Imagine sitting there getting so much negative feedback from the community about the way your ranked system is not only in valorant but also league of legends and you legit say there isn't a problem. I have honest thoughts you guys are just here for the money grab and could not give the slightest rubbish about your community.
My main point wasn't to say "there isn't problems". There are definitely things we want to change, improve, etc. But I do think there are things that people blame ranked for, that aren't true or don't exist.
So my question to you is how do we fix ranked for someone blaming "winners/losers queue" for why they are losing when it doesn't exist?
We also measure community sentiment, in surveys, data, etc. It's very difficult to measure social media. If an influencer complains about ranked, or an issue, then lots of people will agree with them(even people who don't play ranked or Valorant at all). The stuff that we can measure, surveys and data, we have seen increases in ranked satisfaction since the overhaul and each update we've made. Because of this, we believe that yes there are issues, and things that the community may be upset by, but Ranked is not in the worst state it's ever been(as some claim it to be).
I've worked at Riot for 7 months. My salary or any m...
Read moreJay “Sinatraa” Won has violated Rule 8.1 of the VALORANT Global Competition Policy for failing to fully cooperate with the Tournament Operator’s investigation and will be suspended for six months. He will also be required to complete professional conduct training.
After becoming aware of sexual assault allegations made against Jay “Sinatraa'' Won, VALORANT Esports immediately launched an investigation into the matter. Sexual assault allegations are some of the most serious that can be levvied against an individual and the Competitive Operations Team treats all such claims with the gravity they deserve. To conduct the investigation, the Competitive Operations team partnered with Riot’s investigative team to begin an objective and neutral fact-gathering exercise to evaluate whether Sinatraa’s behavior was in conflict with the VCT Global Competition Policy.
During the course of the fact gathering, the Competitive Operations team was info...
Read moreAfter two months of regional Challengers events, we are just a week away from Masters Reykjavik, the first international LAN tournament for VALORANT. More than 2000 teams from seven different regions attempted to qualify for the event, but just 10 will compete in Iceland.
To create the 10 team bracket, the top seeded teams from the two regions with the largest player base, North America and Europe, were seeded into the second round. The remaining teams were randomly seeded with the requirement that teams from the same region could not be put on the same side of the bracket in order to avoid matching against each other in the initial rounds.
wtf
Sick! Nicely done.
That's a nice looking lobster!
Read moreOkay. Here are some tangible, non perception things you can focus on to make ranked better:
1) Reduce MMR fluctuation by a lot, like I said, +15 to win and - 26 to lose is extremely harsh, and even on an average +16/-23 is not equitable. This is iron 2 data and I do not see why your system needs to hard correct in iron ranks
2) How does MMR work anyway? I top frag in almost every match (again, feel free to verify) but somehow my rr deductions rarely come below -23, while it doesn't also go beyond +18 either. Especially considering since you guys can take into account first bloods and understand value, if someone is generation so much value every single game, should the MMR not rank them in the above tier? In order to protect the higher ranks not having boosted people, you've doomed the lower ranks because no one can climb them
3) In my personal, empirical experience, the -rr should not be - 21 if your teammate goes afk. A lot of times a guy goes afk for like...
On your first point, I agree that this feels bad. We are looking at making this "I'm above my MMR and getting pushed down" and trying to make it feel better. The hard part is that no matter what we do or what system we use, something like this will occur. This is because you will be held accountable to the rank you belong to at some point, and this is touching your second point a little bit, but there isn't secret math holding you back or holding ranks down. If a player groups with friends, and gets boosted, or if you are just win streaking, but that win streak is not enough to raise your MMR(but you end up ranking up from pure win/loss) there is a chance that you end up a rank above where you belong. Getting above your MMR is worse in our system because we eat your points when you hit 0 RR of a rank, which means we are deleting Ranked Rating from your rank without penalizing you for it, which leads to inflation of your ranked rating; but if we didn't have demotion protection, it m...
Read moreUnlike some people replying to you I believe this. I'll admit I'm not the most aim heavy player in existence but that is why I use KJ/Cypher or Skye/Breach and they enable me to impact the match while not necessarily relying on my aim and I've been in Plat-Diamond since day 1. If the system was weighted as most people want it to be I would be Gold/Silver. Props to you for explaining the system so people can get a better understanding. The only question I do have is are the games balance by MMR not Visual rank im thinking that is why you can get some high gold and low Diamonds in plat games as their MMR is closer to the Plat 2 vs their visual rank.
Thank you for the kind words.
Yes we do match off hidden MMR, not forward facing rank. This is one of the things on my list that I want to improve on. We fixed and changed a lot of the easier targets, which make ranked feel better. Next, we need time to build the fixes for some of the more complex issues(like it feeling bad when you see lots of different ranked players in your match).
The "might see an increase in queue times" tells me the ranked player base may not be as big as we thought. Is this like League where in NA the unranked is actually bigger than ranked?
Nah, it's a "might" because of how the change works and affects match-making.
It was actually two fixes; we wanted to make two tweaks, which affected low and high ranks differently. The diamond and below change probably won't see a match-making time change. immortal+ might, but it doesn't seem like it. We just wanted, to be honest, there is a chance.
When you tighten match-making, you are saying, "Grab from a smaller pool of players." so now, if there aren't enough players in the pool, you have to wait until there is. So that's why matchmaking time would increase.
I think you can't say that "I'm not saying your experience is a lie because your perception is that you're better" but then list off all these arguments why actually, you're just making a big deal out of it by perpetuating ideas.
I KNOW my experience is not a lie. I think the metrics you use to measure stomps must be wrong too then, because I don't think you lie about running your tests. But as the consumer of the code you create, I am telling you there is a disconnect between your tests and the actual reality. I invite you, dm me, get my riot ID and see the number of stomps and see the leaderboards. Your matchmaking is broken, the MMR system is broken. I'm sorry but it is what it is, and I've played valorant RELIGIOUSLY since it came out, even though it never f**king works without me using the 1.1.1.1 dns
My point is that we get feedback, like the above comment, where "Valorant is broken because of X" - when that X isn't true or doesn't exist. So then it becomes a "why does the player feel this way?".
We can't solve problems that don't exist. A great example is winners/losers queue. There is no such thing in Valorant matchmaking that does this, or even a reason matchmaking would make you think this occurs. So we wanted to figure out why people thought it did. After we investigated this issue, it seems to be confirmation bias that this happens. It doesn't matter how much it doesn't exist; if someone is adamantit does, they blame something that doesn't exist for why ranked is terrible.
Such issues like that, along with people that may not be climbing ranked because they reached their peak, perpetuate problems that aren't true and makes the space harder for navigating. Also, these ideas that don't exist or aren't true spread. A new ranked player gets upset about a losin...
Read moreWhat a cap. Today went 11 mm’s. 9 win in a single row with sage, every game between 19-37 kills, 5-7 entry frags 4-7 plants, a lot of assist from heal, revives, clutches. Te 37 kills game literally got carried by me, if I died once we lost, had to frag 3-4 close every round, clutched 1v4’s 5 times, and I got 15 elo. Every game I gained between 14-17 elo. Then lost 2 matches -27, -23 elo.
Link your tracker.gg if you are comfortable talking about your games openly, we can talk about why you are getting those point values.
True. Clutching multiple rounds because you have team mates who were obviously boosted or in a premade group is hell. Imagine having Diamond players in a team who aim down to the ground and can't even use abilities. I have said this for months and I got downvoted to hell but RIOTs ranked system in valorant is not about skill. They ruined it in season 2 and idk why. Season 1 was perfectly fine and I never had to deal with such bad players in my team.
I see this feedback a lot that "Season X was better, why did they change it" or "Match making was so much better in a previous season"
We actually have not changed how the game determines MMR, or match-making, since launch. We did make one tweak, to help high MMR players(top 50 players in the world at the time) find matches faster. BUT that being said, the system isn't making different matches than it was in Episode 1.
You are getting the same matches you would have gotten 6+ months ago. The most significant difference now is that the community has been getting better since the launch of Valorant, and of course, we've added a bunch of content.
We also measure all of this, to see how often games really are a "stomp", skill gaps between players on teams, the skill fluctuation of players(if you have a good/bad day), and tons of other ways to try and measure "Ranked Health". Lastly, we also survey you, the players, and ask questions about ranked and gather data ...
Read more
devtrackers.gg
