Does the MMR comp system penalize dps that pop off in the match making system?

Or… because they’re doing 20% more damage they win more and earn a higher rank?

Do you have a source for this?

It’s just based on your fire % so there’s more to it than damage or healing, then compared to how often each rank/hero is on fire at that rank.

Let’s say the average diamond support is on fire 20% of the time and average diamond dps is on fire 30% of the time, one who is on fire 35% of the time will get games that will be easier for them to win. This means they’ll get teammates who also belong higher, thereby balancing the rank of multiple people at once.

The issue isn’t that you get harder opponents, but that you don’t get proportionally better team mates.
Whenever I hit master, I start getting teams that are utterly dysfunctional, even when I play well.

Pyah-
I don’t buy your theory of SR chasing MMR, and in fact think you’re wrong. The reason being is that if what you are saying is true then the general conspiracy theory of forced losing streaks is also true. Say you a high silver player is playing with with his diamond or high plat friend all weekend and manages to gain 600 or so SR from 1900 to 2600 basically by getting boosted. His MMR would still be bad at 1900 but his SR would be much better at 2600. Now he plays soloQ so his SR chases his MMR and he gets put with major trash teammates for the next 40 games straight. A play of silver would have a hard time winning in plat even if things were even but if what you are saying is true the MMR system would add even more pressure by placing him with really bad teammates on top of being in a skill bracket that he doesn’t belong in order to push him back to silver as quickly as possible. That would be a forced losing streak.

Others claim that the MMR system seeks to balance the match making so that one team is not blowing the other out completely. This makes more sense. Say every player is assigned a score from 1 to 10. If one player is a 3 then he maybe gets a 7 teammate so the average is a 5. Then the other team also averages a 5 in the middle and hopefully a fair fight. This is why some sort of stats and dmg and healing numbers have to factor in.

If you win 4 games in a row, you will be about 100 SR higher, so your games will be about 100 SR harder, on average. There is a small performance modifier, but maybe that would mean you gain 120 SR in four games instead of 100. This difference is going to be too small for you to feel.

Yes.

Doing this will cause you to gain less SR/MMR, which will hurt you overall.

To see how things actually work, read How Competitive Skill Rating Works (Season 13), especially the summary.

SR doesn’t chase MMR by rigging matches, it does it by giving you more points on a win, and losing less points on a loss (assuming your MMR is above your SR). To see this in action, see 3000+ Skill Rating Data and Analysis (now including DCs). Each match is still typically predicted to be fair.

2 Likes

No, your team mates gets better also. But there is a slight problem on every border of a rank. Because there is lots of people who want to hit a rank and they dont play 100% tryhard after hitting a goal. Its not like masters are worse players somehow in your team than diamonds. That is just dishonest.

There are definitely better players in diamond than some (many) who are in masters.

Rofl.

It really does rig matches. The “on fire” meter is a very real indicator of your MMR. If you’re on fire 20% of the time and the average for that hero at your rank is 18%, you’re going to get easier matches designed to make you climb.

I’ve tested this in many ways on many accounts and it’s just fact at this point. If I play my 3 main heros I’ll stay within a 200 SR spread for many seasons. If I play what the team “needs” (e.g. McCree when enemy has Pharah), even if we win THAT game, I’ll lose more down the line even playing my mains because I didn’t put out the APM of someone who belongs at that rank on McCree. I’ll get unwinnable losses, even playing my mains, for a few games until my MMR adjusts again then I’ll get games I can’t lose.

I’ve also tested “giving up” when the team is throwing. Just sit in spawn if there are others AFK. The matchmaker sees this as lowering your averages/underperforming.
I’ll get unwinnable games for longer after doing that because it’s trying to force me to derank. You don’t have to believe me but I’ve tested this far too many different ways for it to be coincidence.

Do the other accounts happen to be called S23 and MHz?

If you want it to be considered a fact, you actually need recorded and shared data: a list of matches played, hero choice, level of effort, etc. This would allow actual analysis do be done to see if what you are seeing is more likely than not to have occurred by chance rather than rigging.

No not really, you’re probably hitting your skill cap when you start winning and rising in rank then reach an area with equally skilled players hence making it harder. You’re probably better than low platinum but struggle at higher plat maybe. Just keep practicing and lean.

Truth is, nobody besides Blizz knows how MMR works, not me, not you, not that guy over there, not his mom, not Santa, and not Kaawumba, whose post is regularly referenced in these threads.

I have my theory, but it’s spotty and based on one of the few things said by Blizz, which is that MMR aims to make every match “fair”. To me, that means gym class match making, making sure the highest 2 MMR are on opposite teams, basically deamanding they out carry each other, while the lowest 2 are on opposite teams and 1 absolutely gets carried to victory. I’ve been on both sides of the equation and all over the middle. …but that’s just my speculation.

(No offense Kaawumba, I appreciate the analysis)

To be honest, I don’t care if you believe me. I’m just sharing what I’ve experienced, take it or leave it.

I’m not going to argue, just know that I have 7 accounts and would be diamond border if I only had one. I have a TON of time in this game and I’ve tested it enough ways to feel confident in what I’m saying but again you don’t have to believe me. I’ve taken on silver accounts, tryharded for a few games, and then I get 2-3 games in a row where I can literally AFK and win my team is so good. Then the matchmaker catches on and thinks “you don’t belong here” (because I’m doing literally 0 damage, less than a bronze player with 1 hand would do) and I get games I can’t even win as a masters player on a silver account for a few games, then it catches on “wow, you belong in diamond/master” and gives me more games I can’t lose even if I AFK.

One parting thought–If you don’t think MMR is used to design matches to push your SR in the direction it should go, what else do you think it’s used for?

The thing is that we have 1000+ hours played Cuthbert and his followers, who believe the exact opposite, that after dominating for a few games, the matchmaker finds a doppelganger to put on the opposite team to balance the matches and force things back to 50%. If we all say “I played a lot and it seems to me” then we have no way distinguishing between what you say, and what Cuthbert says, and what I say.

This is why we need shared data and statistical significance analysis. That is, science, rather than just people stating their opinions.

It’s in the name → Match Making Rating. It is the rating used to make the matches.

Skill rating is a friendlier, human readable form of matchmaking rating, with the ability to implement features like decay and leaver penalties without affecting how matches are made (matchmaking rating does not change during leaving and decay). See How Competitive Skill Rating Works (Season 13)How Competitive Skill Rating Works (Season 13) → Summary → “Summarize Matchmaking, Rating, and Progression for me” to get the key references.

I think the people who have done the research in light of everything Blizzard has said about MMR over the past two years, and knowledge about how competitive ladders work in general - are as close as you are going to get unless Blizzard defines every detail that DIFFERS from other matchmaking systems.

We know that many variables are considered when matchmaking does it’s magic. many players do not even consider that map knowledge is a MAJOR factor in every single ranked game. The players with superior map awareness and knowledge, will have a distinct advantage over those who do not. Learning maps takes time, it’s a skill just like aiming.

Smurfs disrupt ladders in every competive FPS, and have since the start. It is very easy to conceal your skill from the matchmaker. In Overwatch streamers do this with heroes they don’t play - Lucio is a popular pick. Many people realized that Lucio is also an incredible duelist in the right hands, and he’s easy to grief players that can’t aim well with. Because even good players have a hard time hitting him.

These players enter matchmaking under the assumption that it is very likely that they will be playing whatever hero they normally play on THAT map. When they change this, they no longer perform at the expected level (they may not play the hero they picked as well). There is clear evidence that smurfs also use this method to trick the matchmaker into thinking that they are not as skilled as they truly are.

Players smurf in CS:GO all the time for lulz. it’s very easy to alter your sensitivity to something ludicrous to trick the check for accuracy, and play like a bot to trick the check for game sense. And then join a match, pick Junkrat, and repeatedly solo rip-tire the biggest threat on the enemy team out of the game. There is no recourse, you will just lose the game.

I don’t think there are many true smurfs. I think most of them are high plat and low diamaond DPS players that feed constantly and can’t climb. But when they feed in silver and gold and low plat, they don’t get punished - they destroy unaware players. They’re that average scrub you see in a cs or cod lobby that can’t take a 1v1 gunfight, they would lose to superior aim. They have to flank you and shoot you in the back to secure the kill. In high ELO, your teammate would see the flank and kill him, thus saving your life. In low ELO, your teammate would tunnel vision ahead and not see the guy - and he would get the drop on you. But it’s because your teammate was bad.

Just a few things that I have seen in my 26 years of playing FPS games.

Just wondering what was ur first FPS games and what games have you played past years?

Wolfenstein-3D was my first, DOOM was the first multiplayer (You had to use a BBS connection, the internet didn’t really exist). I played Duke Nukem 3D, team fortress, CS since it was in beta as a mod for half-life (I just got the 10 year veteran coin). I’ve played every Quake game, I have an heirloom hunstman bow and razorback on TF2, every medal of honor, every battlefield (including starwars), more…

Are you into strafing in quake series?

I played wolf, duke, quake series. I played the HL series but never got that into it.