|
On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." when stats clearly shows there are over 10% more Toss players in GM and Masters than it suppose to be. Beside, on all leagues except Bronze Terrans are least performing race ?
GM league Toss players -
America 94/199
Europe 92/200
Korea 81/198
Taiwan 88/198
SE Asia 7/13
China 82/198
Global 444/ 1006
44% Protoss global GM
Masters League -
US
P - 37%
T - 24%
Z - 36%
KR
P - 36%
T - 28%
Z -32%
EU
P - 37%
T - 27%
Z - 34%
On what basis Blizzard states " "There isn't an exaggeration of Protoss players."" ?
At this point, I have no idea if Blizzard's team is blinded by ignorance, a lack of resources, or simply stupidity.
|
On December 19 2013 00:20 Hadronsbecrazy wrote:
I think the leagues are pretty retarded to be honest, I keep facing low gold players whom i beat really easily or diamond/masters players who beat me really easily and im still gold. Sort this shit out Blizzard , either give me plat or players my own skill level
this. it became so frustrating i stopped playing all together T.T
|
On December 19 2013 18:04 paralleluniverse wrote:
A few other observations:
1. There is nothing wrong with MMR decay. If you're inactive, your skill falls. Therefore, your MMR should fall too.
2. Unlike Valve, Blizzard almost never releases information about their ladder system. In fact, this is the very first time they've EVER revealed non-obvious information about how a part of their ladder system works.
3. It seems that the best way to get Blizzard to release information about their ladder system is to spread rumors and misinformation and to get people to use those rumors and misinformation as a scapegoat for their grievances.
We should exploit this, ruthlessly.
I agree that the MMR decay is nothing bad by default, but we are lacking data to judge it. I don´t think that the ladder is as linear as we think it will be more like a wave with downs like release of big other games like gta5 or end of seasons. Or on the other end of the scale after patches or big tournaments i can imagine more people actual laddering.
That can create problems like some people mentioned. Devoted in weird leagues, floods of former master players in platin. To me it looks like the MMR decay is a good thing for individual players or small numbers but gets worse if the number of influenced players is getting too big.
A possible solution could be to adapt the custom games to an iccup´ish system. You can choose a map, a matchup and the process of the player search is as transparent as possible to explain better why you are playing whom.
What Blizzard really missed is on the PR side of the ladder. The threads on the topic are full of "i was master and now i am gold and i hate Starcraft". The league. Doesn´t. Matter. It has nothing to do with the hidden MMR but thats not the point it is a psychological / motivation thing that people don´t like to be "in a lower league".
So what they had to do to avoid all the whining is to rename all the leagues completely. So instead of Diamond or Master we would have id don´t know A call and S class or something else. Just renaming it would have helped immensely.
And on the "Valve does it better", Blizzard just looks like a "old" company right now. They hide as many informations as possible how things work, almost no communication with the community and slow reaction on trends in esports and on making/selling games. The Titan desaster and the complete miss of the F2P trend in the recent years are a pretty big proof of that.
|
On December 19 2013 04:01 captainwaffles wrote:
Being a 15 times masters finisher I was placed into diamond this season and it took 15 games to get back. I don't see what the big deal is the ladder should be hard.
On December 19 2013 04:08 darthfoley wrote:
Yea I was high diamond in WOL. Haven't played very actively at all in the past year with college starting and all that.
Recently decided to get back into HOTS, got placed directly into bronze. Basically beat everyone back into high gold, which is apparently a good ceiling for me right now. It only took my about 15 games to get back to a comfortable level of challenge. Dunno why it's fucking so many other people over
you guys only see your perspective. for a player that is inactive and gets demoted it is not that much of a problem to get back to his old level/ league. but the problem most ppl complain about is the following.
consider an active playing guy who is currently in gold (which is a good representation of his skill). at the moment he will face quite a lot of players that are much better than him, although he plays active. so you see, it's not his activity that creates the problem, its the inactivity of the other players. there will always be a master lvl player that took a break and has to work his way back up where he belongs. to the master lvl player the 90% win rate is not the problem. but to the players he hits in the lower leagues, the players that actually belong into those lower leagues, its a problem.
did that make sense? 
|
On December 19 2013 03:51 Excalibur_Z wrote:
That's exactly what it means. If a season is 3 months long and you play on day 1, day 29, and day 57, then by the end of the season you will have decayed around three leagues' worth of rating.
How can someone implement a system where playing a few games every months decays more than playing no games at all is beyond me (I was in low master/high dia mmr and I played a few unranked games every month and now I am facing bronze players...)
|
On December 19 2013 18:11 ander wrote:Show nested quote +On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." when stats clearly shows there are over 10% more Toss players in GM and Masters than it suppose to be. Beside, on all leagues except Bronze Terrans are least performing race ?
GM league Toss players -
America 94/199
Europe 92/200
Korea 81/198
Taiwan 88/198
SE Asia 7/13
China 82/198
Global 444/ 1006
44% Protoss global GM
Masters League -
US
P - 37%
T - 24%
Z - 36%
KR
P - 36%
T - 28%
Z -32%
EU
P - 37%
T - 27%
Z - 34%
On what basis Blizzard states " "There isn't an exaggeration of Protoss players."" ? At this point, I have no idea if Blizzard's team is blinded by ignorance, a lack of resources, or simply stupidity.
Or maybe, it's the "I'm-smarter-than-Blizzard" misinterpreting things AGAIN.
He was responding to someone who claimed that 90% of active ladder players are Protoss which implies that the guy is facing only Protoss players when he plays and thus doesn't get to play any other match up.
And that's just not true. There is an even number of active Protoss players as any other race.
|
On December 19 2013 18:50 gondolin wrote:Show nested quote +On December 19 2013 03:51 Excalibur_Z wrote:
That's exactly what it means. If a season is 3 months long and you play on day 1, day 29, and day 57, then by the end of the season you will have decayed around three leagues' worth of rating. How can someone implement a system where playing a few games every months decays more than playing no games at all is beyond me (I was in low master/high dia mmr and I played a few unranked games every month and now I am facing bronze players...)
As a software developer 'm not at all surprised. What's really needed is a new variable stored in every player profile that keeps track of accumulated decay. This however would require a change in data structures, databases and communication protocols. It would take time and would be expensive. They wanted an easy cheap fix that could just be slapped on to the current system with no other changes, and that's what it is.
|
On December 19 2013 19:06 Mendelfist wrote:Show nested quote +On December 19 2013 18:50 gondolin wrote:On December 19 2013 03:51 Excalibur_Z wrote:
That's exactly what it means. If a season is 3 months long and you play on day 1, day 29, and day 57, then by the end of the season you will have decayed around three leagues' worth of rating. How can someone implement a system where playing a few games every months decays more than playing no games at all is beyond me (I was in low master/high dia mmr and I played a few unranked games every month and now I am facing bronze players...) As a software developer 'm not at all surprised. What's really needed is a new variable stored in every player profile that keeps track of accumulated decay. This however would require a change in data structures, databases and communication protocols. It would take time and would be expensive. They wanted an easy cheap fix that could just be slapped on to the current system with no other changes, and that's what it is.
That Excalibur_Z guy is just theorizing. No one knows exactly how it works except Blizzard.
|
On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players."
Probably based on historical comparisons in numbers. As has been pointed out repeatedly (and flat out ignored by Terrans trying to whine about balance) Protoss has always had a strong presence on the ladder, even when Terran was so strong it was roflstomping absolutely everybody in Code S. This is due to the nature of Protoss in blind best-of-one format against unknown players along with the diversity of Protoss all-ins which tend to be amazing on the ladder.
Therefore if the current ladder data shows no statistically significant difference to that historical data then there is nothing to worry about. No matter how much people cry about the percentages there just isn't a problem.
|
On December 19 2013 18:57 kheldorin wrote:Show nested quote +On December 19 2013 18:11 ander wrote:On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." when stats clearly shows there are over 10% more Toss players in GM and Masters than it suppose to be. Beside, on all leagues except Bronze Terrans are least performing race ?
GM league Toss players -
America 94/199
Europe 92/200
Korea 81/198
Taiwan 88/198
SE Asia 7/13
China 82/198
Global 444/ 1006
44% Protoss global GM
Masters League -
US
P - 37%
T - 24%
Z - 36%
KR
P - 36%
T - 28%
Z -32%
EU
P - 37%
T - 27%
Z - 34%
On what basis Blizzard states " "There isn't an exaggeration of Protoss players."" ? At this point, I have no idea if Blizzard's team is blinded by ignorance, a lack of resources, or simply stupidity. Or maybe, it's the "I'm-smarter-than-Blizzard" misinterpreting things AGAIN. He was responding to someone who claimed that 90% of active ladder players are Protoss which implies that the guy is facing only Protoss players when he plays and thus doesn't get to play any other match up. And that's just not true. There is an even number of active Protoss players as any other race.
And this statement of yours is also simply not true. According to sc2ranks the overall distribution of players might be close to even (there are roughly 4.5k less T players compared to P overall), but as you go up the leagues, you should notice an increasing trend of fewer T players, resulting in a huge difference of ~10% between T and P in GM/Masters, whereas in a balanced situation the numbers should be way more equal.
|
On December 19 2013 14:13 storywriter wrote:Show nested quote +On December 19 2013 03:57 Incubus1993 wrote:On December 19 2013 00:38 storywriter wrote:
I think the 6% figure is if they only count people who are active. A lot of people play very few games per season and have kept their leagues that way for a long time . That doesn't mean those people have gotten so bad they need to be placed two or three leagues below their original league. They should be demoted though, their slots should be open for people ACTUALLY PLAYING the game. There are so many "Masters" players that play their placement game and leave it at that for the entire season. This inactivity used to inflate masters league but since league placement only takes into account active players, no one would have been robbed of their spot in a higher league. The inflation was kinda annoying and I get the reasoning behind MMR decay but at the moment, it must be horrible to play on the ladder as a lower ranked player.
Only grandmaster league has spots. Other leagues use static league MMR thresholds. These thresholds changed last time in the beginning of June (or if there has been changes since they have been very small). If your MMR rises above a certain league threshold you will be promoted regardless how many other players there already are in that league.
On December 19 2013 16:54 Danglars wrote:I really view this as a reaction against the research TLers have done into MMR decay. They downplay its widespread effects and pound it in time after time. It's MMR decay and the inability to be demoted out of leagues (the adjusted promotion metric). Let's look forward to their changes to fix: Show nested quote +For various reasons since that announcement, the distribution of players across leagues slowly shifted to no longer match those desired targets. Currently the lower leagues like bronze and silver have a larger percentage of players than desired. Meanwhile the upper leagues like platinum and diamond are under-represented. One of the reasons this occurred is due to the way that we maintain those target percentages. Altogether too many paragraphs spent on Show nested quote +Please don't think matchmaking screwed up for inactive players or that we're to blame for tougher lower leagues Blizzard can do better, even in blog posts.
Yes. Blizzard has a long history of using 'PR / marketing language' in their statements. They downplay potential problems in their statements to a point where it often becomes misleading. And masses accept those statements without thinking themselves (for example there is already consensus in the main Reddit thread that MMR decay adjustment is only worth of 'few games').
For example statement: "At its maximum value, the adjustment is small; it’s the equivalent of losing a few games." The actual adjustment is equivalent of about ~20 losses (unless they have changed it silently during this season). There is plenty of data to back this up (MMR tool data for several seasons). Also if people think logically they could deduct it themselves, that it is not "slight" adjustment.
For example if we think about a character account that only plays one placement match during each season. If that character was two seasons ago diamond, it would have placed either in platinum or gold last season. This season it would have placed either in silver or gold (slight change of high bronze too). If the adjustment each time was indeed worth of 'few games' the league system would not matter. It would mean that to get from bronze to diamond you would need only ~10 wins (if few would mean e.g. '2'). In reality to pass each league's MMR range typically requires ~20 wins more than losses (Size of MMR ranges for different leagues differ slightly).
|
MMR decay is not the cause of this mess of leagues I think.
The main reason is the diminution of player base + the unranked play.
1) noob players quit so better players are ranked in low league belonging to newbs that left
2) people play more and more unranked therefore MMR is more inacurate. remember that MMR as any learning machine algorithm is more performant when he have a lot of data. If you play often you will have a pretty accurate MMR but you can't escape to face people with a pretty inaccurate MMR of the same level as yours. This explain the I easily beat this guy or this guy was really better than me.
The only thing that doesn't make sense to me is the league distribution because in theory it should have stay close to what they desire.
|
On December 19 2013 19:41 -Celestial- wrote:Show nested quote +On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." Probably based on historical comparisons in numbers. As has been pointed out repeatedly (and flat out ignored by Terrans trying to whine about balance) Protoss has always had a strong presence on the ladder, even when Terran was so strong it was roflstomping absolutely everybody in Code S. This is due to the nature of Protoss in blind best-of-one format against unknown players along with the diversity of Protoss all-ins which tend to be amazing on the ladder. Therefore if the current ladder data shows no statistically significant difference to that historical data then there is nothing to worry about. No matter how much people cry about the percentages there just isn't a problem.
No. So much protoss on ladder is kinda new. Stop making up fact please.
|
On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." when stats clearly shows there are over 10% more Toss players in GM and Masters than it suppose to be. Beside, on all leagues except Bronze Terrans are least performing race ?
GM league Toss players -
America 94/199
Europe 92/200
Korea 81/198
Taiwan 88/198
SE Asia 7/13
China 82/198
Global 444/ 1006
44% Protoss global GM
Masters League -
US
P - 37%
T - 24%
Z - 36%
KR
P - 36%
T - 28%
Z -32%
EU
P - 37%
T - 27%
Z - 34%
On what basis Blizzard states " "There isn't an exaggeration of Protoss players."" ?
Very simple. Ignore the Master/GM phenomenon and take the whole player base worldwide:
Race Statistics (Total):
T 29.25%
(77,615 Users)
Z 30.28%
(80,349 Users)
P 31.39%
(83,305 Users)
R 9.09%
(24,119 Users)
And lo, magically the problem disappeared. If someone points out Master/GM again, insert a finger in your ears and start singing very loudly.
|
On December 19 2013 23:08 TheDwf wrote:Show nested quote +On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." when stats clearly shows there are over 10% more Toss players in GM and Masters than it suppose to be. Beside, on all leagues except Bronze Terrans are least performing race ?
GM league Toss players -
America 94/199
Europe 92/200
Korea 81/198
Taiwan 88/198
SE Asia 7/13
China 82/198
Global 444/ 1006
44% Protoss global GM
Masters League -
US
P - 37%
T - 24%
Z - 36%
KR
P - 36%
T - 28%
Z -32%
EU
P - 37%
T - 27%
Z - 34%
On what basis Blizzard states " "There isn't an exaggeration of Protoss players."" ? Very simple. Ignore the Master/GM phenomenon and take the whole player base worldwide: Race Statistics (Total): T 29.25% (77,615 Users) Z 30.28% (80,349 Users) P 31.39% (83,305 Users) R 9.09% (24,119 Users) And lo, magically the problem disappeared. If someone points out Master/GM again, insert a finger in your ears and start singing very loudly.
You are missing the point. High skilled top players represents the balance and how well the game was designed, therefore, GM and MASTERS are the most important part of the ladder.
|
On December 19 2013 04:00 saltis wrote:
You are missing the point. High skilled top players represents the balance and how well the game was designed, therefore, GM and MASTERS are the most important part of the ladder.
That was sarcasm. He got the point.
|
On December 19 2013 23:02 Faust852 wrote:Show nested quote +On December 19 2013 19:41 -Celestial- wrote:On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." Probably based on historical comparisons in numbers. As has been pointed out repeatedly (and flat out ignored by Terrans trying to whine about balance) Protoss has always had a strong presence on the ladder, even when Terran was so strong it was roflstomping absolutely everybody in Code S. This is due to the nature of Protoss in blind best-of-one format against unknown players along with the diversity of Protoss all-ins which tend to be amazing on the ladder. Therefore if the current ladder data shows no statistically significant difference to that historical data then there is nothing to worry about. No matter how much people cry about the percentages there just isn't a problem. No. So much protoss on ladder is kinda new. Stop making up fact please.
People have been complaining about Protoss overrepresentation on ladder since almost the start of WoL. Your assertion that "no its a new thing" is the new thing. Again, if its not a statisically significantly difference from past numbers then it isn't something to be worried about. Stop making up things to allow you to keep whining please.
|
On December 20 2013 01:28 -Celestial- wrote:Show nested quote +On December 19 2013 23:02 Faust852 wrote:On December 19 2013 19:41 -Celestial- wrote:On December 19 2013 04:00 saltis wrote:
Can somebody explain, how according Blizzard representative Kaivax "There isn't an exaggeration of Protoss players." Probably based on historical comparisons in numbers. As has been pointed out repeatedly (and flat out ignored by Terrans trying to whine about balance) Protoss has always had a strong presence on the ladder, even when Terran was so strong it was roflstomping absolutely everybody in Code S. This is due to the nature of Protoss in blind best-of-one format against unknown players along with the diversity of Protoss all-ins which tend to be amazing on the ladder. Therefore if the current ladder data shows no statistically significant difference to that historical data then there is nothing to worry about. No matter how much people cry about the percentages there just isn't a problem. No. So much protoss on ladder is kinda new. Stop making up fact please. People have been complaining about Protoss overrepresentation on ladder since almost the start of WoL. Your assertion that "no its a new thing" is the new thing. Again, if its not a statisically significantly difference from past numbers then it isn't something to be worried about. Stop making up things to allow you to keep whining please.
Source or you lie. I never felt that there were that much protoss in GM/ML. And I'm pretty sure i'm right.
|
United States12240 Posts
On December 19 2013 22:46 klup wrote:
MMR decay is not the cause of this mess of leagues I think.
The main reason is the diminution of player base + the unranked play.
1) noob players quit so better players are ranked in low league belonging to newbs that left
2) people play more and more unranked therefore MMR is more inacurate. remember that MMR as any learning machine algorithm is more performant when he have a lot of data. If you play often you will have a pretty accurate MMR but you can't escape to face people with a pretty inaccurate MMR of the same level as yours. This explain the I easily beat this guy or this guy was really better than me.
The only thing that doesn't make sense to me is the league distribution because in theory it should have stay close to what they desire.
The distribution is correct in terms of what they're currently recording, which is based on activity in terms of unspent bonus pool. If you were to filter out all players who have more than ~2 weeks unspent bonus pool, you would see the percentages as pretty close to (but not exactly) 2/18/20/32/20/8. The problem is that decay records inactivity in a different way, so as long as their existing metric doesn't factor that in, they wouldn't have been able to detect it.
|
The response to threads like these lead me to despise the player base. Over-entitled gamers, desperate to blame anything and everything on game developers.
It's an MMR rating system that matches you up with roughly equivalent skilled people. It's impacted by a steadily declining player-base, and the fact that if there wasn't a decay in place, there would be plenty, if not more, 'fuck blizzard' posts talking about how the game is stupid and the experience is shit for not factoring in
The vase majority of the whining is just symptoms of the usual Dunning-Kruger effect, leading to selection bias and over-entitlement.
|
|
|
|
|
|
EPT
