 2500, my team average is 2400 and the enemy team average is 2375 that I would gain a little bit if I won and lose alot if I lost because I was a higher ranked player on a higher ranked team to where I should win it. It might be even more complicated than that if it ties in your game performance to it. All I know is that losing 2 games and then winning 3 should be a net positive in your SR and the other night I had a string of games like that that ended with a lower SR than when I started, which is frustrating.

I think it's pretty natural to assume that the level 25 restriction on competitive is for the game to get a QP SR ranking on you before you do placements, and then it uses that SR as the starting point for your competitive SR. You can even see this in effect if you get a smurf account; by level 12 on my smurf all of my QP games were against low- to mid-Diamond players (I'm still only level 16 or something, so we'll see where I actually place).

Using your in-game performance would be a tremendous amount of work (how do you accurately measure performance?) for a gain that people would barely notice if at all, while introducing a ton of variables that could screw with your matchmaker. That's the type of thing any reasonable business manager wouldn't touch wtih a 10-ft stick.

Papercut
Aug 24, 2005
I only caught the tail end of his explanation for why he was stopping, but iddqd pulled the plug on his 600-Grandmaster stream. From the sounds of it, because of reasons brought up here, that it was lovely to the people he was playing. :(

e: apparently it's because of a Reddit thread where people threw a shitfit about poor sportsmanship, lol

Papercut fucked around with this message at 19:42 on Jan 18, 2017

Papercut
Aug 24, 2005

Fojar38 posted:

So complaints about Symmetra usually get a response of either shooting her in the head or playing Winston but it takes one and a half full winston clips to kill a Symmetra shield generator or teleporter

Why does it have as much health as a tank

Tracer builds ult MUCH faster than Symmetra and can easily use pulse bomb to take out the shield generator. That's an easy trade. DVa ult is also a good counter to it.

Papercut
Aug 24, 2005

Fojar38 posted:

Wait did you just suggest I spend Dva ult on a Symmetra teleporter

It's a totally fine trade because DVa without ult is a million times more useful than Symmetra without ult, so the net result is your team gaining the advantage.

Papercut
Aug 24, 2005

Oxyclean posted:

I think the best earthshatter counter is don't be clumped up near an enemy Rein. If only gets one or two people with it that leaves the rest of the team to just melt him if he tries to go for a kill.

Pro teams still get caught in Rein ults all the time so "just don't get hit" is much harder than you make it sound.
https://www.twitch.tv/overwatch_nge/v/115687370?t=01h02m53s

Papercut
Aug 24, 2005

Oxyclean posted:

I can't watch the clip now, but I'm just not sure I'm understanding what the problem is. Do Pro teams occasionally get caught in D.Va ults? Do Mercys occasionally get full team rezs? There's not much wind up on her rez, and that sort of thing seems far more game swinging then any other ult on its own. I don't think the counter to Rein's ult is supposed to be something you can react to - you play around the fact it's something he can do.

I apologize if I'm coming off as facetious or something, but I'm not sure I follow the reasons for why Rein's ult needs more windup / a nerf.

It's just a clip from the NGE tournament yesterday of Rein catching half the opposing team with an ult.

My point is just that "don't clump up and you won't get hit" trivializes the difficulty of following that advice. Even pro teams get caught in Rein ult frequently. The only time pros get killed by DVa ult is pretty much when it's wombo comboed with either Zarya or Rein ults.

Papercut
Aug 24, 2005

Wildtortilla posted:

When I quit playing this weekend I was on my 14th straight loss.

Did you group to get to where you were or something? 14th straight loss means you somehow ended up waaaaaaaaaaay higher than you should be. Just lose your way down to your ability level.

Papercut
Aug 24, 2005
Did they change loot boxes? Previously i thought you were guaranteed a blue and white minimum in every box, so even if you got a purple or something then i the remaining three would be blue/white/white at worst. But since the patch most of my boxes with purple or gold have been that plus three whites.

Papercut
Aug 24, 2005

Broken Cog posted:

Nah, he's actually pretty bad, but he's fun. And useful on lower ratings.

Edit: My point is, Junkrat could use a little love.

I think they should keep him as is, a training dummy for like 3/4 of the cast.

Papercut
Aug 24, 2005

Kamikaze Raider posted:

I hate CTF as much as the next guy, but in the several games I have played of this, I have had this happen exactly once. This just doesn't happen in my experience.

I've played a bunch of games trying to get my arcade wins across two accounts, and 0-0 draws are realllllllly common. That game mode just sucks

Papercut
Aug 24, 2005

DrSeRRoD posted:

Checkout Mystery Heroes if you just want some arcade wins. It's actually pretty fun (until a team wipe brings them back as a real nasty comp).

Yeah that's what I've been doing. 3v3 is by far the fastest way to get arcade wins, but holy poo poo does it get boring fast. Mystery heroes is the best compromise between having a decent chance to win and actually being somewhat fun.

Adbot
ADBOT LOVES YOU

Papercut
Aug 24, 2005

Nevvy Z posted:

I thought you did? I think you should always keep it, personally.

You do not, unfortunately

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply