DotA 2 MMR
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.
You are currently reading a thread in /v/ - Video Games
You are currently reading a thread in /v/ - Video Games
>sup /v/
>dota 2 fag here
>fucked up mmr on TBD games
>got stuck on low mmr
>tips for ranking up since i play better that the fags mmr gives me
>can't rank up alone, since it's a teamgame
>win together, lose together
>can play mostly anything
>don't care about party mmr, solo is moar important
>all tips, tricks, hints are welcome
Is this your first post on 4chan you faglord?
Lurk more and stop shitting up the place.
>>332517410
Don't make fun of the new guy, just tell him
GIT gud
Check out this broken as hell Mirana 1v5.
https://youtu.be/qmzetnvBCWU
>i play better than the fags mmr gives me
you just believe that anon
you're as bad as them
>>332517547
noes, cuz on party I play with @1.5k higher mmr friends, and am able to keep up, actually seems easier than with dense low mmr fags
5k here if my name doesnt make it clear.
I lost all 10 tbd games on my smurf, still got to 4.2k. So stfu about fuckimg up calibration matches.
My tips on climbing:
Get better at farming
Get better at laning
Get better at teamfight micro
Learning a specific hero comes after all these things
>>332517509
Obviously fake or 40 mmr
>>332517410
This. You want to actually fucking talk about games?
>>>/Reddit/ is that way
>>332517775
>.97 on farming, 0.82 pushing
>problems occur on some teamfights
I'm not a newb, 3200+ hours, 2700 matches, 1500+ wins, just started MMR recently, but stuck on low.
Get it nao?
>>332518069
the pentagon means literally nothing
you can have that many hours and still be dogshit
what's your mmr?
Forget Ranked, go play Ability Draft only. I know I can never be a Dota master, so I'll just play the party game mode.
>>332517728
>cuz on party I play with @1.5k higher mmr friends, and am able to keep up
literally the eternal litany of the shittler
your high MMR friends are doing all the heavy lifting for you and you don't even realize it because you have literally no game sense
>>332517410
This
>>332517346
Fuck off
>>332518140
calibration gave me 1.5k solo, 1.4k party
worked them up to 2.1k solo, 1.9 party
Losing streak, 8 loses, currently 1.9 on both
played different roles, played my best and fav heroes, still stuck @2k, can't seem to push out
>>332518069
.97 on farming and .82 on pushing
What the fuck does this mean? Is this ur graph on yhe profile? That doesnt count for shit.
I average 550 gpm and xpm on my profile, whats urs?
Post dotabuff, any DECWENT (3-4k mmr) player will have 5cs/min AT LEAST whenever olaying mid/carry, which i fucking bet you cant get
Dont you fucking dare be proud of ur abikity if ur mmr is as low as it is.
I am 5k and NEVER SATISFIED with my ability, i fucking strive to get better, thinking you are already superior to your teammatws will only make u worse at improving urself, because u turn a blind eye to urself
>>332518329
dude you're fucking dogshit
anything below 4k is absolute trash-tier cancer
honestly 4k is pretty awful too
just do what >>332517775 says
if you're playing carry or any kind of non roaming meme hero then aim for NOT missing any fucking cs (100 every 10 mins is pretty good after the first 10 minutes). done you are instantly a 3k mmr player
>>332518235
we motly play CM, and I do the drafting
i'm the only one of them to actually read the entire changelogs, patchnotes and game changes
i know their roles and preffered stuff, so i get a best overall for everyone
everyone does their job, nobody cares about mmr then, just about wining
>>332518463
oh yeah that's how you guys got to the high realm of 1.9K party MMR
>>332518630
nah, they are above 2.7k mmr
>>332518329
i was high masters in sc2 so i think playing dota 2 came pretty naturally to me, didn't take long to get to 5k+ -
if you don't have those kinds of micro, macro (in terms of map awareness, game sense etc) skills built in its going to take hard legit practice to get better.
watch your replays, get a coach god forbid or just ask a 4k dude to watch your replay and give you some pointers
>>332517346
post match ID, we'll make fun of you for free
>>332517346
Just play at the rank you were given, you'll get better over time and rise naturally. So long as you have a 51% winrate.
Conclusions:
>admit you suck
>git better
Improve these:
farm
awareness
fights
picks
micro-skills
Pray for no cyka blyat or allahu akbar in team.
Got it /v/
10x
Tip for increasing MMR:
Hope you don't get paired with whiny entitled retards that all pick carry or another mid hero when the position and pick is already filled.
>>332518329
>2k mmr
LMAO WHAT
Are there people really that bad at the game taht 2k is a thing people can't get out of? I'd say 4k to 5k is where shit gets super hard, because you have the mix of really good players and really awfull ones. 2k is straight up shitters town.
>>332519353
welp, 2k is a mixture of 3k and <1k
so... whalecum to my world
>>332519491
post
ID
>>332519598
private profile, i'm at work, can't login to change privacy settings
>>332518329
I gave dota 2 to my wifes 10 year old son, 5 hours in he is already 3k mmr.
Jesus dude.
>>332519748
>dem feels
>>332519491
>>332518329
Assuming you're not trolling, a good way to work you're way up to 3k is by spamming mid heroes that are difficult to counter for noobs but are still viable late (because your retard teammates will not know how or when to push or take objectives). Think Slark, Weaver or Brew. Anything invis is good too. Until 3k you should be absoluetly dominating the game: if not, you are probably just not that good.
>>332517346
Bait
Me and my friend have been playing this game since 2011. We're both mid 4000s and have been since the ranking system was introduced YEARS ago. We're literally ~200 mmr within what we were placed at, after 1000s of hours of ranked play. I have shitty smurfs that are in the mid 3000s and I have a hard time ranking those up as well. I think the system is just ass.
>1.4k mmr
>anything below 2k is just you're either on the shit team or the winning team
How do people seriously get out of the trench?
>>332519146
also one thing I notice most low mmr players doing is fucking NOTHING
ie. make sure you are always doing something - be it farming, roaming to gank, teamfighting, sitting behind your carry if you're supporting etc. - if you find yourself running around in circles gaining no xp or gold DO SOMETHING (this includes buying a smoke and making your dogshit team do something with you)
>>332518452
Dude...
3k is like the top 10% of players. The vast majority of players are near 2k or a little below.
>>332519871
>it's not me that stagnating, it's the system that's wrong!
Also, consider that if you ranked at the beginning, everyone else on the ladder had time to improve. It's not that you didn't learn anything, it's that you didn't learn faster than everyone else.
>>332520003
what is your point? they are still bad at the game
>>332519951
hit a lot of creeps and then hit buildings
>>332517346
dont know whats your mmr
I am at 5019 and I assume because you said you are low mmr that you are 2k or 3k but from boosting my friends account you can usually pick any hero that can snowball mid easy.
The mistakes people do at 2-3k are just too big leaving you with much freedom and space to abuse it.
>>332520003
>this is what shitters tell themselves to sleep at night
AHAHAHAHAHAAHAHAHAHAHAHAH
average MMR is 3.2K
https://yasp.co/distributions
>>332519852
Not trolling.
Thanks for the tip!
>>332519871
you're probably not actually improving, just doing the same shit over and over
I was the same as you, but last year actively started trying to improve (watching my own replays from enemy perspective all that shit) and im mid 5k now so get good shitter
>>332519852
Brew? Really?
Anyway OP, git gud at Invoker/OD/Spectre/Death Prophet
They're all overpowered in the current meta and will give you a positive winrate if you're remotely decent at them in your given matchmaking bracket
>>332517509
>buys 2 tps instead of boots of travel.
>>332520190
not disagreeing because it's probably 100% legit, but isn't this a biased sample because the only people who sign up to yasp are probably somewhat competent at the game
>>332520493
>Includes players displaying MMR on profile and sharing public match data (Sign-in not required)
>>332520418
noted
Slark has always been a hard to kill cunt, gonna play it moar.
morons itt
if you are 'stuck' on mmr then that is how good or bad you are. end of story.
get better at dota and your mmr will grow over time.
in particular get better at these things:
>farming
look at any low mmr game and you see people miss easy last hits left and right. they are literally throwing away free money.
>map awareness
the minimap exists for a reason. always keep an eye on it. alway visualize where the enemy heroes are and think about what they might do next.
>fighting
use your fucking spells and items. make sure you have enough mana and hp do do that everytime a fight happens. too many morons die without using their shit.
>positioning
stay away from what can kill you. get close to what you can kill. this only works as well as your map awareness is - if you dont know where the fuck everybody is then you cant position yourself.
>objectives
hit the fucking tower and barracks instead of wasting your time chasing the enemy into their t4 towers and fountain.
when in doubt, HIT BUILDINGS
>picks
pick good heroes. dont pick stupid shit.
dont pick shit that gets countered by the enemy heroes. pick heroes that counter the enemy.
>>332520995
>Do all these
>Still lose because >>332519251
This community is pure cancer and toxicity. How can you win a team based game solo when you have retards actively working against each other on your team?
>do great in bot or top
>get some kills, stack and pull creeps so waves of xp denied to the enemy
>other lanes come and fuck your shit good
>lose match
>rinse and repeat 8 times in a row
>>332520638
Be somewhat careful with Spectre though. Although very good, she does rely on some babysitting in the very early (which your teammates won't understand).
>>332520418
Brew might not be very good currently in high mmr games, he still stomps noobs who don't know how to counter his ult, especially when well itemized.
>>332521186
the same thing is happening in the other team
take advantage of it instead of whining
If you're gonna have to contest cs with someone else on your lane then get a quelling
when playin support, carry sucks and vice-versa
hao can i improve like that?
also, aside from improving those, best idea is picking metas and personal best heroes
>>332519951
Play a lot of games and do the best you can.
>>332521294
Most of the time it isn't. Whenever I'm paired with retards not willing to pick a cohesive draft and work together, the enemy team always has a solid comp and actively group ganks and pushes.
>>332521186
if you are better than the average player at your mmr then your matches are
4 retards + you vs. 5 retards
in the long run you will win more than half of your games and thus rise in mmr.
hell, read this
>http://www.playdota.com/forums/showthread.php?t=1398477
all the "muh bad teammates" whining has been disproven over and over again for years now
>>332518329
With 20 hours I got 2k MMR how bad are you?I now have about 2.4k hours
>>332521539
>me ooga
>team booga
>other team ooga
>>332521434
protip: you suck just as they suck, you just don't notice it because it's you.
post your dotabuff if you dont believe us. good players will gladly point out all the games you sucked ass and lost on your own
>>332521601
Oh boy, anecdotal evidence! My favourite!
>>332521802
>anecdotal
>144 games
Were your parents siblings?
6k MMR here.
Don't bother climbing too high. People up here are just as fucking retarded. Don't believe me? Turn on some high MMR streamer stream, their teammate can be just as fucking stupid. Also to grind out MMR focus on a role and certain heroes. That is the best and fastest way to climb. The minimum for what I consider competent at Dota2 is 5k.
tbd games don't have that big of an impact on your mmr
if you got placed low it means you were low hidden mmr in unranked during your 3000 games and just didn't realize
>>332521601
Not him, but I managed to consistently land gold ranks in LoL in my first 10 matches.
Last season, however, I got so many DCs and rage feeders/quitters I landed upper silver and got stuck there until I duo queued with a m8.
TL/DR: elo hell does exist.
>tfw 5k
Will I ever git gud and reach 6k?
>>332521802
nigger what else do you want
>watch low mmr games
>everyone plays like total trash
>watch medium mmr games
>people play bad
>watch high mmr games
>people play good
>put proven good player into 3kmmr
>he goes on ultra winstreak and immediately rises back up to 5k
yet you insist your whining has any merit to it?
>everyone is bad but me!
>i just lose because my teammates suck all the time!
do you even listen to the shit you're saying? are you seriously unaware of how deluded you are?
>>332517346
>>332517728
You type like a 12-year-old retard, fuck off.
>>332521892
>144 games from a single person is indicative of the hundreds of thousands of games every hour
>>332521892
>Anecdotal: "Based on personal observation and not rigorous or scientific study".
Yes Jim, it's anecdotal.
>>332517346
>fucked up TBD games
Buddy unless you lost every single one in very unfortunate ways, you're where you're supposed to be. Party MMR is a joke and nobody takes it seriously.
>>332522234
post your own dotabuff then
show us your profile
>>332522234
the total size of the population being tested has no effect on the necessary size of a representative sample
learn to statistics idiot
>>332518329
>calibration gave me 1.5k solo, 1.4k party
>worked them up to 2.1k solo, 1.9 party
Ok thread is over, this is just embarassing.
>>332522262
How are 144 recorded games not a rigorous test?
hey OP, some real talk for you and im going to teach you math.
ive been rank 1 useast a few times in wc3 random team ladder in the titular mode of east, 4v4. this is the kind of complaint i heard all the time there and here as well as what i said before i got better. im also over 4k in dota
so let me tell you some tips to help you in dota
first there are 5 players on both teams, if you are as good a player as you say then your team will always be at an advantage, why? because you are guaranteed not to be a shitter which means there are 4 chances for your team to have shitters and 5 chances for their team, this means statistically over time you will have an advantage in all your games
the most important thing is always improving your personal skill and knowledge through practice and reading guides, watching tournaments etc. on the road to becoming a rank 1 rt player i spent a lot of time playing solo and improving my individual skills of micro and macro beyond and kept up their practice, far beyond what is considered necessary for an rt player. what you will find is that these individual skills actually make a significant difference over players who just show up and dump their skills
yes there are games where your team has utterly thrown and purposefully fed and abandoned but if that isnt the case then the vast majority of time that game was winnable if you played better. ive won 4v5, ive won with feeders on my team etc.
the two most important things i can tell you thought is this:
1. you arent better than your rank until you prove you are and get a better one
2. if you wanna get good you gotta play a lot and get over the mental block of fear of losing/losing rank
>>332520617
Most people hide their data. Average is around mid 2k.
>>332522354
>Two men we studied were bigots, racists, and rapists
>Therefore all males are bigots, racists, and rapists
Yup. Sound logic you got there.
>>332522449
Scientific means controlling for variables. He lost a lot but it's unlikely he was using the same character the whole time, playing with the same people with the same connection, that some people didn't drop out or he loses a team on some games and not others etc.
Yeah. Nothing is controlled. This is just 144 games that this one person experienced in a certain was as compared to the hundreds of thousands of games happening constantly giving more meta-data. It practically slots in to the definition of an anecdote.
>>332522638
>this data is biased, here is some stat I pulled out of my ass, it's much better
>>332522638
>Mid 2k
Is that what you tell yourself everytime you lose a game?
>people complain about bad teammates
>people insist they are much better than the retards they get matched with
yet none of you have the balls to post your own yasp or dotabuff profile.
you know why? because then everybody would see that you are full of shit, that you are throwing games left and right, that you are just as bad as the retards you whine about, and that there is no merit to any of your complaints.
go on, keep whining about the system. maybe it takes another 2000 hours of stagnation until you finally realize how shit you are and how to improve. maybe you will get so mad you quit dota forever. or maybe you will just keep getting mad at strangers, keep being unaware of your own flaws, and keep stagnating forever.
enjoy
>>332522449
This is a decent formula for determining an adequate sample size (admittedly i did IB HIgher level maths in school, but a simple google search for "how do you determine if sample size is adequate" gives this
Necessary Sample Size = (Z-score)2 * StdDev*(1-StdDev) / (margin of error)2
If you want, say 95% confidence level (levels of 95, 99, and 97.5 are usually used), 0.5 std deviation (which is acceptable, for most normal distributions anyway)
You would need 380 ish sample size.
Anyway, this is all retarded anyway,simply because a person who has played only 100, 200 or even 500 games of dota isn't representative of someone who has played the game enought o actualy understand its mechanics, hell, 100 games of dota you haven't even played every hero once, let alone bae a representative sample.
I'd say 500+ games is representative when talking about a hero, for example I can confidently say I'm a good TA player because i've got 1k games of ta at 60% winrate, a person who has 90% winrate at 100 games doesn't matter for shit desu.
I'd say u need 2k games of dota for your accounts dotabuff to have any real meaningful data.
>>332522694
Except it's not 2 here, it's 144.
>>332522726
>He lost a lot but it's unlikely he was using the same character the whole time, playing with the same people with the same connection, that some people didn't drop out or he loses a team on some games and not others etc.
How does that matter in a test? I don't think you understand a single thing about statistics.
Your hypothesis: winning in the trench is down to chance.
Confidence level: 95%
Sample size: 144
Resulting confidence interval: 8%
He would need to have a winrate below 58% for your hypothesis to hold.
He got 85%
Don't even try to pretend you understand statistics.
>>332522465
agreed
but please tell me that matchmaking has nothing to do with my real skill
i only go on steaks
5-8 wins, 5-8 losses
never moar, sometimes less, and i can't seem to be able to overcome losing streaks after say... 2-3 losses
i just lose all the way
>>332517346
>sup /v/
>dota 2
Fucking faggot.
Do newfags really not lurk anymore before posting or even work on their writing skills? Typing like goddamn 14 year olds.
>>332523174
>isn't representative of someone who has played the game enought o actualy understand its mechanics
that's not what's happening here, we have some OP pretending he's better than his level, and a sample of some guy who's actually better than his level playing
>You would need 380 ish sample size.
you somehow didn't include a confidence interval in your calculation, I don't even know what you're calculating here
>>332523178
I have a degree in Psychology and Forensics so I had to do some stats and Research Methods but I admittedly wasn't amazing at it. This wasn't what I was arguing anyway; I don't want to make a research question, I just wanted to say that one person losing some games is anecdotal and not scientific.
I got a 2:1, chill out.
>>332523227
Post MMR and Dotabuff. If not gtfo.
>>332523227
>but please tell me that matchmaking has nothing to do with my real skill
it does
your a shit
>>332523178
Oh, so 144 convicts picked up from a single location is fine to represent an entire gender. My bad. So much better logic.
>>332523227
much of it is down to your mindset and mood
when you lose a game, take a break. do something else, play another game. dont immediately requeue.
>>332523336
Confidence interval = Margin of error, at least iirc. But im pretty sure on this.
>>332523468
What do you mean a single location? Your hypothesis is that winning in the trench is down to chance. This guy experiment proves it wrong.
A single counter-example with enough of a statistical sample is enough to prove you wrong. Can you stop with the shitty analogies too? You're only confusing yourself.
>>332523227
Post your dotabuff.
>>332523538
>Confidence interval = Margin of error
yes but confidence level /= confidence interval
>>332517346
Don't listen to memers who say you have to play carries. I've come to realize that it's way easier just to play support to grind mmr. Just create situations where only a retard couldnt win. It's fucking hard to carry games solo, but if you just feed kills to your carries, rotate and tp often so people don't get free kills, and have vision so that any ganks you can see from a mile away, then it's way easier to win than dealing with the stress of having zero vision with people dying left and right because of shitty supporting.
>>332517346
Ignore the secret club fags.
If you truly deserve to be higher ranked, you will make it there. Its 4v5 of people at that mmr + you.
Pick what your team needs. Be positive to your team from the very start. Moral is important for winning, and if you open the game with "I'll support the fuck out of you at bot, you just last hit and win us the game" they are less likely to talk shit to teammates in a positive atmosphere.
>>332523728
I already stated confidence level of 95%
It is used to calculate the Z-score used int he calculation
>>332523654
It's completely relative, fuckwit. Single person, 144 matches. Exact same scenario. Keep going on with your anecdotal evidence and shit, invalid logic.
>>332523881
Post your MMR and Dotabuff.
>>332523881
nigger do you not understand the situation at all?
>shitters claim bad teammates keep them in low mmr even though they play good
>actual good player gets low mmr account
>shitstomps all the shitters and gets back to high mmr in no time
>HURR ANECDOTAL EVIDENCE
>ITS ONLY VALID FOR HIM PERSONALLY
thats the fucking point. he's good, so he wins a lot against these scrubs. if you were good too, you'd be winning a lot more than you are.
Just pick night stalker mid and carry games senpai :^ )
>>332523839
>>332523174
the margin of error here is 35%, not 5%, don't just copypaste what you found on a website
>>332524034
>Shit, he got me
>Better deflect and go for something to change the subject now that I'm outed as a mouth breathing retard
>>332523881
A single example is enough to rebuke a generalization like "winning games in the trench is down to chance"
lrn2logic bitch
>>332523881
To be fair, any mmr expermient in dota would be empirical, that doesn't necessarily reduce its statistical value.
But enouhg of this bullshit.
Think about it anons. I've played on my friends' 2k account (I'm 5k) and absolutely stomp almost every game, if not all of them.
I bet EE, RTZ or anyone would.
Hell, a 4k person would do it, stomp 2k games.
So if you aren't stomping, then you know you aren't that good of a player compared to the ones yo get matched up with
Reading all this dumb shit makes me glad i stopped playing dota 2 2 years ago.
Literally a bunch of autismal children .
>waah i keep losing because of all the other retards
all you shitters are simply lying. post your match history on yasp or db and you will be shown how much of a bumbling retard you are.
>>332521738
Don't carries/cores have a higher impact in low MMR games though?
>>332523178
>Don't even try to pretend you understand statistics.
You have absolutely no idea what you're talking about. Please stop posting.
>Sample size: 144
The sample size is actually 1. Kill yourself.
>>332523174
By the time you've played 2k games, the first 1k will be irrelevant because how much you should have improved.
And by the time you've played 1k games of a single hero, the first 800 are probably irrelevant because of any patch changes or just the meta.
>>332524402
that holds across all MMR brackets.
playing core is intuitive. playing support is not.
pretty much every player (me included) does a better job at playing core than at playing support.
>>332524482
>The sample size is actually 1
it's 144
we're testing your hypothesis that winning is down to chance
we have one counter example which has been tested 144 times
it's enough to prove you wrong, just like one stone going toward the roof when I drop it would disprove the theory of gravitation
How is this thread still up?
>>332524305
The thing with mmr is that at lower brackets (2-3k) you can always stomp left and right if you are at least 1k above. BUT, when you reach 4k everything is different, as said in the experiment posted above, everyone has this shitty attitude where they think they are the best player on the team, and when they might be good at the game they end up losing because they die once and decide to spend the rest 10min using allchat to call gg. I've gone from 3.7k to 5k,, now down to 4.5k and I can't take this bracket anymore, people tilt so easily because they know they aren't good, and they know they made a bad paly, but they decide to blame whoever is closer to them and proceed to shit talk over and over until game is over.
Even when I have a million comends because people tell me im chill and always add me after games, EVEN THEN I get people talking shit to me every 2 games or so.
Mid 4k bracket is trully the most horrible place in dota.
>>332524658
Not me.
In any game I prioritize teammates over myself.
>>332519951
You can honestly win games completely alone in the 2k bracket, if we assume there is no smurf in it.
2k players are literally to 100% people with an IQ of 80.
They dont know how to farm properly, when to push or how to set up a working gank.
If you are a strong mid player you should easily be able to win these games solo, because of how shit everyone is.
The harsh truth is that people grossly overestimate themselves and win about the mmr trench / Elo hell, when they themselves suck shit at the game.
>>332518732
thats still below avg.
Dota especially in extreme low MMR zones is literally farm heroes and win game, go mid, crush mid, take tower, crush people trying to stop you, rotate, crush other lanes with that advantage, repeat until you,win game. Its that simple
>>332524806
This. 4k much like 1800 for HoN was the worst bracket because everyone is ready to fucking throw the instant they misplay.
You can carry a retard, but you can't carry someone legitimately throwing a game on purpose because they can't fess to them fucking up.
>>332523227
If you want to get an higher MMR you should focus more on improving than anything. The MMR is fair and works, if your real skill level is higher than your MMR reflects, and you think everyone in your team is worse than you, it means the other teams skill level should in average be about the same as your teammates, meaning you are more likely to win.
As other people said in lower MMR it's easier to climb the ranks by picking heroes that are hard to counter rather than going by what is currently played by the pros in the meta, Ursa, Weaver, Clinkz or Bristleback come to my mind. Also if you really are objectively better at the game than your teamates, or at least understand the game's flow better, you should guide your team when necessary (calling when or where to push, gank, go Roshan, etc). The truth is in low MMR I don't think it matters too much specifically what heroes you pick, but it always help if you're not countered by the enemy picks, I usually play position 5 and I already lost the count of the number of times I have carried party games with my lower MMR friends using fucking Crystal Maiden, only because I rotate a lot and gank early game and know better to them how to position myself in team fights to minimize my chance of dying.
>>332524942
this desu
people have a natural bias to not recognize their own mistakes, thus they complain about their team even though they aren't really any better.
>>332524868
That's kinda the problem with low MMR support players. They mention all these self-sacrificial things they do like "I bought wards and courriers, I bought a mech, I died to save my carry, etc." but that's missing the point of what makes a good support.
A good support knows how to get ressources for himself, and be aggressive.
>>332524942
-Sincerely a 3k Dotalord.
>>332524670
First of all, a sample size of 144 is incredibly small anyway. Try playing 3000 games.
Secondly, your sample size is 1, because you've sampled 1 player.
Thirdly, a 95% confidence level is not appropriate in every circumstance.
Fourth, you have poorly defined your hypothesis.
>>332524658
I almost always play support, so I kind of feel like im better at that. It's really frustrating playing core and not having teammates to back you up, so I just started playing supports
>>332525049
>You can carry a retard, but you can't carry someone legitimately throwing a game on purpose because they can't fess to them fucking up.
I play at 4.5k and haven't seen someone like that in years, maybe it's your region. If someone were to throw a game before it's over they would just get reported, unless they're in a 3man stack, in which case you should just wait for the game to be over.
>>332525134
That parts nothing different from being a carry though.
I'd say the key to a good support is knowing how to value your farm, life, time, whatever, in comparison to what you could do for the team.
>>332524868
see >>332525134
buying wards and mek for your team isnt what's hard about support.
what's hard is finding farm for yourself without taking from your team.
what's hard is roaming and ganking without dying.
what's hard is creating opportunities throughout the game.
tl;dr knowing what to do and when to do it is really hard. plenty of supports end up wasting a lot of time and not getting shit done.
>>332517509
God that was fucking terrible.
>>332525340
Support doesn't actually mean support, it means setting up ganks and taking towers.
>>332525095
yeah it's pretty amazing the amount of people who mention extremely precise micro errors to their teammates after a fight in low bracket
like, focus on playing your own hero for God's sake.
>>332525250
>First of all, a sample size of 144 is incredibly small anyway
It's not, I posted the numbers. And another anon posted a formula which gives a sample size of 54 (not 380 like he posted). You're free to post how you came to this "3000" conclusion, but don't expect us to take you seriously if you're pulling that number out of your ass.
>Secondly, your sample size is 1, because you've sampled 1 player.
"it's enough to prove you wrong, just like one stone going toward the roof when I drop it would disprove the theory of gravitation"
>Thirdly, a 95% confidence level is not appropriate in every circumstance
Also works with a 99% confidence level actually.
>Fourth, you have poorly defined your hypothesis.
I have done so three times. Your hypothesis is that winning at low MMR is down to chance, mine is that your hypothesis is wrong.
>>332525157
I played Hon and dota for over 10k hours and was 1900 mmr at hon and 5,5k at dota.
But believe whatever you want.
I have seen a billion people whine ingame aswell as in forums about getting stuck at low mmr despite being good themselves.
Guess what each and every one of them i actually saw play, was shit at the game and deluded himself because of ego reasons, that he is competent.
I played around 10 accounts up to 5k in dota.
Someone in Hon actually gave me an 800 mmr account, which is pretty goddamn low and a week later it was at 1700( equals about 4k in dota)
TLDR: People are deluded and shit.
>>332525381
No, what's hard about playing support is watching you win a tri-lane, pull properly, get yourself XP and gold from the stacks, but watch your carry still not able to fucking last hit.
>>332525250
Sorry anon, but you are wrong and the other anon is right.
This is the theory "it is not other people holding you back, it is yourself"
So we have 1 guy, who has disproved it 144 times, by winning games even when people on his team are 2k players etc.
Statisticaly speaking, the chance of this 1 person, to be able to consistently climb 144 games is very, very unlikely.
So mathematically speaking, either this experiment was a very, very very lucky one, or the original hypothesis was wrong.
The fact that it is 1 guy doesn't make a difference., For example, say you have one die (dice if urr a retard), rolling it 10000 times is a good sample, even though you only have 1 die to cast.
>>332525381
>buying wards and mek for your team isnt what's hard about support.
I don't think any position 5 purchases mek anymore, your timing would be most times too late for it to be too effective.
>>332525078
It's a nice thought in theory but it's not necessarily true. You're also leaving out number of matches played which is arguably a bigger factor.
>>332523227
all matchmaking does is try to find 9 other players with a rating as close together as possible, sometimes it wont find an exact match, sometimes it will.
the point is, its a comptuter adding numbers together, not god.
just because a player who regularly plays vengeful spirit quite well has a certain MMR, doesnt mean on the random day he plays with you and decides to give lone druid a try for the first time that hes going to be equally as skilled.
or perhaps the enemy team picks a hero hes not very good at countering, or maybe his item choice is fairly good in most games, but in this particular game he gets a bit greedy and goes for something he shouldnt, or his usual choices are rendered bad because of other heroes in the game etc.
there are so many millions of factors at play, and at lower MMR levels, consistency basically doesnt exist, players might show a lot of skill one day, and be a complete retard the next.
most people who blame their own team/the other team etc always seem to be of a mind that they are static factors and dump them into either "bad player" or "good player" bins.
the best advice is to just focus 100% on your own game, dont look to blame others and watch your replays a week or so after youve forgotten the game and try to pick out all your bad mistakes and flaws in your playstyle.
you will see so much more of how bad you are when you arent trying to play a game of dota at the same time.
>>332525682
>watch your replays a week or so after youve forgotten the game and try to pick out all your bad mistakes and flaws in your playstyle.
>you will see so much more of how bad you are when you arent trying to play a game of dota at the same time.
this so much
watching your own replays really is an eye-opener to the gorillion of mistakes you make every game.
>>332525682
It's good advice, but it doesn't change the fact that you're at the mercy of your team mates in any team game.
>>332525956
since they all have the same MMR it doesn't matter
>>332526043
MMR is a very rough tool to determine ones ability. You're putting too much faith into a system with a high amount of noise.
>>332526148
noise is equally distributed so it doesn't matter either
>>332525956
I think everyone who complains about this should actually go and git gud at fighting games.
There is nothing to blame but yourself and i bet you the very same people wouldnt get good at those either and find some lame excuse instead of actually improving.
"HIS CHAR IS OP", "I PRESSED THAT BUTTON" "OMG LAG"
>>332526219
Sorry, I'm not into gymnastics.
>game measures everything, including APM
>decide that measure players abilities solely on wins and losses is the best method of ranking
>>332526148
if you look at a signle player and a single game, MMR is not very accurate.
if you look at a single player over the course of several hundreds of games, MMR is pretty damn accurate.
>>332526309
Aren't you?
But you're stretching yourself so well to justify your low MMR.
>>332517346
That's not how math works
Lets say there is one shitter in every game that loses the game
The chance of it being in the enemy team are 55%
The chance of it being on your team are 44%
Ergo you are the shitter
>>332526327
But it is.
Do you want SC to put you in Masters league just because you have a high APM too?
>>332526375
You've just proven my point. Your attack is useless because I don't even play DotA.
>>332526251
>>332525956
Hell even if bad teammates ARE actually the reason you loose, you can still do something about it instead of bitching.
You can make friends with the players in your games that you deem good and actually recruit a 5 man party, but no these insane people throw themselves into the solo grinder a billion times and hope for different results.
Take things into your own fucking hands like
a)git gud
b)recruit people
The other option is to make lame excuses and rage endlessly, which leads to getting a heart attack at 40.
>>332517346
>fucked up mmr on TBD games
Your mmr is determined in the 150 or so games you have to play before ranked is even unlocked. The 10 placement games hardly matter
>>332526347
>if you look at a single player over the course of several hundreds of games, MMR is pretty damn accurate.
Uh huh, and if your MMR fluctuates from 4k to 5k then to 2k and then back to 5k, do you think that's accurate?
>>332517346
BUT WHO ARE YOU QUOTING???!?!??!
>>332526548
>things that never happen
>>332526487
>You've just proven my point.
What point?
>>332526641
You're performing mental gymnastics. In fact, instead of proving me wrong, you continued to do more.
>>332526548
>if your MMR fluctuates from 4k to 5k then to 2k and then back to 5k, do you think that's accurate?
Are you >>332526487
Why are you talking about a game you don't know?
>>332526731
No I'm not that person. Nice post.
>>332526682
>mental gymnastics
I'm sorry, are phrases like "equally distributed" too complex for you?
>>332526548
MMR isn't accurate. It's a rough tool. People with high MMR are obviously going to be biased and put their faith into the system. Why wouldn't they? They feel entitled to a large number that they can show off for all of their hard work.
Honestly MMR is a trash system and everyone knows it. Just play the game for fun and try not to worry about the horrible rating system Valve implemented.
I can't tell you how many times I hit 6k only to drop down to 5k, but I can tell you just playing unranked and not giving a fuck about MMR makes the game more enjoyable.
>>332526849
t. 2K
>>332526815
Point still stands. Maybe you can find an account with such a MMR record.
>>332526848
No, it's just wrong.
>>332526864
This. I stopped at 5k and just played unranked. I still get mad at the shitters, even if they're doing what they're supposed to.
>>332526943
>noise isn't equally distributed
don't use words you don't understand
>>332526731
The MMR system isn't exclusive to DotA. It's a math and logic issue.
Queue in USW server, OP.
I'm 1.8k mmr (forgot). Haven't played in 2 months.
I'm a jungle gook so I play on SEA but holy shit USW is utter dogshit.
>>332526936
>Point still stands.
You didn't make a point genius, try to read.
>>332527008
ironic post
>>332526849
Anon it depends.
If someone stabilized in a certain MMR region with 50% and has like 3000 games he completely belongs there.
If someone has like 50 games and is at a high mmr i would understand your point, because he hasnt stabilized yet.
But i have never seen someone with a lot of matches seen fluctuate all that wildly, so its most likely a working system.
>>332527012
>It's a math and logic issue.
Then maybe you could start making some mathematical and logical arguments.
Is it true there is a secret behaviour MMR?
>>332527106
>But i have never seen someone with a lot of matches seen fluctuate all that wildly,
You probably haven't looked at any.
>>332527180
>secret
you can see it with a console command
it's called behaviour score
>>332527084
the point is that doesn't happen
you're free to prove otherwise
>>332527273
What command is that?
>>332527143
Every time someone does, it's met with neanderthal chest thumping: "you just need to git gud faggot"
>>332527180
What is it supposed to measure?
It would be hillarious if the system would put ragers into matches with other ragers on purpose.
Since that is sure to incite people going apeshit.
>>332527265
I have.
What now?
>>332527307
Because it often is a "scrub is overrating himself" issue. People have tremendous egos man.
>>332527452
>What now?
You actually go and look.
>>332525956
but thats true for every player in solo queue, so long as you play well and improve, your chances of winning increase and your mmr will rise.
people still try to cling to this whole "forced 50%, valve are queueing 8k 5 man teams against me and giving me all shitty russians as teammates" meme.
its all just perception.
>>332527362
>It would be hillarious if the system would put ragers into matches with other ragers on purpose.
I'm a rager and I never get matched with other ragers. Unfortunately.
>>332527301
Developer 1
dota_game_account_debug
>>332527307
well then try it, because it hasn't happened in this thread yet
>>332527362
it starts at 10,000 and lowers every time you get reported or abandon
>It would be hillarious if the system would put ragers into matches with other ragers on purpose.
it totally does. There's an anon on /vg/ who lowered it to 1. Matches must be pretty shitty at that level.
>>332527452
It goes both ways.
>>332527493
I like how you convinced yourself you're bamboozling people here when what really happened is you got rekt and you're trying to save face by going the "I was only pretending to be retarded" route.
>>332527516
>forced 50%
It is forced 50%. When I played HoN my winrate was 61%. Now it's 50% in Dota 2. I've played a lot of games in both.
>>332527635
>It goes both ways.
As is "scrub is underrating himself"? Don't see that too often, no.
>>332517346
>pick AM
>farm
>farm even more
>push mid or any side-lane and just throne them
and this is how you win a game at 3k.
just your problem will be you have no idea how to farm efficient.
>>332527591
Hm, I'm on 9075 even though I've only abandoned once and had one report. Doesn't seem to stop me from ending up playing with raging russians.
>>332527680
I like how upset you are.
>>332527516
Yes, it'll increase your chances. Just like playing basic strategy in Blackjack will increase your chances. It's still not a guarantee.
>>332527818
>Doesn't seem to stop me from ending up playing with raging russians.
They are still in the top of what doto has to offer. Whole community is pretty rude regrettably.
>>332526219
its more accurate with more playtime though.
a guy who just played 10 games probably wont have a hugely accurate mmr.
also especially at lower levels, people tend to vary wildly, not have the consistency of a higher skilled player.
sometimes they win by huge margins, sometimes they lose by huge margins, and they generally are not in control of swinging things either way, they just fuck up less than the other guys.
>>332527839
>literally "lol u mad"
called it
>>332527968
>sometimes they win by huge margins, sometimes they lose by huge margins
and you're the only constant
This thread went totally off...
>>332527591
>it starts at 10,000 and lowers every time you get reported or abandon
uh oh
>>332528141
do it anon
how low is it?
>>332527985
>being this mad over nothing
calm down, try making a reasoned argument
>>332527493
I did.
It proved me right. :^)
No really without any actual data everyone can claim eitherway.
>>332527635
Yeah it probably does, but i have seen actual bad players do it so often over the years, that i am tired of it and usually reply with longer forms of "git gud". I dont doubt some people are actually stuck without being shit, but i believe for the majority its because they are bad.
Sometimes i even used to post suggestions on which heroes to pick and which positions to play. And then all you get is: "lmao thats boring i dont wanna tryhard".
You give them a solution and they tell you that they like to play suboptimally on purpose.
>>332528141
>mfw I have been chat muted for several weeks at a time
>mfw it's still relatively high despite multiple times getting told that I am among the top percentage of reported players
>mfw I still shit talk bad players and subhumans anyway
>>332528107
Does it? I'm pretty sure we're still on the "me ooga, team booga" route.
>>332520617
>>332520617
>>332522638
It is more likely that people with higher MMR will share their MMR on their profile. Notice the bumps every thousand. So average is propably a bit lower. Considering normal MMR and all players its propably even lower, for active players its propably higher.
>>332528225
I uninstalled, brb
Dota? Arent you guys ashamed to live?
>got calibrated at 2.5k
>never broke the trench
>years later make smurf
>calibrate at 3.9k
>rise to 4.8k
it's not worth it to scrape up from the bottom OP
just remake your account and live without your cosmetics
worked for me
>>332528279
>No really without any actual data everyone can claim eitherway.
The point of making claims is to convince other people, which you have failed to do.
>>332528325
>Notice the bumps every thousand.
Pretty sure that bump is there because of all the "holy shit three more games until I hit 4K/5K/6K"
It just doesn't hold, you're telling me people wait until 6K to share their MMR?
>>332528441
And you made your claim first and didnt convince me either.
So what?
Just as i said without a large data sample this wont go further.
>>332528441
You claim there exists such people.
You only have to post one. We can't prove a negative.
You make a claim, you support it, it's not rocket science.
>>332528432
That just means you are better now, I could probably win 9 of 10 games if I were to play at less than 4k.
>>332528634
>That just means you are better now
He's exactly the same as he was before idiot
>>332529206
No player with enough skills to play even games at more than 4k would have troubles wining all their games at 2.5k, it's such a big difference in skill. I can tell you from my own experience that you can win those games by yourself, regardless of the position you play in the game, or if you want to do it easy mode just go to the offlane 1v2 or 1v3 and you have 90% of chances to win the lane and have 2 or 3 kills by minute 6, unless you're playing against an smurf account that is still getting calibrated.
>>332527756
you understand what the purpose of a matchmaking system is right?
to match you with equally skilled opponents.
if you had to play against yourself forever, what do you think the average win ratio would be?
>>332518026
Thats sad.
>tfw <2k mmr
Is playing support even worth it?
>>332532919
>wasting support on retards
It absolutely isnt, even if you are actually supporting super good they wont make use of it.
One primitive example would be that players who dont even look on the minimap dont profit from the fact that wards are up all the time.
They generally dont know how to farm decently in that bracket anyways, so picking support is a waste.
>>332530947
this is what i dont get
people thing theres some sort of hidden conspiracy of forcing 50%
its not hidden, matchmaking is supposed to find players of equal skill so you win 50% of the time
thats literally the point, if youre 50% its probably working and were you should be
Why does anyone care about other MOBAs in a world where Paragon exists?
I'm low MMR and I hate it, but not because of the quality of play. I'm actually quite happy playing against people of my level, I'm not much of an RTS guy so I'm not very good at mobas.
My issue is with the other fucking people. I genuinely can't stand them. I can understand it if you're a genuinely good player and you're telling people what to do, but you aren't. You're the same trash tier as me, even if you have got a few kills. Stop whining and saying "Report X" and start trying to help the team. Give advice, not constant abuse.Is this different at higher levels?
>>332536309
Yes, but only at the very top.
According to Juice 4K-4.5K has the worst attitude.
>>332517346
play offlane
im on my other pc but on my main i got a screen of winning 20 game sin row with clockwerk at 5k mmr
pretty much my go to mmr climber if i lose some games
why offlane clock?
1 you dont have to argue over mid
2 dont have to argue who carry
3 people usually dont wnat ot off
4 can rape enemy carry or ruin his farm if you know how to play
5 can rape enemy mid when you hit 6 or pick up haste or invis
6 only few heros can 1v1 him, like lina or silencer
7 high impact entire game, even late game you can 1v1 a carry by just spamming scepter ulti into abbysal
i always build blaidmail into scepter or invis or bkb
when i get invis is run up and only use ulti to chase down kills with blinks or other escapes
your bane are bkb nukers like lina, balde mail wont save you and you need your own bkb
