One more thing, have you noticed that they have removed the "behavior score" from the Match Making? I am seeing such poison in my games in the last week that was refreshingly gone for the last couple of months... I am not a big fan of this. I liked to play with good behavior score people more...
LoL also has a drastically better MMR system which allows you to rank up by placing you with actually good teammates (for me I had russians sadly but I ranked up anyway), that is not the case with Dota
Single MMR is super abusable. Even taking yourself as an example, you'd be ruining loads of solo queue games if you were matched off of your party mmr.
No?
For example this girl I'm playing with is getting boosted party, right now at around 5.9k
But it doesn't matter because she will never play party with anyone except super high mmr friends to carry her. If she plays on her own with "fair" teammates she will lose every game.
If you combine solo/party, so what? She will NEVER play solo because she will lose every game.
But most players will end up with a more accurate MMR, boosted accounts are much rarer than everyone seems to think. Its much more common to have someone in your game who took a long break and now sucks, but the system requires hundreds of games to recalibrate them properly.
And as far as "MMR validity" goes, you would only need to play 1 game with her to realize she deserves 2-3k MMR.
Not even gonna read what you wrote because honestly I'm too lazy. Just gonna say that I personally feel the medal system caters to divine players and not much else. My games have been so fucking unevenly matched it's insane. My medal means I'm around 1960 MMR, today there was an Ancient 1 in my game, which means he is around 4200 MMR. Does that not seem a bit dumb?
At first I agreed with op. But Marlan makes some very legitimate points. It would be too easy to get boosted if it were a single mmr.
Also real quick, I'm in a similar situation where my party MMR is hundreds of points lower than my solo because when I play with friends we tend to joke around and throw a bit (i.e 1v5'ing as TB or solo ulting a squishy support just to watch them explode). It's to the point where I don't even play solo games anymore because I honestly don't see myself ever increasing in medal. No one on my friends list has even gone up once from what I can tell and I'd much rather have fun playing the game than I would praying the people I get placed with have two brain cells to rub together.
My point was that it doesn't matter, and that solo/party should be combined.
Players who abuse parties to get booster aren't going to be playing solo anyways, and even if they do they will lose.
Its not like they can't just pay someone to boost them in solo queue as it currently stands.
@Anime Profile = Win Game What you said in your first comment is completely untrue. Ranked system doesnt cater to divine players and thats what i can tell you from my personal experiences. Recalibation was designed in a way so it deflates mmr of high ranked players. That resulted in everyone from 6.1k to higher tiers of 6k and even some lower 7ks being brough down to flat 6k/ low 6k. Everyone below 6k wasnt even touched. My mmr right now is around 5.9k and the bracket around me is clumped up with so many players who were much higher mmr than me before the recalibation and the skill gaps in games can be huge even though the differences in ranks arent that big.
Concerning the issue about an ancient 1 player being in your game you should try turning on strict solo matchmaking in the options so you dont get matched with party players. The other possibility is that his rank might either be bugged or that it presents his party mmr while his solo is lower. Anyway badge has no meaning in your games as they are balanced around the avg mmr just like the ones before the update. The badges were just a useless cosmetic upgrade.
@Metsis Behaviour score wasnt removed from matchmaking. A few weeks ago i tried searching searching for Solo ranked games at the same time my friend does. Our mmrs were pretty close (and behaviour scores both normal) so we would consistently get matched together and it was pretty fun playing either with him or against him. A few days later he got a few abandons because of his internet being bad and ended up in low priorty which decreased his behaviour score to D. My behaviour was still normal and after he got out of LP we tried queing again a couple of times and never got matched together because of the difference between our behaviour scores.
I mean now what's the use of playing party mmr if it isn't going to make medal higher than ancient 5?
I just saw I girl with 6.2k party mmr ancient 5 so I mean playing party till u get ancient 5 is ok but after that it's just useless to play party
Most people who play with their parties play less serious, they are just fooling around and doesn't care because it's just party mmr
It's way to easy to get boosted in this manner. You don't even need to hand over account details.
Doesn't matter if you're party mate will never play solo. The fact is if she does, it wouldn't accurately reflect her ability to perform in a 6k avg game.
I can see ringbone has no friends which is why he would question people about playing party mmr lewl..
Eddie I'm not questioning people I'm stating a fact that it's only useful playing party till u get ancient 5 then there's no use playing party mmr
And how can u say I don't have friends? Even my party mmr is 4k which I got by playing with my friends and I can also see that ur 2k which means u don't have any friends who can play party with u to boost u
@Umbranox About the behavior score thing. This just happened about a week ago. My solo & party games were a joyous support fest until about a week ago... And the old "cancer" returned to my games at that point. And it is not only "my feeling" that makes me think this.
Another thing supporting this is the fact that the search times have seriously gone down during my "find matches". So something has been "relaxed" from the match making... That is a fact. We had multiple minute searches sometimes even 5+ (in the shit tier) just a short while back and now we are seeing matches in about 30 seconds. And the team mate behavior has changed dramatically.
This is why I think this is so.
I am aware that the combination of the MMR thing will meet with some resistance and that is to be expected and sure it will create some rough matches from time to time. People buy accounts as I am seeing some quite high badges in my games and the players aren't clearly of that level and it would be nice to see the "current badge" when going into the game so you could see what the current skill level is. As the highest badge may reflect "the other" MMR etc. This just needs to be changed... I am not shamed of my bad solo badge at all.
What do you guys think of the "performance" points that I would like to see added to the MMR system? So if I play really good, I don't lose as much MMR as the rest of the team. A small reward for "doing your job well" kind of thing...
@ringbone all my friends are 1-2k and my party is still 1k mmr, so how exactly am I going to boost my party mmr? You just stated that there's no use playing party once you hit 5k party mmr, why is there no 'use'? Are you that brain dead to realise that most people play party to have fun with friends and not get their 'boosting' done? So you use your friend to boost your party mmr? Lmao. You're weird
https://www.dotabuff.com/players/379757546 look at my friends party mmr and realize how cancer in party can be :)
The thing is ur just 2k or 1k whatever nobody cares if u loose or win in that bracket but as players who has higher mmr it takes a lot of time to find a match and they win like 2-15 mmr for a win so most likely no one plays for FUN in that bracket while people like u get 25 mmr
One of the reasons ur still 2k is this that u just play for FUN.
I mean ask 5k players if someone's got both solo and party 5k what would they play party or solo? Most likely they will focus on solo and if someone gets like 9k mmr playing party what's the use of it? No one's gonna know ur party mmr is 9k
If there was a performance modifier added to the MMR adjustment, the top tier players would benefit most from it as they usually out perform the "average" by quite a lot. So they would actually gain more MMR from their Ranked matches than they currently do. And it would take a little edge off from that -42 MMR loss as well.
And this party is nothing and solo is the thing debate is just why it would be so much better to have a single ranking... Maybe have the average solo game give +25/-25 and the party game +15/-15 to make solo games more meaningful if you are looking to climb the MMR ladder or something.
It would be so much easier to balance out overall as a single rating...
it wouldn't. you're just being naïve.
boosting would be on steroids - get 5 clients who want boosting, play as a 5-stack of boosters, finish every game in 15mins.
don't forget that party matchmaking currently takes your solo mmr into consideration.
Why are people talking about merging party and solo mmr. Please god never let that happen. The only ones who want that are the 2k retards hoping they climb out of their trench with the help of their friends.
itt: retards who can't think for themselves.
If you have 1 boostee, and 4 boosters in a party, that's already going to cost the boostee 4x as much as he is paying for 4 boosters now instead of 1. The only "advantage" things brings to the boosters of dota, is that instead of having a ~90% winrate playing solo they will have a ~99.9% winrate for example. End result is that merging party/solo won't change the boosting market at all. It will still be much cheaper for people to be boosted solo than in party (even if there is only 1 booster in a party - because now they have to deal with scheduling bullshit, and having 1 anchor on their team, making it hrader to win.)
Sure, friends can boost friends, but that rarely happens, in fact I've literally never seen it happen in all of my party games EXCEPT for girls. Girls are almost ALWAYS being boosted. But that won't change. They've already been boosted and combining MMRs won't change that. They don't play solo queue anyways.
Most of you apes can't use your brain. You just see "merge party/solo" and have a sperg fit without even thinking about it critically.
Even if some whale wanted to pay for a 99.9% winrate boost from a 5 stack of boosters, who gives a flying fuck? It's actually BETTER for everyone. He's going to be boosted anyways, but now he will ruin WAY less games on the way up. It doesn't mean there will be more boosted accounts, because this kind of boost would cost a LOT more, so the demand will be low.
Anyways, Valve should either allow party MMR to gain badges up to Div5, or merge party/solo.
Oh and for all you morons who think Party MMR doesn't matter. You don't have a brain either. Congratulations for being on the Reddit bandwagon though. Maybe think about it critically for yourself though.
Common ape arguments
- Party is easier cuz coordination > I guess you're not playing vs. other parties either, m'bad.
- you can get boosted ezpz > so can solo players, and probably much more often. I've never seen someone get purposefully boosted in party.
- nobody cares about party mmr just memes hehe! > This is just groupthink. Almost nobody I know goes into party playing to lose, or not trying. People only say this to feel less bad about losing in party as an excuse. "hehe i wasn't really trying!"
- Party MMR is inaccurate, it's an average of your friends > Yea, but solo isn't accurate either. Theoretically it is, but for you to have an accurate MMR it requires hundreds of games (every time your MMR moves a significant amount). And, due to how fast dota changes, can become inaccurate very quickly. If you always play with your same group of friends, who cares, your average MMR will be accurate. (And if you shuffle friends, eventually your real MMR will show anyways, as you play with different skill levels it's very similar to if you were playing solo queue)
TL;DR: You're all fucking retarded, and Valve should combine party/solo MMRs.
I know plenty of people who get boosted playing party with their friends but are not as good solo and also play solo. That is enough reason to tell you not to merge the two.
I know plenty of people who get boosted playing solo but are not as good solo and also play solo. That is enough reason to tell you not to have solo queue.
^ your logic
Boosting an account is quite stupid if you ask me... Why would you like to get your MMR to a level that you can no longer play the game even half decently??? I don't know how this has become an industry.
I'm not asking to get free MMR or anything like that. And my "new higher MMR" wouldn't be that much higher than what I have right now, it is just weird to have 2 different ratings... And to make it more fair the new MMR could be calculated like this (Solo MMR + Party MMR)/2= New Single MMR. So you would be in the middle of your current MMR's somewhere. Or it could just be like "we are merging the two different MMRs which one would you like to retain as your new MMR?" question like thing. I know most would pick the higher one (I would) but I have a friend who has been dropping his MMR with the new calibrations (this latest one and the TI calibrations) if it was a possibility for "easier games". So it is not like this is a lock that everyone goes up. And he has always climbed back up to his current probably highest ever again and again after each drop.
Does someone disagree with the "drafting screen shows your current relevant MMR badge" thing? If yes, I would like to know why? Currently without any visibility to the current status you are going into the draft blind about your team. The profiles have some information, but not much and you should be focusing on the draft and not on how good your team mates are based on their profiles.
Some people just want a bigger number/badge on their profile. But w/e, let people do what they want to do. It doesn't hurt anyone if they don't even play solo.
I'd much rather have a game where my matches actually mean something.
I hate playing solo, it's not fun for me even when I win (last time I played 40-50 in a row I had 70% winrate), but I'd rather just stay in party queue that is actually enjoyable. But now I'm stuck with this disgusting divine 1 badge even though my party MMR would put me at divine 4 currently, and very close to divine 5.
The incidence of people getting boosted solo is isolated to them subscribing for such a service. You cannot really deal with it by designing a better mm system except by somehow increasing the cost of making New accounts.
The incidence of casual players playing party with much better friends - who likely are on their smurf - is very high.
I mean there really is no comparison here. I see you're trying to bring in the fact that people will get boosters if they want. But it's a question of accessibility. Making a single mmr makes it way too accessible for casual players to be boosted (esp for those who get boosted unintentionally). Meaning instead of those who would pay for boosters, you have a whole group of players who are going to be boosted unintentionally.
Case closed. Thanks marlan for your input. But your model just doesn't pan out when you lay out the blue prints.
@HanSolo
By this logic, if you play support you are always being boosted. As you cannot win the game alone and need the team to win the game...
The absolute BOTTOM line in Dota is that your team will do as good as all the people in your team! You can play like a god and still lose and you can really suck and still win... If your team comes through or fails hard. This is a team game! And you are only as good as your team mates are. And in party queue you know at least 1 more person and his/her skills going into the game and you most likely have better communication with the said person and thus better coordination between your actions... That is the difference between solo and party queue. It does not magically make you play any better than in solo queue so I don't see why we need different ratings here. You just talk more in party games... It is basically separating your "no communication" games from "some communication" games. You are not any better or worse skill wise in either setting.
If you play ranked with a high mmr friend, firstly, eventually your MMRs will average. The low mmr has a overskilled teammate. The high mmr has a 500 ton anchor tied to his leg.
They will average out.
Most players do not want this. I won't play with some dogshit player when I'm on my main account.
Yea, I'll use a smurf when I play with my low mmr IRL friends, but I just pick support or fuck around. I know some people have some kind of superiority complex and tryhard even vs. 2ks but I don't see the point.
Anyway, people are going to get boosted no matter what you do. Sure, people can get boosted in party sometimes, but it's very rare someone will specifically boost someone (tryhard every game and try to carry their feeding ass) What's more likely is they are just friends who play together a lot. Their MMR's will average out and it doesn't matter.
-> If you play with 1 group/person only, your MMR will average -> No issue.
-> If you play with multiple groups/people, your MMR will eventually reach YOUR MMR. It's exactly the same as queueing solo, albeit slower.
I understand the concept of averaging out very well but I feel like high mmr players smurfing is very common (could be wrong). As I said it's unintentional boosting that is the worry here. But also intentional boosting becomes easier. But let's ignore intentional boosting.
I'll give you a few scenarios:
1) friends play in pairs all the time. They understand each other's habits well and play well together. They have good synergy. They both can end up with a higher mmr rating because they play together.
2) One high mmr player who is still improving rapidly. His party is severely underrated because he hasn't played it in one year. He plays with a Low mmr player who is at his true rating. That player gets boosted. This is actually very very common scenario.
3) high mmr player smurfing to have easy games. Low mmr friends get boosted. Games are not serious but Low mmr friends are boosted nonetheless
4) for averaging out of mmr to actually work, the Low mmr player needs to actually play core. Let's say he only ever plays supports, his impact in games are minimized. The progression of mmr rests more on the higher mmr players ability. This is a less significant example of getting boosted.
1) That doesn't matter at all. Dota is a team game, why do you want to punish teamplay? Other players can earn MMR by playing well with others too. That's like saying it's unfair to be good at CS'ing or any other Dota related skill.
2) That has nothing to do with merging solo/party. That's a different issue with matchmaking taking WAY too fucking long to calibrate your MMR properly. The same issue exists in solo queue.
3) Sure, but most people don't smurf that often, and the ones that "do" probably shouldn't be smurfing anyways. I get extremely bored smurfing for more than 2 games in a row.
4) Assuming they play at relatively similar skills, support players DO matter quite a lot. Regardless, this is still basing your opinion on some isolated situation that doesn't even mean anything. Their MMR still averages. As a unit they are matchmaked fairly.
Unintentional boosting is not really a thing... You may end up on the winning side of a smurf game or you may end up on the losing side of a smurf game... This is not really a thing. A game here or there doesn't really effect your MMR in any significant way.
And what comes to the #2 part on the @HanSolo post, would not be an issue if we only had one ranking... The other ranking would not lag behind even if you don't play it, so it would not be an issue.
Smurfing is one thing that is hard to get out of the game... The phone number adding almost a year ago helped against this to some extent, but you still see smurfs from time to time. But it is the same as the unintentional boosting thing. A game here or there doesn't effect your MMR growth in the long run all that much. And if you don't actively play with someone smurfing, the effect is probably about 50/50... You are as likely to be teamed up with one as you are as likely to be playing against one.
1) playing with one guy over and over is very different from play with randoms which test your ability to adapt to different play styles and game scenarios.
2) agreed. But that's one of the fundamental problems. Which is even if there's only one mmr, there are a lot of players who don't play ranked and dont have mmr up to date.
3) I would argue that people smurf very often for party.
4) supports matter at high mmr (your games, so I see why it's relevant for you). And I'm talking about high mmr core Low mmr support, not equal mmr players. So there's a big difference.
I have quite a wide gap between my solo MMR and party MMR ... ... The difference is several hundred points
Wow, slow down there! "Quite a wide gap" would be 2000 mmr at least, anything less is basically no difference
Supports matter... At all levels. At my bottom tier the support is usually quite bad, but if you know the fundamentals of playing support you will be great.
The biggest problem in my solo games is that I cannot trust my team mates to know the basics of the game. It is a total dice roll every time you go. In the team games I play with my friend about 500-600 MMR points higher the players usually know the basics of how to play Dota. Nothing all that special, but they do know the basics of the game and can usually handle their heroes and participate in the game in a positive way.
I still think the MMR system could use an overhaul... And now is the perfect time for it. Update it for Season 2... Merge these two into a single rating. And it will be fine. Smurfing, boosting and all this stuff is not supposed to be going on... It is forbidden and these players should be banned. Change of the MMR system will not change this... It will not make it more rampant as people who are ready to pay are already doing it and they will be doing it after this is done. The gaming experience for these players will not change no matter what the system is.
But it would help us to play meaningful games no matter if we play solo or party... And this is what I want. I want to play games that have a meaning for me... I don't like to play games where I think it doesn't matter at all how it ends. It takes the edge off of my game and this mentality doesn't help my personal performance.
Basically your argument against this is that "people break rules"... This argument doesn't hold water... Like at all.
1) the only thing you learn from playing solo in comparison to party is how to deal with different stupid shit your team does, and that's not a marketable skill unless you want to be a booster or do well smurfing.
3) yea, but what I meant was most people don't care when they smurf. So they don't try. The ones that have to "try" every game likely shouldn't be smurfing in the first place because they don't deserve their main mmr either. Any sane person would get bored of stomping people way less skilled than them, if you were actually way above skill.
4) There are definitely decent supports in lower MMRs and can help you win games. Regardless, like I said, it doesn't matter. Their MMR will average out, and as a unit they will be balanced fairly anyways.
@Metsis - How to fix boosting/smurfing 101. Valve needs to implement a system that moves your MMR quickly (say 10-30 games) as opposed to a fucking ice age. If MMR moves quickly, boosting/smurfing dies immediately. "But marlan, a booster could get a 3k account to 6k in "10-30" games!!" k? Then the boosted players goes back to 3k in 10-30 games. Now only 20-60 games were ruined by this account, as opposed to the current system of 200-600 games. "But marlan, what if they just don't play ranked!" Well, who cares then? If you see a player with Divine 5 and NEVER plays ranked games, and his normal queue games are 2k. It's pretty fucking obvious he got boosted. It wouldn't even be hard for Valve to flag accounts with abnormally vast differences between ranked/normal for early re calibration. Congrats, now he needs to pay for a boost every time he even plays ANY 30 games because his normal MMR keeps dropping to 2k. At this point the ONLY way around the system is to simply never play dota. WOAH
Party mmr players thinking their rank means anything LUL. Either form a real team or play solo queue, ap in 5 man stacks is absolute dogshit. If they combined ranks every retard and his dog would just get boosted by their mates instead of currently where they have to pay Russians.
People thinking Party mmr doesn't mean anything LUL.
Like I swear you people are actually retarded sometimes. Party MMR isn't perfect, but neither is solo MMR. Saying it doesn't mean anything is downright stupid. Solo MMR isn't perfect either, so I guess it doesn't mean anything. And don't pretend you think Solo MMR is perfect. Grab any 2 apes from the same MMR and one will be clearly more/less skilled than the other. It takes way too long to calibrate your MMR accurately, so almost nobody is even close to being accurate except for people who grind lots of games.
Like I guess some 3k party mmr guy will just stomp some 5k party mmr guy by your logic 
Solo queue is awful, or at least it isn't fun for me. And forming a team takes way too much time, I have a job.
99.9% of dota players are not divine. Everyone else can ALREADY boost their MMR in party since it goes up to ancient 5. So, yea.
Party is just less accurate because there are variables that you can affect rather than them being all random. Party ranked que is biased but that doesn't mean it means nothing.
Please sign in to post comments.
I know I am a bottom tier player, but there is an issue with the badge system that I don't really appreciate...
I have quite a wide gap between my solo MMR and party MMR and as such, basically the lower side of the fence doesn't matter... The difference is several hundred points and I don't have time to play enough to catch up to the other rating. The thing is, when I play solo (my lower rating), it just doesn't really matter at all... I can win or lose as much as I want to and I will not gain or lose any badges for it. This is really annoying... Also when entering a game, your badge should reflect the mmr of the rating relevant to the game you are entering. As in, if you play solo, your current MMR solo badge should be shown and vice versa.
Your profile could and should show your "season high" badge as a merit badge for you to show off...
And do we really need to have two separate MMR's??? I don't think there are many players that have solo mmr 2k and party mmr 5k... A single MMR would solve a lot of the issues and there are games like Overwatch that have only a single ranking and it seems to work just fine, why couldn't it work in Dota? Sure there would some outcry when this was done, but it would settle it out in a few weeks time and I don't think anyone would miss the two different MMR's...
And I think they should add some type of a performance meter to the MMR change as well. If you have a really good game and still lose, you should lose a couple of points less and if you have a really bad game and lose, you should lose more MMR... And gain the same based on your performance. One good metric I have found is from OpenDota "Benchmarks". As in if you are playing your hero way above average, you should be rewarded for it and if you are just lucky to win and suck compared to the average performance on the hero you should gain a couple of points less than the others in the game. Some of this is in Overwatch too...
I have played some Overwatch, but am not a huge fan of the game. I just find the MMR system in it to be better than the one in Dota. Is it my fault that "I play better" when there is at least one less foul mouthed feeder russian in the game???
The point is I don't like the feeling of playing for nothing in Dota.
Another thing about this seasonal thing is that the calibration games were just called that. I have not heard a single report about a huge gain of MMR through the calibration process. Make it optional and make it a real calibration and forget the "old ratings". The season 1 calibration was just games called "calibration" and it dished out the same +25/-25... The only big changes I have seen are for the top 1% that was calibrated way lower than before.