I agree with triple. The possibility has definitely increased. But no where does increased queue time confirm anything of what hidden pool retards are saying. Of course, I'm not actually an unreasonable retard and stand to be corrected.
It's just fucking skilless retards using it as an excuse for their own behaviour triggers me
ywn, I did not know top 1% in Dota is bad. I got it in 600 games too, which is less than most people. Keep crying so I can start a salt farm please.
Jacked, I never said the LC was toxic. You assumed that because you are a biased moron. Hidden pool is approximately 30% toxic players, 1% game ruiners, 69% bad players. The main problem is getting bad players. You assumed toxicity was the main problem because you were biased once again.
8 months ago, I literally got low priority every 5 to 10 ranked games (confirmed from my match history) for months, which was why I had to make a lp removal bot. I posted about hidden pool, and there you were to spout nonsense about how my claims were false. I definitely had hidden pool, but you denied it because you are a biased moron. This was not about whether both team had hidden pool players but whether it actually existed.
If players in hidden pool could only get each other, toxic or game-ruining professional players with high mmr, like Cncc, would basically never find games. To get 10 hidden pool players at 9k mmr is unrealistic.
Triplesteal is not so bad after all. I will stop calling him Trippleretard.
It only says that there's an extra search parameter based on behaviour score, not how heavily this parameter is weighted. The system will probably try match people with similar behaviour scores but we don't know what the range is and how much priority is given to it. It might be that there is a cutoff and people with a behaviour score difference that is too big will never be matched together, or it could be that it will group them together if there are no other people to match.
The post doesn't confirm hidden pool but it makes its existence more likely. For a low BS person to be in a hidden pool it would have to be impossible for them to get games with high BS people but that's not confirmed. The only thing that's confirmed is that its probably less likely for them to get matched with high BS people but it might still be possible if there are not enough other low BS people searching.
i dont know why you keep arguing with someone that is called ' trippleretard' ' ywn' or ' jackishited' doesnt make sense rly . lul. i mean i teached them same leeson 1000 times over and over again, with proves from GABEN itself lul, they dont understand. deal with it. why it bothers you? jsut try do understand them , dont hate them, they victims not bad persons. ' obssesion in mmr' sure has it own side effects. some person can handle that, some not. those 3 not at all. im not even trolling this time.
fact of the matter is if i played any of your low behaviour shit accounts i would still win most of my games and you dont cause youre bad and instead of tryign to get good you flame eveyrone else
More like you got nothing. Dreng gets 10 minute queues at 2.7k mmr with hidden pool, which is 100 points above the average mmr in Europe. Now, Cncc has 9k mmr and is in the top ~0.0000026% of the mmr distribution. The conduct summary says 23% of players have more than 2 reports. Assuming hidden pool is based on reports, his matchmaking pool would further be split by roughly 5 times. This would place him with 0.0000005% of players. Basically, it would take him very long to find a game. Allowing the game to have a non-hidden-pool team vs a hidden pool team would "halve" his queue time.
Now that I consider it, if it was only hidden pool together, hidden pool players would still be playing harder games. The hidden pool population is ~1/5 of the normal population. This means there would be inflation in the normal population when compared to the hidden pool.
the calculations make no sense due to much higher mmr disparity of higher level games
as a 2.7k u get games 2.6 to 2.8 average, as a 9k u queue into high 5ks sometimes. The MM prioritizes matches closer to your mmr over those far away from it, but it´s not clear how much weight it puts onto this parameter, or, in other words, at which point does the mm "give up" and put 9ks and 4ks together
yeah where did yo uget this lul. how can you know that? like how can you know , ' if rain is created by clouds, it will fall down into earth? how can you know that? waht kind of genius are you ? jsut cause you not named tripple retard doesnt mean you genius brah lol.
funny how are we constantly swinging between
"hidden pool doesnt exist, you are bad" and "existence of hidden pool was confirmed long ago this is nothing new"
yeah where did yo uget this lul. how can you know that? like how can you know , ' if rain is created by clouds, it will fall down into earth? how can you know that? waht kind of genius are you ? jsut cause you not named tripple retard doesnt mean you genius brah lol.
Seriously tho it based on an account flag, meaning when you meet the requirement your account be flagged. The question is how long does the account flag last.
I have 3 accs with corresponding que times, one with 20-30 minute que times at 2.7, this one with 15 minute ques at 2.9, and then another account with 4 minute ques at 2.6
Also on the 30 minute que time For confirmed that all my team mates had 15 minute or more ques and the all the enemies had 5 minute ques. then on my 4 minute que alt account, I was matched against several of the players who were on my team on the long que time account, and stomped them to Oblivion but never got them on my team on the short que account.
I got the numbers from the references and the conduct summary from Valve. Actually, Cncc’s exact percentile is unknown because there is a rounding error of 0.5*10^5. Though, the exact percentile did not really matter. The point was to show Cncc had a significantly smaller matchmaking pool. If rampageacademy (drengr(?)) has 10 minute queues at the average mmr, Cncc would have absurd queues. This leads me to believe Valve will allow a hidden pool player to match with other normal players if the queue takes too long similar to how Triplesteal said the system would ‘"give up" and put 9ks and 4ks together’.
No-one outside of Valve knows the matchmaking system but the calculation was like this. Since hidden pool players are x5 rarer, it would take x5 longer to find one. When someone queues, that person must find 9 other players. If it was hidden pool vs non-hidden-pool, 4 hidden pool players would be needed and 5 normal. If it was hidden pool vs hidden pool, 9 hidden pool players would be needed. If x is the time needed to find a normal player, 5*x is the time for a hidden pool player. There are 2 cases: the system simultaneously searches for all players or the system searches for each player individually. If the system searched for each player individually, the time needed to create a hidden pool vs hidden pool game would be, 5*x*9 = 45*x. The time for a non-hidden-pool vs hidden pool game would be, 5*x*4 + 5*x = 25*x. Then 45*x/25*x *100 = 180%, so it would take 80% longer. If the system searched for all players simultaneously, the queue time would greatly vary but hidden pool vs non-hidden-pool will almost always be faster because 4 hidden pool players are needed instead of 9. Also as Triplesteal highlighted, the system prioritizes games with mmrs closer to the player. This would also affect queue times. Say they tried to match 9 and 10ks together for a 9k average, if the match failed, they would have to drop some players and find replacements to form a lower average. The less players in their matchmaking pool, the more the drop and replacements will occur.
"hey guys i was right im stuck on hunden pool see u guys are wrong haha!!!!"
I agree with triple. The possibility has definitely increased. But no where does increased queue time confirm anything of what hidden pool retards are saying. Of course, I'm not actually an unreasonable retard and stand to be corrected.
It's just fucking skilless retards using it as an excuse for their own behaviour triggers me
My que times went from 15 minutes to 5 minutes.
my q time is always around 2 min now, which is the shortest i ever had, q-ing on euw/russia or euw only
did language prefference affect Que time?
some of the devs back in the days posted that language preferences are secondary and do not affect much the q time, and are often disregarded
No I think
ywn, I did not know top 1% in Dota is bad. I got it in 600 games too, which is less than most people. Keep crying so I can start a salt farm please.
Jacked, I never said the LC was toxic. You assumed that because you are a biased moron. Hidden pool is approximately 30% toxic players, 1% game ruiners, 69% bad players. The main problem is getting bad players. You assumed toxicity was the main problem because you were biased once again.
8 months ago, I literally got low priority every 5 to 10 ranked games (confirmed from my match history) for months, which was why I had to make a lp removal bot. I posted about hidden pool, and there you were to spout nonsense about how my claims were false. I definitely had hidden pool, but you denied it because you are a biased moron. This was not about whether both team had hidden pool players but whether it actually existed.
If players in hidden pool could only get each other, toxic or game-ruining professional players with high mmr, like Cncc, would basically never find games. To get 10 hidden pool players at 9k mmr is unrealistic.
Triplesteal is not so bad after all. I will stop calling him Trippleretard.
It only says that there's an extra search parameter based on behaviour score, not how heavily this parameter is weighted. The system will probably try match people with similar behaviour scores but we don't know what the range is and how much priority is given to it. It might be that there is a cutoff and people with a behaviour score difference that is too big will never be matched together, or it could be that it will group them together if there are no other people to match.
The post doesn't confirm hidden pool but it makes its existence more likely. For a low BS person to be in a hidden pool it would have to be impossible for them to get games with high BS people but that's not confirmed. The only thing that's confirmed is that its probably less likely for them to get matched with high BS people but it might still be possible if there are not enough other low BS people searching.
Well now you know ; )
shut up clopper
i dont know why you keep arguing with someone that is called ' trippleretard' ' ywn' or ' jackishited' doesnt make sense rly . lul. i mean i teached them same leeson 1000 times over and over again, with proves from GABEN itself lul, they dont understand. deal with it. why it bothers you? jsut try do understand them , dont hate them, they victims not bad persons. ' obssesion in mmr' sure has it own side effects. some person can handle that, some not. those 3 not at all. im not even trolling this time.
fact of the matter is if i played any of your low behaviour shit accounts i would still win most of my games and you dont cause youre bad and instead of tryign to get good you flame eveyrone else
you see? you understand the meaning behind that silence after your comment ywn? NO ONE CARES ! lul.
?
lol shit mafioso didnt know u taught me lessons, guess i wasn't paying attention.
@OP your argument just doesnt make sense. im too lazy to go into detail why it's completely wrong.
More like you got nothing. Dreng gets 10 minute queues at 2.7k mmr with hidden pool, which is 100 points above the average mmr in Europe. Now, Cncc has 9k mmr and is in the top ~0.0000026% of the mmr distribution. The conduct summary says 23% of players have more than 2 reports. Assuming hidden pool is based on reports, his matchmaking pool would further be split by roughly 5 times. This would place him with 0.0000005% of players. Basically, it would take him very long to find a game. Allowing the game to have a non-hidden-pool team vs a hidden pool team would "halve" his queue time.
References:
https://www.opendota.com/distributions
https://www.reddit.com/r/DotA2/comments/6n1u5o/the_true_mmr_distribution/
https://www.reddit.com/r/DotA2/comments/6ndcr5/infographic_mmr_distribution_2017/
Now that I consider it, if it was only hidden pool together, hidden pool players would still be playing harder games. The hidden pool population is ~1/5 of the normal population. This means there would be inflation in the normal population when compared to the hidden pool.
what if the hidde npool isnt based on reports, but account flags? checkmate atheists
the calculations make no sense due to much higher mmr disparity of higher level games
as a 2.7k u get games 2.6 to 2.8 average, as a 9k u queue into high 5ks sometimes. The MM prioritizes matches closer to your mmr over those far away from it, but it´s not clear how much weight it puts onto this parameter, or, in other words, at which point does the mm "give up" and put 9ks and 4ks together
Increased wait time BECAUSE IT MATCHES YOU ONLY WITH OTHER HIDDEN POOL PLAYERS WAKE UP SHEEPLE
where did u get this from lul
yeah where did yo uget this lul. how can you know that? like how can you know , ' if rain is created by clouds, it will fall down into earth? how can you know that? waht kind of genius are you ? jsut cause you not named tripple retard doesnt mean you genius brah lol.
LMFAO
That has got to be the strangest and most disjointed logic I've ever seen.
Wtf is this thread
funny how are we constantly swinging between
"hidden pool doesnt exist, you are bad" and "existence of hidden pool was confirmed long ago this is nothing new"
no now its just: "hidden pool exists, but you are still bad"
Lmfao you guys got jebaited so hard, was fun trolling u guys for a couple days on this thread.
Can't believe none said that this is a troll.
XD
Cuz no one knows who TF OP is
He popped up recently
Seriously tho it based on an account flag, meaning when you meet the requirement your account be flagged. The question is how long does the account flag last.
I have 3 accs with corresponding que times, one with 20-30 minute que times at 2.7, this one with 15 minute ques at 2.9, and then another account with 4 minute ques at 2.6
Also on the 30 minute que time For confirmed that all my team mates had 15 minute or more ques and the all the enemies had 5 minute ques. then on my 4 minute que alt account, I was matched against several of the players who were on my team on the long que time account, and stomped them to Oblivion but never got them on my team on the short que account.
why are u lying?
I got the numbers from the references and the conduct summary from Valve. Actually, Cncc’s exact percentile is unknown because there is a rounding error of 0.5*10^5. Though, the exact percentile did not really matter. The point was to show Cncc had a significantly smaller matchmaking pool. If rampageacademy (drengr(?)) has 10 minute queues at the average mmr, Cncc would have absurd queues. This leads me to believe Valve will allow a hidden pool player to match with other normal players if the queue takes too long similar to how Triplesteal said the system would ‘"give up" and put 9ks and 4ks together’.
No-one outside of Valve knows the matchmaking system but the calculation was like this. Since hidden pool players are x5 rarer, it would take x5 longer to find one. When someone queues, that person must find 9 other players. If it was hidden pool vs non-hidden-pool, 4 hidden pool players would be needed and 5 normal. If it was hidden pool vs hidden pool, 9 hidden pool players would be needed. If x is the time needed to find a normal player, 5*x is the time for a hidden pool player. There are 2 cases: the system simultaneously searches for all players or the system searches for each player individually. If the system searched for each player individually, the time needed to create a hidden pool vs hidden pool game would be, 5*x*9 = 45*x. The time for a non-hidden-pool vs hidden pool game would be, 5*x*4 + 5*x = 25*x. Then 45*x/25*x *100 = 180%, so it would take 80% longer. If the system searched for all players simultaneously, the queue time would greatly vary but hidden pool vs non-hidden-pool will almost always be faster because 4 hidden pool players are needed instead of 9. Also as Triplesteal highlighted, the system prioritizes games with mmrs closer to the player. This would also affect queue times. Say they tried to match 9 and 10ks together for a 9k average, if the match failed, they would have to drop some players and find replacements to form a lower average. The less players in their matchmaking pool, the more the drop and replacements will occur.
Quality thread.
10/10, new shitpost thread? lmao
SHUT UP ALL OF U UR ALL SHIT TY