Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Season 1 I made it all the way to Plat 1 (almost P2) then suddenly teammates would not work together for almost 40 games in a row so i got demoted back to Silver 2. out of 50 or so games I lost 40 games. Does that mean I suddenly suck or am deliberately being teamed up with people that suck. It has to be the latter.
Yes, the goal of the algorithm is to keep everyone hooked. If you suck, it helps you win so you don't give up playing the game because it knows you give up easily. If you are good, it helps you lose so you keep playing the game because it knows you are competitive.
The metrics that would count good plays don't exist because you can't measure positioning, you can't measure time chased by two tanks and a strategist on Jeff, You can't measure holding the off angle and creating crossfire keeping the enemies held back.
Good plays are impossible for the coders to code for.
The best system would be
Step 1 Ranked placement matches
Step two 20 points for a win -15 for a loss +2 points for every win in a streak above 2
No rank resets and allow players to settle and find their spot and have games with similarly skilled players.
Team death match is conquest I think it's called
I like your solution idea but regarding your comment about measuring positioning. It is very possible to measure positioning. You don't think its possible for Marvel Rivals algorithms to be able to use an unit that measures how away and close a player is to the mission area throughout the game? And then use it for matchmaking? It is very possible. The games that my team loses quickly (under 7 minutes) and barely any kills, those teammates barely target the mission area and I even have saved clips of this. My teammates dying near the spawn while I flank enemies to the mission area, repeatedly throughout the game. BUT my teammates blame me for the loss due to lack of kills when I was one of the only ones trying to capture. I always get these teammates when I rank up too much within a short period of time. Instead of playing daily I like to play for several hours on the weekends which backfires because of this. How is it that matchmaking is suddenly so bad when I'm ranking up too much in a short period of time? How is it that Plat matches are much easier than Silver matches until suddenly you get teamed up with monkeys? How did monkeys even get promoted to Plat by not targeting the mission area in the first place when that is the entire game? It doesn't make sense.
See screenshot in the URL below of a heat map of some video game.
https://www.reddit.com/r/tf2/comments/kh4or/heatmap_of_deaths_on_dustbowl/?rdt=36567
You don't think the algorithms could analyse players based on heatmap tech like in that screenshot. But I'm not saying they are using heat maps, all I'm saying is that positioning is measurable. Also if you look into EOMM vs. SBMM, Netease made a whole paper 4 years ago describing how they would be implementing EOMM into their games all to increase engagement. To increase engagement, you need to create wins for non-competitive people and losses for competitive people. Without EOMM, you have competitive winners complete the season quickly and losers give up the season quickly.
thats the most insane sh*t ive ever heard. almost the same bs that a guy on YT said, that the game has some social MMR lmao.
but yea, i dont disagree about the games matchmaking, it is super weird, wether its called eomm or sbmm, its just really bad. but yea.. nothing to do against.
maybe they gonna adress it soon cause there is a lot of players complaining about it non stop and it could actually kill the game
What about games where 2 or 3 of your teammates carry you. Is that fair or insta-win?
What about games where your play style is incompatible with your team composition. Is that fair or insta-loss?
You are just coping.
The MM sure have problem, but not the ones you are trying to "solve".
35% - Neither team's team composition has the advantage against the other, so the match is even.
30% - Your team comp has the advantage, so you win easily.
30% - Their team comp has the advantage, so you lose badly.
5% - The game is actually unbalanced in some way, like with a leaver, smurf, or cheater.
A shocking number of people don't realize that playing, say, Magik / Squirrel Girl into double flier is just auto-lose. This has gotten even more prevalent this season as many polarizing comps got heavily buffed, like Fliers and Mr. Fantastic / Iron Fist / Captain America.
It is somewhat fixed in higher ranks, where people understand the counters better and you can also ban out some of the problematic team comps. The stompy matches are probably only around 30% total there, but it's still way more than it was last season.
EOMM and SBMM are not the same thing.
SBMM tries to put you against similarly skilled players. In theory, SBMM would lead to 100% of your matches feeling close. The complaint around SBMM is that there is no time to "relax" because when you know 100% of the time will be against people roughly as good as you, you have to try your hardest 100% of the time. SBMM is great for people who play games because they like the thought of improving as a player, but is bad for people who feel entitled to a 100% win rate because they have played similar games in the past and feel like they have "payed their dues" practice-wise, or people who just want twitch clips of them going on insane killstreaks against people 1/10th their skill level.
EOMM, however, just does whatever it takes to make you keep playing in the hopes of getting you to spend money. If you are a player who gets on loss streaks and refuses to quit until you win one, it puts you against better opponents endlessly because you winning=you leaving. If you are someone who quits the second you lose, it always gives you a few winning matches the second you come back. There is no integrity, just a matchmaker that builds a profile on each player based what the last few match outcomes were before they quit, and tries to avoid doing whatever it was that causes that player to quit. EOMM is really good for people who acknowledge that it exists, and takes the time to train the profile it gives you to give you the matches it wants by always quitting the second it gives you something you DON'T want, but its really bad for people who play games for the personal satisfaction of self-improvement and a feeling of fair competition.
Both are bad, and both can make modern games nearly insufferable to play, but at least SBMM has positive intentions behind it.
If the enemy team has 3 flying enemies and you have no hitscan dps it will feel one sided even if its not because the game is counter heavy. Thats the game style of hero shooters. Its why role queue breaks the game and they dont want to include it.
Its not EOMM, There is no hidden boogieman coming to hold you down. You can read the actual EOMM papers online and see for yourself that the game has no rigged aspects.
You think twitch streamers get given free wins? Do you even hear yourself? You can just watch their streams and see thats not true.
There is no punishment for ranking up quickly. People were going straight to masters and celestial when they are good at the game before the game challenges them. This is normal. Your crazy if you think the game is punishing people for playing the game correctly.
None of what you have said stands up to basic scrutiny or logical parsing.