Thinking of returning to comp after several seasons - will MMR reset?

Hello all, I have recently returned to OW after taking a ~5 season break from the game. I finished my final season with a (likely) mid-diamond mmr, having finished mid diamond in S4 and only done placements in S5.

I’ve been relearning in qp and am thinking of doing comp again, but I’m a little nervous about how the game will handle my SR if I return . I know it used qp mmr for first placements, and my qp mmr is fairly low on this acct since I’m not the most mechanically skilled player, and I’d never played an FPS before ow (my very first season placement was in low gold, after which I climbed quickly and stayed in plat most of the season). I’d hate to get placed with players 1200sr lower than my previous rank if they go back to qp mmr, and I probably won’t even bother returning if that’s gonna happen, lol. (Then again, pharah might be a blast to play at that rank).

Does anyone have any experience with entering comp with accounts which were idle several seasons? If so, was your rank based mostly on your old comp ranking, or did it sway widely?

After I took a 4 season break when I returned I ended up roughly where I was before I left.(a little lower)

I think the algorithm uses your previous comp as a benchmark and calibrates it based on your performance throughout the 10 placements to make sure it’s right about where you should be.

So if your skill is about the same, you are going to end up near where you were when you left.

2 Likes

It doesn’t use quickplay MMR in any way to determine initial rank.

You will prob restart diamond, but at worse around 2500

Actually, that’s not true. It does use quickplay MMR if you have never played comp before. Otherwise it’s all about your performance in comp.

No. Quick play is never used. Initial Competitive Skill Rating, Decrypted

You will start off with your old competitive MMR/SR, but it will be much more fluid for a while (probably about 20 games, including placements).

See also: Season 9 Placements Analysis, High Variance Explained

if you’re going to join comp again, this is the season to do it. From what i’ve seen, the community as a whole has been much more pleasant with the addition of commendations and the lfg system. Good luck!

Did this change?

I’m nearly positive it was stated by someone on dev team that it used your qp mmr as a baseline when throwing you into the normal comp algorithm if it was your first time playing comp. Then it moved you around rather rapidly from there before you get your first placement.

Every alt I’ve made had its first placement game in a ~2500 sr match even though I was playing mostly diamonds/masters in qp.

No, all he did was run what he felt was a test with a sample size of one and then declared that it “proved” QP is not used for one’s first comp ranking and everyone just gets gold. Yet there are other threads around here of people who only played arcade, then tried comp and crashed into silver.

So clearly gold is not a default sort of placements. I have little doubt one’s QP is used for their first base line comp setting as it’s how lower SR players are able to vault up in rank on new accounts. If the system only used placements only then their issues of silver playing like a silver in comp (as in placements) would thud them back into silver. So the only way I can see them jumping a bunch of SR is from racking stats vs low effort levels of play in QP.

as for your SR after a long break, if you go into LFG you can see the green/yellow/red for the SR of groups wanting to play comp vs your own MMR. Like I last played comp in S4 and had mid 3k SR.

If I go LFG right now for comp a gold group will show as red -1000 and one made by a person at say 3.3SR will show +/-150 green for me. So the system still thinks I’m mid 3k, as it should really. It’s not like you forget the fundamentals.

They might have done that in older seasons, especially 1 and 2. But I have no data from then. I also have no quotes from a dev saying that that was the case. Just many random assertions from random forum users.

Please actually read what I wrote before declaring it garbage, please.

And one measurement is enough if the signal to noise is good enough, as it is in this case. For example, one measurement is sufficient to measure the size of a piece of paper with a ruler.

I have read them… and I’ve never outright bashed your efforts but the simple fact remains that you are trying to push a narrative based off a very limited sample size.

Your views on if QP is or is not a factor in one’s first comp rating carries no more weight than anyone else’s view just because you wrote out a lot of info. I’ve seen players duck QP 100% pre comep and get silver. I’ve seen low SR players place higher even after wonky placements and had friends who did placement and got the exact SR they had on other accounts near 3k.

I highly doubt any of that is being accomplished in just a few placement games as it’s just not enough data for the system to work out a players standings. The entire reason most people seem to place gold is nothing more than that’s the 50% break point anyways.

Now it would be nice if the devs would just step and share more info about all of this stuff. Yet until (or even if) that happens the simple fact is the only way player harvested data is on some of these issue is ever going to work it if some group just goes full nerd effort with 100% new accounts. Stuff like a silver plays 25 levels of QP but then a masters does placements on that account. Then a masters players QP but the silver does his placements.

anything short of that sort of effort just isn’t enough info or a large enough sample size to really know what is going on.

Looks like you need to read it again. Cause you didn’t understand it.

I never said that. I said the first game of placements is in gold. And the range of ranks after placements is likely silver to diamond, with the stated caveat that I did not measure players with extremely good or poor performance in competitive.

Everything I wrote is consistent with these anecdotes.

I did an experiment like this. Quick play was 100% thrown. Competitive was not. I tracked everyone’s SR during placements to estimate my own.

I understand it, but that changes nothing about the issue. You think QP doesn’t play a role based off your very limited data set and I still think it does based off mine.
Being neither of us are devs and they are the only ones that know for sure, so we end up in a tie.

your sample set is just too small to make any claims of knowledge never mind pitch it as definitive as you tend to do. Being the issue in play is too complex.

The devs stated around the time competitive launched that QP isn’t factored into anything.