EPTSo stupid, so fucking stupid.
This MM is a JOKE, it doesn't make any sense because stacks of players play vs solo queuers.
Ranked matchmaking coming to Dota 2 - Page 77
| Forum Index > Closed |
Please keep the QQ to a minimum if you do not like this update. We are happy to hear your reasoning for not liking a ranked system, but no "OMG VOLVO WHY" posts. | ||
|
BlitzerSC
Italy8800 Posts
So stupid, so fucking stupid. This MM is a JOKE, it doesn't make any sense because stacks of players play vs solo queuers. | ||
|
Alur
Denmark3900 Posts
| ||
|
DrPandaPhD
5188 Posts
I don't know if it is because I've played in a more "serious team" (we still sucked) and stuff but I can't take Ranked MM very serious at all. It's not even close to team vs team serious level so I don't treat it as such. I just play and enjoy the game. Atm most of the good dota players seem to queue ranked so I do too for better games pretty much. | ||
|
thOr6136
Slovenia1775 Posts
| ||
|
Kemy
105 Posts
On December 30 2013 03:13 SKC wrote: No? Ranked still has stacks, players still play it with vastly diferent mentalities and, more importantly, the context of that quote was not a game with matchmaking. If you are playing in a tournament, an arcade, bnet, etc, the amount of games you win is a good measuremente of skill. You cannot say the same when someone puts you against an equally skilled opponent every game. The premise that the mmr system puts you against equally skilled opponents every game is not true. The mmr system as it is now takes a relatively long time to adjust your rating according to your skill: - you gain 24-28 pts per win or lose 24-28 pts per loss when you soloqueue. - this is a teamgame and the individual can only do so much to win a game. Thus winrates for most players range between 45-55%. This means that even if you are on average better than equally rated players it may take hundreds of games before you gain rating and get matched with players that have an equal skillevel. Thus, winrates are not meaningless in the sense that a winrate above 50% indicates that your rating will likely increase over time while a winrate below 50% indicates that your rating will likely decrease over time. This may be the result of you improving faster than the average opponent with the same rating or because you were better from the beginning but have not played enough games yet. Of course, as your rating climbs your opponents get better which may decrease your winrate until it is averaged out at exactly 50% when you are as good (improve as fast) as everyone else on that specific rating. Winrates are only meaningless if you want to compare playerskill based on just winrates alone. You cant compare winrates because they are only an indication of the future rating development and thus make only sense if you take your current rating into the equation. | ||
TheYango
United States47024 Posts
| ||
|
Kemy
105 Posts
On December 30 2013 04:42 TheYango wrote: The problem is that there are so many things that muddle winrates. A player who has played for 2+ years creating a fresh smurf is likely to reach his current MMR with a much higher winrate than on his main account because of how that initial climb where he's not at his appropriate MMR makes up a larger percentage of his games, and because of how the vast majority of his games were played with a different player-pool than the current DotA 2 player pool (e.g. the very early beta player pool was probably much more skilled than the current one because of how pros got into the beta relatively early and in general the early beta favored players who were experienced in DotA already). This is an extreme case, but in general variations in games played and time spent playing DotA 2 make it hard for winrates to be meaningfully compared except when two players have played DotA 2 for a similar number of games over a similar span of time. That's what I said above: winrates are an indication of your personal rating development. The smurf will have a very very high winrate for the first few games (rating will increase massively and relatively fast) until he gets to the same rating as the main account and then the winrate will drop until it gets close to 50% again (rating will not change anymore because he is as good as everyone else on this level). A high Winrate alone doesnt tell you anything regarding that players skill other than his rating will/has increase(d). | ||
|
rudimentalfeelthelov
Finland268 Posts
Let's take for example a game where both teams have players with following MMRS: player 1=4500; player 2=4000; player 3=3500; player 4=3000; player 5=2500 Now let's say that on dire team player 1, player 2 and player 3 are stacked together and on radiant player 3, player 4 and player 5 are stacked together. Now I know that some people won't admit it, but let's look truth in the eyes, dire team has a huge advantage. And when you play solo with stack of 3 or 4 or two stacks of 2, if you are the highest mmr it's really hard to carry them, because stacks communicate over voice and make their own decisions without consulting with you. So for those who queue alone, the solo players only option is a necessity for balanced and entertaining games. | ||
|
Fatalize
France5210 Posts
it was 4 stack + me vs 5 stack | ||
|
Alur
Denmark3900 Posts
On December 30 2013 05:45 Fatalize wrote: Just won 32 mmr in a game, has this ever happened to someone? it was 4 stack + me vs 5 stack Think I've gotten 30+ once or twice. | ||
|
TRAP[yoo]
Hungary6026 Posts
 edit: early beta was hell for people that never played in a good team in dotaleague. too many good players at the start | ||
|
Nevuk
United States16280 Posts
| ||
|
Atreides
United States2393 Posts
Note that I am not complaining about this, because I understand the concept that as long as you never do it yourself it is 25% more likely to happen to the other team than yours! (That said, I have been "unlucky" with those percentages. :D) | ||
|
MotherOfRunes
Germany2862 Posts
On December 30 2013 03:53 Alur wrote: I hope you're in the minority. I really enjoy playing ranked with my friends without having to be 5 or having to wait a very long time. well it seems that some people dont have friends :D btw @blitzer: didnt you say you quit dota few weeks ago? | ||
|
Atreides
United States2393 Posts
| ||
|
JamazVu
Peru77 Posts
I have 3.1k mmr atm, ( 3.4k after placement / I play AllPick) but no way this is my real "level", my real mmr, Im just like hitting a wall... there is something wrong with this system, I have tried switching role, from carry to support to jungler... nothing changes PD: Just an example... 2 hours ago I got a teammate that randomed invoker( we had pudge) and he took mid... and then he said it was his first time playing invoker. im just dropping mmr with this ppl and this BAD system | ||
|
DrPandaPhD
5188 Posts
http://dotabuff.com/matches/434541174 Check radiants team. We also had an offlane deathprophet Ranked MM are still just pubs, don't take them too serious even now when there is a number to it. You are probably going to hover around +/- 300 rating for a long time. Just enjoy the game instead of focusing too much on rating. If you want faster progress you should make a team and practice more serious with them. Then after 500+ games with them you can go back to pubs and you would feel pubs being noticeably easier. Improving in pubs is always possible. | ||
|
Kupon3ss
時の回廊10066 Posts
Though the laning seems rather subpar | ||
|
Emnjay808
United States10660 Posts
Dont let the 5k standard get to you, thats literally pro level. | ||
|
TheYango
United States47024 Posts
On December 30 2013 11:18 Kupon3ss wrote: lineup seems perfectly sound; offlane PL, offensive jungle chen mid SD jungle bat solo safe DP Though the laning seems rather subpar There are also other reasonable ways to lane that if your team isn't as enamored of off-lane PL as Kupo is, and I don't think any of them involve an off-lane DP. Radiant's lineup is also manageable, if a bit greedy and rough if lanes don't go well--though judging by the items and farm, it's probably a safe bet that the Alche refused to play support which then presents a major problem. | ||
| ||
