Winrate vs Performance - how to climb

So amidst all the hooplah about this GOD AWFUL match making system in overwatch, (hots was the same way, r.i.p dedgaem) i see a lot of claims about…

its not about the winrate… its about your performance to rank up…

Does anybody have ANYTHING to suggest that is true? Anything said by gamers, streamers, blizzard… anything more than just trollish rumor? I wouldn’t be shocked… theres no telling what has caused overwatch to have such awful MMR results. But i just would like to see something to back up this claim ive seen so many people make…

So please… discuss… what do YOU think causes rank up?

edit: Please try and back up your claim with SOMETHING decent.

3 Likes

Cant be performance based. I have been in stomp after stomp with POG and perfect stats - still bronze 5 70% WR

I even started to play selfishly and farm kills. It’s pathetic how I run circles around these other teams. I feel bad for them.

1 Like

If i may ask and request… how many games have you played since placing bronze 5, and if its more than 30, would you mind taking a snipit screenshot showing that and link?

Okay so I am using DPs queue to test this. I played incredibly aggro to farm kills and stats.

NEVER RANKED IN DPS QUEUE OW1

All Mei:

7-2-1 BRONZE 5

Top Frag every game, POG 7/10 games. I seriously carried so hard most games that I never topped 4 deaths any game.

7-0 Update….BRONZE 5

Undefeated. Seriously making people quit during the matches they were so frustrated with my Mei. I got POG 6/7 matches, never topped 4 deaths. I tube my Crit accuracy was ~50% I was cracked.

I’ve recorded every games stats and it’s bonkers. I feel so bad for these other teams as they have no game sense and let me run circles.

It’s simply broken in concept even if I have to “climb out” with 30+ wins at this rate.

I’ll keep grinding and see what happened but these poor new players grinding 50 wins only to get face slammed in Bronze 5.

I haven’t had this comp issue until I reached masters 1 with a 70% wr once I won 7 games thinking I’d promote to GM 5 “I didn’t” my theory is there’s technically an SR system with numbers we just can’t view which was pretty much told to us by the devs, the theory is say you can be bronze 5 but you can be sub 500 SR technically so even tho you’re winning a bunch of games you don’t actually have the amount of SR needed to actually to tier up and a lose is counted as -100 maybe? Not sure on numbers but that could be the reason why people are stuck in bronze 5 they prolly have sub 500 SR

2 Likes

I have multiple accounts. On one I haven’t played on in a long time (2 years) that was 2700 I placed Silver 2 and climbed to Plat 1 within 5, 7 win placements. I had another that was 2400 placed Gold 5 and climbed to Diamond 5 within 4, 7 win placements. I have a bronze account used to play with my IRL friends (800 sr last season) that was placed Gold 3. My actual Diamond account placed Plat 3 and I will have an update on what I rank to with that tn or tmrw. However my winrates, besides the bronze account, were rarely, if ever positive; yet I was still able to fly through the ranks.

Sub-diamond was always in ow1 performance “based”. You would still lose sr when you lose a game and gain if you win but you would lose a lot less and gain a lot more if you “in stats” outperformed people playing that same hero on that map. This was done to boost smurfs up so they don’t get to stomp low ranks for hundreds of games. I doubt blizzard changed anything about this.

1 Like

To add on league has a similar system with LP and MMR and having a HIGH mmr determines how much SR you get which translates to the tier systems. So if your mmr is super high you’ll gain insane amounts of SR-Tiers and if it’s low it’ll cause you not to gain much therefore staying in your current tier.

What does a screenshot of perfect stats look like :thinking:

3 Likes

I’m still trying to figure out why blizzard though a ranking system with even less transparency would be a good idea?
its genuinely baffling.

7 Likes

I mean, my ana stats are looking pretty good…
https:/ /www.reddit.com/r/Overwatch/comments/y0jyer/comment/irsdvo8/?context=3

1 Like

everybody keeps using this word transparency… but what good is seeing the inside of a dead engine?

I have the opposite experience as most people in this thread.

I placed plat1 in dps despite being ~4.1k at the end of ow1. I have had an awful win rate since then. I cannot explain it I have never had this many terrible games in a row in my life, but regardless I have 21 wins and my 3 like sets of 7 wins were something like 7-14, 7-17, 7-15

I have a 30% or less winrate with every hero that I tried.

Despite that I am now diamond 2 because I went from plat 1 to diamond 4 and then diamond 2

3 Likes

Two historically gm accounts. Same player with tons of rust.

1 account gets placed in Diamond V. Player gets wasted with beers and plays noticeable worse. Like, way worse. Closer to the end, player didnt care and only wished everything finishes fast. 2nd account gets placed in Gold IV.

Gotta try this saturday again. Because guess what, I have more high rank accounts :smiley:

A little late but I believe performance plays a heavy role. I placed around gold 3 in support and grinded it out of the first day. 35 wins later I’m diamond 4 close to where I was at last season. Even though my winrate is about 44% due to an unlucky amount of DCs on my team I still ended up climbing every 7 wins.

1 Like

anecdotal bc based on own exp but all i focus on is not dying and providing max value(only placed supp so far) and it seems to b working i only have 14Ws so far but i went up two divisions from where i placed

But there can be other factors at play there… Possibly the mmr of the enemy team was lower in one game? Im not disagreeing, im just searching for something concrete to support the claim so many ppl are making.

edit: the thing that i think is most “undisputable” is that the MMR system in ow1+2 is terrible. no other game ive ever played have the majority of games been one sided.

Maybe the new rating system has a deep elo for sub bronze 5. Like 0-500 for bronze 5 and 501 to 600 for bronze 4.
I think the decay system might have something to do with it. Like maybe it has been decaying for many seasons and many players are starting a 0 elo. A lot of my friends were placed bronze 5 or bronze 4.
They did say that players that decay will go up and down faster to get back into things. Maybe players have a 0 elo and a high decay rating going into placement. Then maybe performance now has a much more major role and a bad performance on decayed player really sets them deep into the low ELO. But if they perform well on all their placements they get highly rewarded.
I can’t figure it out honestly, but there are for sure some friends of mine that are getting placed bronze 5 to bronze 3 when they should be closer to plat or diamond.
I have personally been placed bronze 5 on my tank and support and would have finished my damage today, but characters were locked.
Once I finish placing all three I am going to test if deaths have anything to do with it. I hope it is not accuracy, especially because we have characters with infinite ammo.

Ok so reading other players instances, I think I have an idea how this works.
We know that you can go up even if you lose more than you win. You can also not go up if you win all 7. Some people report a 50ish win rate and went down a bunch.
So here is what I think:
Players decayed from OW1, they said that when decayed players returned they would go up or down elo to get back to where they belong quicker. So we know there is a variable, which we can call volatility (v).
Let’s just say as example.
Wins = 5sr and Losses = -1sr
Bad performance= -7sr
Average performance = 0sr
Good performance =7sr

You win 7 and lose 14. So 35-14= +21
Performance is “average” based on accuracy or deaths or something, but you have one bad game. So -7sr.
You net 14sr multiplied by your V, which let’s say is a 3. so you get +42sr. Which does not rank you from bronze 5 to bronze 4.

Another player, plays sigma the entire time, games calculation for tanks is buggy and goes up 5 ranks.

Another player has 3 bad performance games multiplied by the V and ends up losing 100sr, even though they had a 60% win rate.

That’s just my thoughts, that there is something like that going on, but more complex. If performace is taken into account I have no doubt they are doing it wrong. Each map, each hero and even each part of the map would have different stats.
A player could only play Mei when they need to hold the last point and just dies a bunch, with horrible accuracy and 0 ult kills because of the short play times. Then get deranked.

Hopefully it’s not like this, but I have a strong feeling it is. if so I don’t mind staying lower ELO to play for wins instead of performance, but new players will find it hard.

I have a theory. The ranking IS performance based, but they are calculating performance incorrectly. For example, if you play tank they may be comparing your damage mitigated and objective time to the average stats of everyone that plays your hero. But this doesn’t work because in stomps you will contest the objective less and have less damage to mitigate, so depending on how bad OW programmers are at understanding data, it’s possible they coded it so you can only climb by playing close games. Especially because if you are stomping you are likely playing against enemies with lower rating than you so you will rank up more slowly. In certain rating systems, like Chess, you will literally gain no rank from beating someone significantly below you in rating. The way to climb is to beat someone above you in ranking, and the matchmaking may be screwing you over by only placing you against bad players.

1 Like