EPTGeneral Discussion - Page 2337
| Forum Index > Dota 2 General |
IRC chatter should remain in IRC - http://webchat.quakenet.org/?channels=tl.dota2 Posts that relate to topics with their own thread, such as in News, Tournaments or Strategy should go in those threads. | ||
|
eluv
United States1251 Posts
| ||
|
NB
Netherlands12045 Posts
On March 09 2013 18:27 Shaella wrote: you're assuming the goal of the matchmaking system is to enforce 50% WR In reality, its trying to make equal matches happen, every time, so if you reach that proper skill level, your winrate should stabilize and slowly drop towards 50% as a result of the stabilization Seriously. Are you stupid? Its not the goal, its a side effect. And as we all know, WR =/= skill level or MMR. Seriously? Insulting people out of a question? And yes, i do know that it is a side effect. But that simply means that the answer to the original question is yes assuming that you are playing solo queue exclusively. The only thing that is disrupting such phenomenal right now is party MMR which will never be calculated correctly. I have never said WR = skill level or MMR. What i said was essentially exactly what you said, worded differently.... On March 09 2013 18:31 eluv wrote: What do you mean by "enforce"? If you mean that the system tries to match you against people of equal skill, and that winning increases its estimation of your skill, causing it to match you against tougher opponents - then yes? But once you've got more than a couple hundred games, winning 5 in a row really isn't going to be so significant that it's going to "try to make you lose." Fundamentally, you can choose to look at it as the system being out to get you, or you can see it as an attempt to match you to the most appropriate opponents based on your in-game performance and in the end the result is the same. Only your attitude towards it changes. enforce is a strong word in this case but i feel like it is appropriate since it is such a common result for a skill based match making system to generated 50% in a big sample size. Looking at starcraft2 for example, once you win way too much in Grandmaster league, you will end up with gaining 1 point for each win and losing 20, 50 times more points for every lose. In League of legend, top players could even lose point after a win(which was identified as a know bug but was generated by the nature of the match making system). If the match making system of dota2 is skill based(which we know so far that it is at least is influenced by), then the eventually result of a solo queue player with a big sample size will be 50% winrate. As i mentioned above, i think the only reason we are seeing otherwise is either people have yet to play enough games to reach the right sample size OR using the party MMR system where it is impossible to calculate accurately. | ||
Firebolt145
Lalalaland34507 Posts
There are two different approaches to MM: - The MM system tries to match you up against players of your skill level, and thus you average towards a 50% winrate (but not always) - The MM system tries to force you towards a 50% winrate, and as a side effect deliberately tries to screw you over when you get winstreaks etc The first theory is correct, the second one is completely false. | ||
|
NB
Netherlands12045 Posts
On March 09 2013 18:40 Firebolt145 wrote: You said a bunch of stuff about how MM forces you into a 50% winrate and also deliberately queues you into stacks if you're in a winstreak. That stuff is all very wrong and it's annoying that it's a widely held belief in the Dota community. There are two different approaches to MM: - The MM system tries to match you up against players of your skill level, and thus you average towards a 50% winrate (but not always) - The MM system tries to force you towards a 50% winrate, and as a side effect deliberately tries to screw you over when you get winstreaks etc The first theory is correct, the second one is completely false. i was under the assumption of both theory are the same thing in term of final result(which i assumed what the original question was about). And yes, i agree that the 2nd one is false. | ||
|
Andre
Slovenia3523 Posts
I don't think the MM system is purely skill-based because of it. | ||
|
BurningSera
Ireland19621 Posts
While i am 100% sure about queue time has nothing to do with winning/losing streak because it doesnt matter what happen nowadays my solo queue time is almost always 4/5+mins, and it can go up to 16mins -_- (not server down but not peak hour either). I stop caring about MM anymore after so many games of facing those stacks (again nothing wrong with tryharding but if you watched notail stream you would know what i mean and we have faced the same finstack so many fcking times). Games are so damn boring and it doesn't even matter what you pick now, there is almost always a mofo in team who will screw the team one way or another (while obviously the oppo stack with competitive line up). AR all the way when i solo queue hell yeah! | ||
|
BurningSera
Ireland19621 Posts
On March 09 2013 18:47 Andr3 wrote: ^Doesn't the MM system also take into consideration the amount of wins/games you have? Looked at separately it might not matter, but it should be a factor. I don't think the MM system is purely skill-based because of it. I notice this too, they must have changed something after many people whined about having team mate with 10wins while everyone has 1000wins etc lol | ||
|
RuskiPanda
United States2906 Posts
On March 09 2013 18:54 BurningSera wrote: I notice this too, they must have changed something after many people whined about having team mate with 10wins while everyone has 1000wins etc lol Yeah after playing almost 1k games I randomly check people's accounts out of curiosity during load time and its invariably a collection of level 30-60s with 500+ games played. Only had a few obvious smurfs with like 7 games total that were probably queing with a stack. Not that I mind since even if its like page 20 on Live games everybody still knows what they're doing and games are usually relatively close. | ||
|
FinestHour
United States18466 Posts
| ||
|
Cool Cat
United States1644 Posts
| ||
|
cecek
Czech Republic18921 Posts
 They set up all this cool shit and then have huge technical difficulties and the stream is barely watchable. | ||
|
WindWolf
Sweden11767 Posts
| ||
|
Grettin
42414 Posts
On March 09 2013 22:12 WindWolf wrote: Casters have been talking about changes to the jungle recently. What are they talking about and what patch was the change made? Last patch i guess(or very recently anyway). It's mostly to prevent or at least make it harder to pull mid. | ||
|
cecek
Czech Republic18921 Posts
On March 09 2013 22:26 Grettin wrote: Last patch i guess(or very recently anyway). It's mostly to prevent or at least make it harder to pull mid. You can't pull the small camp to mid on Radiant, and you can't pull the small camp to easy lane on Dire. | ||
|
Talron
Germany7651 Posts
lolwut. Is it going to be awesome or awesome? | ||
|
zz_
Sweden1022 Posts
On March 09 2013 23:40 cecek wrote: You can't pull the small camp to mid on Radiant, and you can't pull the small camp to easy lane on Dire. Sure you can, it's just harder to pull off consistently. | ||
|
cecek
Czech Republic18921 Posts
On March 10 2013 00:01 zz_ wrote: Sure you can, it's just harder to pull off consistently. Yeah. I think you can do it just with some creep camps but not all, or something. Basically means you can't rely on it, that's the important part. | ||
|
SilverStar
Sweden18511 Posts
On March 10 2013 00:00 Talron wrote: Take considering Homestory for Dota2: http://www.gosugamers.net/starcraft2/news/23114-take-i-am-considering-dota-2-as-homestory-discipline?utm_source=twitterfeed&utm_medium=twitter lolwut. Is it going to be awesome or awesome? He already did it with lol. Was kinda meh. Maybe if its only 4 teams it would work better. | ||
|
Grettin
42414 Posts
On March 10 2013 00:00 Talron wrote: Take considering Homestory for Dota2: http://www.gosugamers.net/starcraft2/news/23114-take-i-am-considering-dota-2-as-homestory-discipline?utm_source=twitterfeed&utm_medium=twitter lolwut. Is it going to be awesome or awesome? I was just going to post about the interview. To be honest, HomeStoryCup has always been one of the entertaining events in StarCraft, so i wouldn't mind seeing Dota 2 there. The only concern i have how well team games (and teams) stood up compared to StarCraft which has individuals playing. Interview here, HSC talk starts around 6:40. (Dota 2 part, 20:40 and forward) | ||
|
SilverStar
Sweden18511 Posts
| ||
| ||
