|
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. |
On January 06 2014 02:49 DrPandaPhD wrote:Show nested quote +On January 06 2014 02:28 Andre wrote:On January 06 2014 02:03 Alur wrote:On January 06 2014 00:44 LeLoup wrote:On January 05 2014 21:35 cilinder007 wrote:On January 05 2014 04:45 LeLoup wrote:On January 05 2014 04:43 grs wrote:On January 04 2014 23:17 kaztah wrote:
After a couple of weeks with ranked and about 30 games, I must say I enjoy the whole implementation.
Overall it's exactly like the old system, except it gives us a shiny number to get a picture where we play. That's really all I wanted or needed from it.
My friends and I mostly play all pick, we mostly random 3-4 heroes, don't give a fuck about counterpicking and have a grand ol' time. The only difference is the overall number that tracks our win/loss. I don't see myself ever playing non-ranked again, because I just don't see the point.
Just one thing I would like: to be able to judge what the shiny number actually means. How to judge the meaning of something that means nothing no, I'm quite certain it does mean something Yes it is a game's best guess at how 'good' you are at it that it uses to try and make a fair match. Considering everything that goes into it, it means literally nothing because it can't physically account for the 100 plus factors that affect each individual player on a game to game basis. If rating meant nothing, 5 3k rated players would beat 5 6k rated players 50% of the time. I'm not sure about <5,5k but it seems to me that above 5,5k rating differences don't really matter much. There's only like what ~8 known people to have 6k and they don't really stand out compared to other players at 5,5k++. Above 5.5k still difference a lot. Let's say 5.5k is Grandmaster in sc2 to make it easy to compare. There is a big difference in Grandmaster and top grandmaster. And then there is a difference from top grand master to actually solid pros that does fairly well in tournaments. Then there is a difference between them and the pros who wins. Compare the bottom of NA grandmasters to the likes of Taeja Innovation etc and there is a massive difference. If you watch Sings stream, he's 6k+ and how often he destroys people mid even if they are 5.5k+. It is a huge difference in skill.
I don't think the two scenarios are easily comparable. It's not like players like Jerax don't lose mid to random 5.5k+ players.
|
United States12240 Posts
On January 06 2014 03:05 Alur wrote:Show nested quote +On January 06 2014 02:49 DrPandaPhD wrote:On January 06 2014 02:28 Andre wrote:On January 06 2014 02:03 Alur wrote:On January 06 2014 00:44 LeLoup wrote:On January 05 2014 21:35 cilinder007 wrote:On January 05 2014 04:45 LeLoup wrote:On January 05 2014 04:43 grs wrote:On January 04 2014 23:17 kaztah wrote:
After a couple of weeks with ranked and about 30 games, I must say I enjoy the whole implementation.
Overall it's exactly like the old system, except it gives us a shiny number to get a picture where we play. That's really all I wanted or needed from it.
My friends and I mostly play all pick, we mostly random 3-4 heroes, don't give a fuck about counterpicking and have a grand ol' time. The only difference is the overall number that tracks our win/loss. I don't see myself ever playing non-ranked again, because I just don't see the point.
Just one thing I would like: to be able to judge what the shiny number actually means. How to judge the meaning of something that means nothing no, I'm quite certain it does mean something Yes it is a game's best guess at how 'good' you are at it that it uses to try and make a fair match. Considering everything that goes into it, it means literally nothing because it can't physically account for the 100 plus factors that affect each individual player on a game to game basis. If rating meant nothing, 5 3k rated players would beat 5 6k rated players 50% of the time. I'm not sure about <5,5k but it seems to me that above 5,5k rating differences don't really matter much. There's only like what ~8 known people to have 6k and they don't really stand out compared to other players at 5,5k++. Above 5.5k still difference a lot. Let's say 5.5k is Grandmaster in sc2 to make it easy to compare. There is a big difference in Grandmaster and top grandmaster. And then there is a difference from top grand master to actually solid pros that does fairly well in tournaments. Then there is a difference between them and the pros who wins. Compare the bottom of NA grandmasters to the likes of Taeja Innovation etc and there is a massive difference. If you watch Sings stream, he's 6k+ and how often he destroys people mid even if they are 5.5k+. It is a huge difference in skill. I don't think the two scenarios are easily comparable. It's not like players like Jerax don't lose mid to random 5.5k+ players.
Rating differences don't guarantee outcomes. If the rating difference of two mid players is like 500, then that's going to translate to what, a 75/25 probability between them? It's never 100/0, and everybody knows there's a lot more than just a 1v1 mid that transpires in the game. If you're favored to win then you're favored, not assured.
|
Well it's not like jerax is supposed to have a lower rating
|
On January 06 2014 03:28 Erasme wrote:
Well it's not like jerax is supposed to have a lower rating
There are many players who are a lot better than him and have lower rating though which strengthens the argument that MMR is not meaningful after a certain threshold.
|
Doesn't matter when he has mastery over Earth Spirit
|
On January 06 2014 03:34 gaymon wrote:Show nested quote +On January 06 2014 03:28 Erasme wrote:
Well it's not like jerax is supposed to have a lower rating There are many players who are a lot better than him and have lower rating though which strengthens the argument that MMR is not meaningful after a certain threshold.
More like people don't understand what MMR is supposed to be. If having Jerax on your team playing ES brings you as much as Dendi playing Sniper would bring, so be it. Jerax is a case of MMR working as it should, and showing why it isn't a measure of "skill".
|
On January 06 2014 04:03 SKC wrote:Show nested quote +On January 06 2014 03:34 gaymon wrote:On January 06 2014 03:28 Erasme wrote:
Well it's not like jerax is supposed to have a lower rating There are many players who are a lot better than him and have lower rating though which strengthens the argument that MMR is not meaningful after a certain threshold. More like people don't understand what MMR is supposed to be. If having Jerax on your team playing ES brings you as much as Dendi playing Sniper would bring, so be it. Jerax is a case of MMR working as it should, and showing why it isn't a measure of "skill".
MMR measures your ability to win pub games, if dendi picks sniper every game that effectively makes him a worse player. Obviously the game can't account for dendi not caring about pubs, not even 1v1 games can have a rating system that does that.
|
I guess it depends on a game. In sc2 you had people 6 pooling their way to gm. 6 pool was their Earth Spirit in a sense. Maybe chess has a true measurement of skill. But even there you have some cheese openings i guess.
|
On January 06 2014 05:07 thOr6136 wrote:
I guess it depends on a game. In sc2 you had people 6 pooling their way to gm. 6 pool was their Earth Spirit in a sense. Maybe chess has a true measurement of skill. But even there you have some cheese openings i guess.
Chess is a game with perfect information, SC2 or Dota isn't. This adds an uncertainty that makes "cheesy" things stronger, but technically people getting away with them is just a case of the game not being developed very well. The more the overall skill in the population playing the game rises the less it's possible to "cheese" someone out of a win.
|
also chess is alot less real timey so you have alot more time to not make mistakes if you were in theory capable of not doing them
|
On January 06 2014 04:12 Pokebunny wrote:Show nested quote +On January 06 2014 04:03 SKC wrote:On January 06 2014 03:34 gaymon wrote:On January 06 2014 03:28 Erasme wrote:
Well it's not like jerax is supposed to have a lower rating There are many players who are a lot better than him and have lower rating though which strengthens the argument that MMR is not meaningful after a certain threshold. More like people don't understand what MMR is supposed to be. If having Jerax on your team playing ES brings you as much as Dendi playing Sniper would bring, so be it. Jerax is a case of MMR working as it should, and showing why it isn't a measure of "skill". MMR measures your ability to win pub games, if dendi picks sniper every game that effectively makes him a worse player. Obviously the game can't account for dendi not caring about pubs, not even 1v1 games can have a rating system that does that.
Pretty much a rough system to measure what he said, the ability to win pub games. As there is no real way of putting a number to something like player skill, even if one played only a certain hero every game other factors are also variable like team composition, and skill levels of other players, matchups in lanes, team composition vs team composition. Dont get why people are trying to use it as a measure of skill when its just a number to attempt to match you more effectively against people supposedly at your skill level.
|
On January 06 2014 05:13 r.Evo wrote:Show nested quote +On January 06 2014 05:07 thOr6136 wrote:
I guess it depends on a game. In sc2 you had people 6 pooling their way to gm. 6 pool was their Earth Spirit in a sense. Maybe chess has a true measurement of skill. But even there you have some cheese openings i guess. Chess is a game with perfect information, SC2 or Dota isn't. This adds an uncertainty that makes "cheesy" things stronger, but technically people getting away with them is just a case of the game not being developed very well. The more the overall skill in the population playing the game rises the less it's possible to "cheese" someone out of a win.
Well Chess also has the problem of the majority of the games resulting in draws, and upsets are especially rare. So there's obviously a balance somewhere.
|
|
|
Hard to win as much when you stop playing ES every game
|
i think it had more to do with the 15+min queues he had or something
|
On January 06 2014 20:22 Fatalize wrote:Hard to win as much when you stop playing ES every game
I'd say playing 15 ranked within the last 4 days is defenitly playing ranked. More than what I've played anyway...
|
Yup if you watched the stream you'd know he just hates queues, and they often went to 20+mins for him which is not at all helpful if he's trying to stream. So he's just playing unranked.
|
On January 06 2014 21:37 AndyJay wrote:
Yup if you watched the stream you'd know he just hates queues, and they often went to 20+mins for him which is not at all helpful if he's trying to stream. So he's just playing unranked.
they actually fixed queue times. but now u just get 3~4k rated players on ur team and lose coz u cant beat the 5 5.5ks alone.
every game is basically a 1v5 and ur brown/orange player single handedly loses u the game before creeps even spawn.
also my rating is broken, i win and get +1, i lose and get -49...
|
Thats what i wanted to know, do you know if you're the only one ? I don't really watch sing or jerax
|
|
|
|
|
|
|
|
EPT
