now, a possible error with this method might occour where if people simply just ragequit then games will be really short(20-30 minutes only).
that means the enemy team will either superstomp them if the game goes bad at the start or if the enemy doesn't do that and the flamer team has advantage the game should go super long.
that being said, the accounts given have to have at least a 500-1000 difference from the previous account
meaning i wont take an account with 6300 BS after an account with 5900BS or after an account with 6400 BS
Cuki
Hmm the most common bracket is 3k i have looked into the stats.
I cant check it now but i will later
But honestly i never looked into my teammate
I can borrow you this acc and This https://www.dotabuff.com/players/389864725
However the second acc is around 2.5k mmr according to opendota and it havent unlocked ranked yet
beh score dont affect matchmaking, its how many times you got reported in your 25 games.
At one point i had 1k beh points but 0 reports but got decent teammates at 4k which is a miracle
Cookie, how would you playing on a 2-3 k account prove the influence of behaviour score on matchmaking? Obviously, your skill would be much higher compared to that of your teammates as well as the opposite team and you will end up stomping the match. Wouldn't it be more relevant, if a person at the same skill level (with higher BS >9k on his main) play 4 games on an account with lower BS <5k, since he would lack the skill to solo carry the game?
Also, if you decide to play on 2-3k account with lower BS, could you please play classic support heroes like CM/Dazzle/Bane, who lack pushing skills and are squishy? Just curious to see how it would turn out to be.
What was the initial hypothesis? Some are saying you get in a hidden pool and play with toxic players. Some are saying the toxic players only get in your team. Which is it?
The former is still balanced match making - meaning while games may be unenjoyable you still have equal chance to prove if you are worth your skill.
A lot of people are whining About their teammates but if it's still balanced match making in a hidden pool then this whole exercise is meaningless as you are not addressing the actual issue which is selective memory/confirmation bias/ dunning Kruger
^this is good advice
The skill difference is huge. Plus those blaming beh score don't claim to be 6k buy rather 1kor a bit more above their actual potential. So people that much difference can prove that beh score doesn't impact.
For eg 4ks playing on 2.8k sth account one with 9k beh score and on one with2k-3k beh scorw. Will give you actually figure.
>it's the episode called "i'm doing a research please give me your 2k accounts so i can brag about my winrate against people who play dota for a month" again
Then you realize even if you are in the hidden pool the potential amount of toxic intentional feeding fucks on the enemy team is 5 while your team have 4 potential
Unless you're one of them
I can play some games on the 2k accounts. Though I do not have any 2k accs to borrow myself :/
Btw when are we gonna play those weekend games that dude made group for. I was hoping to show my 3k supp skills
so basically if we get multiple accounts(10-20 of them) with 500~1000 difference we'd see the effects of behavior score shrink as we go up in behavior score.
if there's no effect of behavior score then an account with 500 score would get the same amount of flamers/feeders at 9500.
so basically all i'd do is track the amount of flame in a game and how much they work as a team, example their picks or if they group up or not when someone suggests it, the way that they suggest it(agressive or friendly) etc.
my skill would end the game after 30 minutes, meaning if i ignore my team and just not talk to them they would start showing their natural traits.
so for 30 minutes i'd have basically all information i'd need.
here's the things that make me think that it's not a balanced option where yu get an eaqual amount of flamers and non flamers
that would basically PUNISH people for being nice to their teammates by giving them un-controlable morons every single game and would create an overall negative experiance.
valve wouldn't be stupid to create such a system IMHO.
also the skill of the match wouldn't be an issue for the fact that someone could be 2.5k because he's nice to his teammates
or someone could be 2.5k but with way more mechanical skill but an asshole to his teammates
meaning if you match team A where all people are nice against team B where all people are assholes, the overall skill stays the same.
that's why i said, skill shouldn't be an issue.
only way that this wouldn't show any reults is if the system would actually balance it, which like i've said basically punishes you for being a nice decent human being.
but if that does happen, we'd see a backwards trend where in 1k you'd get nice/neutral teammates, in 5k you'd get only nice and really bad, and in 10k you'd get only bad and neutral teammates
OR if the system just selects players disregarding their behavior score.
i have a 2.6k account that i wont use for a long time because i dont have a sim
afaik its still on lpq tho
actually its this account that i use to post with l m a o
but eh if u dont have an extra card then too bad i guess
Like i said i can borrow you this acc for a while. If you borrow this acc i will play on my smurf but i wont be playing much either
Why not make a google form or questionnaire asking the dota communities who are willing to do it, about their bs score and their recent game, whether or not they have quitters, flamers, rage, and so on.
because this research is trying to be least bias, i'll record all games and i'll say literally nothing for all games.
and upload on youtube
basically saying, if you ask someone about his day
he won't tell you objectiveley how he woke up, ate cerial or went to work
if they felt nice they'd say ''i woke up quite early, had my favorite cerial then got ready and went to work''
but if they felt bad at that instant they'd say '' i woke up like a zombie from a grave, the cereal tasted like shit and worst of all i can't be late for my shitty job''
but basically both scenarios were exactly the same scenario.
same if you ask someone who's a flamer to tell you about his teammates
he'll select only the bad things and blame the system
but if you were asking a nice guy, he'd say quite the opposite.
that's why i said, anyone who's 4k+ is invited to come and do the same testing on the same accounts
because the conclusions i'll make from the data can be biased as much as i'm trying to make them not
I have been across the board from 3kBS - 9kBS.
I can confirm that Behavior Score DOES have an impact on matchmaking to some extent.
I have been researching behavior score since I first learned and heard about it in either December 2016 or Jan. of this year.
The only thing I need to know is what do brand spanking new accounts start at? What is their behavior score? That is the only thing I do not know, and I need to know it to confirm or do more research at 9kBS.
Anyway,
3kBS is as low as I have ever gotten so 2kBS and lower, I cannot speak for.
3kBS and even 4kBS actually has GOOD players. The only probably is the level of toxic, flame, and tilt. If they do not get their respective lane, or they bitch to the support, or some little minor thing irritates them, they make sure the team hears or type constantly and ocassionally feed.
The most "I'm done, feeding" and the player actually buying sentries to bock your own entire jungle and feed in as many ways as possible you will ever see is in 3kBS & 4kBS. But they rarily mess with the courier at least. I never see courier feed much which is nice. They greif, feed, and ability abuse. But the odds of winning those games is very, very low.
5kBS & 6kBS is okay. It is very hit or miss, but I notice the players are either VERY passive or VERY aggressive. I like the very aggressive players more because they actually get into the fights and aren't vaginas, and go in the fights when someone initiates properly.
The most common thing throughout Dota is actually the most common (I notice this in my research at least) thing in 5kBS & 6kBS and that is flame-wars via typing. The win odds % of that game drops because two idiots will go back and forth and instead of playing properly or at their best efforts, they sit somewhere, type, get ganked, die, and then tilt even more than they already were. The other player insults them, then they get ganked, die, tilt even more than before and then you get a ping-pong of flame in your team and are virtually left at a 3 v 5 game. Even if other players suggest (I always suggest this when I see people type too much and play poorly as a result) that they mute each other, they keep the flame going.
Voice chat flame happens a lot too, but the win odds % stays about the same since they still land their skills and do not play entirely retarded.
7kBS is an awkward spot. It has nice players, and assholes. It is like... Behavior Score Limbo. How you act there determines where you will be in the next 10 - 50 games lol
8kBS is nice. Not much toxicity there , but no one is perfect and we all have bad days and get that gamble-addiction scratch when it comes to our mmr.
9kBS... Is weird.
I need someone to confirm for me what the behavior score of a BRAND NEW ACCOUNT is. Meaning 0 or 1 games, due to smurfers, I have a feeling people start at 9kBS, but I need this confirmed. I never smurf, so. I can't really talk about this area yet.
0kBS - 2kBS I have no clue. Never been that low and wonder if that even exists.
Got any questions, let me know.
yes, the question is do you have data or any 2k accounts.
because whatever you just said can be biased to your own experience.
yes, the question is do you have data or any 2k accounts.
because whatever you just said can be biased to your own experience.
It's not bias, I always do my best to see my bias and put it aside for best result.
I was in each thousandth spot for at least 3 weeks and paid close attention despite winning or losing or even on my tilts and bad days or good days.
No, I do not smurf at all so that should answer the "do you have other accounts" question.
Most of the games were 3k when grinding to climb and reclimb back to 4k.
3k has a large pool of players as well. Not as large as 2k, but fairly large.
Not sure if you want me to join a party game with a 2k account you borrow (I am 4k+), but you are free to add me so you can test it out with me.
My current BS is 6.9k, my next report should put me back in 7kBS.
I can queue almost any serve comfortably.
edit: Oh, the only data I have is mental notes. I never thought of storing data on this, and now I feel stupid since I always have an empty notepad that I am trying to find a use for.
I could have used my notepad to wrote down dates, match id's, how the match went and the score with it.
And even some of the games, I was a little toxic and bitchy. My score dropped as low as 3kBS because of real life things that resulted in having to abandon. (1 emergency, another was 5+minutes to find the source of why my web was acting weird)
https://www.dotabuff.com/matches/3169709535
Hello cuki. Take a look at my this game. Everyone in this game is solo with no party. I am actually playing with behaviour score of less than 2k
South east asia server, some called it the most toxic server...
If thats not enough, it is the legendary 2k mmr trench tier coupled with players with an average behavioural score of less than 1k. Holy jeez..
I dont believe in BS, but, if it was real, how could you see the difference in each accounts? I mean, you will stomp them anyways, because you are much higher MMR player, than them. If you are better enough to carry your team, not even the most toxic player would say anything or do anything against you or teammates, becasue they are winning because of you.
I'll check my behaviour score and post it this evening.
Im not playing dota a lot now because i have no time.
I sincerely hope my items stay on my account though
Fun fact = most people I know who whine about SEA server being toxic are bad and delusional as fuck
Don't understand that part about nice player being punished. That's irrelevant. I thought The logical assumption is that a nice player in a team means that team has a higher chance of winning because of slightly more cooperation.
Instead what the uneven matchmaking model predicts is that the toxic teammates will constantly feed themselves into oblivion i.e. Only 9 k average games are full of nice ppl and 1k is full of ppl feeding
In my main account I have 8600 BS, and is really common to be with really toxic teammates, feeders, afk, courier feeders, bad itemization on purpose, and talking shit during the whole match. BS means nothing.
Hmm i cant say that sea is toxic because i treat them as a bot. I could die from crippling depression if i treat them as a normal human. I have played in us east once and its not that different from sea but sea had more flames than us east
I am the one of BS believer
And i play in my friend account which have 700 mmr and 5k BS because he abandon too much
I go cm jungle carry and win the game
No flame because i carry them
And maybe that what will happen to you if you play in 2k
Ppl just suckin ur dick after u rekt enemy mid
Or maybe ppl in sub 1k actually not flame at all because they dont even understand what to flame 
The thing with 1k is there are lot of people who just don't care about winning and losing. All they want is to enjoy a good game. So that bracket cannot be taken into consideration. It is a mysterious bracket.
They will report you for communication abuse/ feeding on purpose but not for bad gameplay as they themselves don't play that great and they know it.
The real problem starts with 2k+ who think they are in wrong bracket. They won't see their mistake at all and only worry about teammates. So even if they give lot of time and effort they don't improve because they don't actually know what their weakness is.
The only problem I see with this is that some people might only flame when they think they are losing the game, therefore the fact that you 4k people are going to win almost all of your games could impact the results, which is why it may be better for people that are actually in that bracket to play the games.
^ this. I play occasionally with 1-2 friends of mine with 4k BS. Everything goes smooth, till you lose a team fight or die 1-2 times. Then the flame-blame games start. When I solo queue, people seem to be more friendly.
Having said that, I have not played enough games recently to substantiate that observation!
Моля, впишете се, за да публикувате коментар.
okay, so this topic has been annoying me for a while now
basically every noob player makes excuses why his games are bad or why his winrate is lower because he keeps ''getting bad teammates''
teammates who flame or feed intentionally
now obviously unintentional feeding can be played with and obviously boosters land 80-95% winrate in low mmr bracket regardless of the noobies on his team.
that being said, everyone keeps bitching about how their teammates are always the mean ones making the game less enjoyable because they're stuck in some hidden pool/low behavior score or someshit.
i'm here to put an end to this, let's finally just do the research and be done with this ridiculous topic so it doesn't have to keep coming up.
So what do i need you to do?
i need accounts, we'll do the research simple: we'll test on accounts in 2000 to 2999 mmr with different behavior scores
the reason i took 2k mmr is because it's the most common bracket and stuff like this will be the most potently visible, also because skill won't be an issue and we can isolate that.
i'll play 4 games on each account, anyone else is also free to come and do the testing with me so i don't have to play a shit ton of games, but he has to have at least 4k+ mmr(like i've said, so skill isn't an issue).
for consistency play only 1-3 heroes.
in none of the games the tester will say anything to any teammates, only observe the amount of intentional feeders/flamers and record it ON BOTH TEAMS.
so if you have any old 2k accounts, feel free to notify me
for the moderators: i'm not doing any boosting here nor advertising any account boosting, buying or selling services. i'm just trying to do research because this topic just annoys me. i don't care if i win or lose or any of these accounts.
my predictions:
if BS does affect your games:
accounts with lower BS will have more feeders/ruiners, ending up with out of control matches which will result in them going longer than matches with high BS. and this will be easily detected.
reason for this is because people will be raging/flaming/feeding at each other, basically they're wasting time on not playing properly.
as a lack of teamwork, games will be harder to close and will result in longer duration matches as low mmr players tend to be bad at pushing.
if BS doesn't affect your games:
accounts with lower BS will have on average the same amount of feeders/ruiners, ending up with on average the same results overall.
to check your BS on your 2k accs just do this:
Type "developer 1" into console, then "dota_game_account_debug"
and just copy/paste it.