General Discussionwhat does really matter the most? seriously

low winrate high rmm high kda means solo player

bitch pls

how are u supposed to get high kda when theres quad mid lane of russians in every single game in solo ranked matchmaking ?

there is no hero in this game that can survive 4 hero nukes at minute 5

These things are marginally useful in determining who is good at dota. But really if you want to know you just have to watch ppl play the game and you really need to watch them play in a game everyone is taking seriously against other good players.

Then there are so many different aspects to the game. Different players have different strengths and weaknesses.

There are roles people understand well or don't, (carry, initiator, ganker, support, semi-carry, pusher).

Then there are lanes and match-ups that people understand to various degrees.

Also you have to consider raw mechanics of clicking on things quickly and accurately That is really the primary factor in determining someone's stats like MMR.

Besides this drafting and hero match-ups is a different level of dota play.

The most complex, difficult thing to understand is what IS correct positioning on the map and in team fights.

All of this and more factors into who is good at dota and most of it is very hard to see from stats.

^ Vouch, because watching this doesn't tell you much.

http://dotabuff.com/matches/449423384

>implying kda matters
Actually when volvo is calculating mmr with their antismurf rating kda matters a lot.
But to be a winning player kda isn't the thing that matters the most, if you want to have perfect kda just get 1 kill and never get out of fountain in your life, here you go - infinite KDA.
Also what vaikiss said, literally every single game in solo is with at least 3 people showing up mid @ min 4-5 meanwhile our team is feeding the rest of 2 guys on other lanes and we do not have wards or ganks or even counter tps to mid.

If you want to pick something that matters most it would be ranked party MMR. Those would be the most legitimate games with full teamplay. It still doesn't mean the higher MMR party will always be better because what server, time of day, language option, and mode still can change the difficulty of matches a lot.

But ranked party MMR would be the closest your could get to a real measure of dota skill in one number at this point. Ranked solo MMR will give you a good idea of raw mechanical skills, but not decision making and teamplay. And of course you really need these numbers to be well established with tons of games. As you can see from the forum ppl's ranked MMR's are changing a lot from their initial values.

Yes. Ranked party MMR is the true MMR. But still, I don't like it because, as it seems, it works like:
if you win, +(random number)
if you lose, -(random number)
Sometimes I stomp some noobs and get +27, sometimes I win hard games vs very good players and get +4
Sometimes I lose because I played poorly and get -4, sometimes I lose because the other team is a stack with some of the best SA players and get -27

there are no stats that matters..
only skill matters
@reletless not everybody tryhard 5stack in ranked party so..~

yeah but IN THEORY the stats should measure one's skill

y but dota isnt about theory

if i follow ur logic everybody with good stats are good players, which is not true, there is a LOT of ppl with 60%+ that are not anywhere near good, and there are ppl with 50% with pro lvl so..~

the only way to mesurate one's skill would be to watch a cw when everybody tryhard i think, or ih with only pros

Theres another thing called "Game impact" sometimes you are usefull more than player with 4 times higher kda. just sayin

And there goes your rmm, a fckin wannabe junglers doom afk 20min doing nothing wont win you a game ... thats why i hate those 50% idiots picking tinker ancients, doom woods.

tinker ancient is fucking retarded nowadays, but doom jungle it's ok... when I go doom jungle I usually gank before lv3 which usually results in a kill (troll trap = kill most of the times)
but of course most people prefer afk farming til lv16

i have good winrate with meepo for example, but kda is average. dying a lot has a huge impact on your kda. as a matter of a fact not amazing kda on any other heroes, because being very aggressive player you (i) will die a ton. and at times i mistake myself for a roshan and dive like a baws.

eventually after playing this game for so long, i dont need a number to tell me how good i am and how good i am not. same goes when judging people. nitpicking on certain numbers was never a good approach to begin with.

Thats the quality of most jungle dooms ^^

http://dotabuff.com/matches/450297739

what REALLY matters is your SOLO queue winrate, nothing else.

in stacks u can get carried so that is the most unreliable shit ever, and solo mmr is broken as fuck cos before ranked was introduced there was no distinction between solo queue and party queue so in reality its not even solo mmr.

if u know u havent stacked one single game and yet manage to get 60% winrate u are gods gift to dota2.

Ranked party mmr is true mmr??
I stack with 4 braindead irl friends who have 32 cs in 50 minutes as naix with midas

^^

And those games looks like:

win: http://dotabuff.com/matches/450553194
loss: http://dotabuff.com/matches/450529774

:DD its kinda shit for them when u queue your 5k+ rating with their 3k rating T_T game is very hard for them and easy for u but u still lose :-)

This is why I don't understand people who think KDA is important. Giving an anti-Mage an entire game to free farm is way worse than having five fewer kills than deaths. I see players all the time who refuse to play actively or aggressively because they're afraid that they'll die and instead they decide to play passively even if the enemy team has far better late game carries. That's why you see five man carry teams in pubs, because no one is willing to play aggressively. People just want to get six slotted and then a-click on the enemy ancient and hope they win. To do this they will sit in jungle for 30 minutes and then be surprised when the enemies, who they let win all three lanes, are so much more farmed. You can go 20-0 as a bane by constantly killing supports and lose to their carries and you can go 0-20 as CM and still allow your team to win by creating space for your other team mates and warding well.

Kasile (or whoever you are, smurf with zero games/stats off). Dota is not a solo game; it is a team game. The solo que score will give some indication of how good you are at clicking on things. Mechanical skill is a very important aspect of the game, but its only one aspect and really not the most important.

It should be easy to see that playing solo is not as good as playing as a team with just a little thought, even if it's not inherently obvious to you. All you have to do is see how people who stack more have higher winrates. The object of the game is to WIN. If you think a stack is "getting carried" you really missed the entire point of a team game. Each player has a different job to do. Spectre is absolutely terrible at supporting. Bane can't be a carry no matter how much farm he gets. Sniper is horrible at initiating fights. OD can do plenty of dps to heroes but not to towers. Chen is great at jungling from the start, but OD has absolutely no way to do it.

The team needs all types of heroes. Some heroes need lots of farm to function. Some don't. Dota is not a game played 1 v 9 to try to get the best stats. Its a game played 5 v 5 to collectively WIN THE GAME. Stacking as 5 as a team is a superior form of dota play. That is why doing this wins more games.

If you want to play with bad friends that is your choice. But there is a reason all the pro dota teams keep the same people together. The ones that win the most keep them together for a long time. Newly formed teams struggle at first no matter how experienced and talented their roster.

You can play solo all you want. I do play solo maybe 80% of my games. It's convenient. I like to just que for a game right now and not wait for anyone. But stacking IS better. It's not cheating. It's smart. When you play with the same group and actually have a plan or at least certain "plays" you can run you play the game at a far higher level. The decisions you make as a solo player not counting on your team to coordinate... those decisions are fundamentally the WRONG decisions. Against a coordinated team an equally mechanically skilled group of 5 random ppl will nearly always lose. That is because teamplay is superior dota. Nearly everything every hero does in dota is dramatically more powerful if correctly combined with teamplay. Solo play is for little kids, for fun, not something anyone who knows anything will take seriously.

solo mmr > pt mmr, nothing to discuss here

When playing solo you can easily blame matchmaking for teaming you with someone who wants to play random hero and sucks with it. But of course, this can happen to opponent team, and then you get easy win!

solo mmr and mmr are two entirely different things.
this is not equation. there is no <=>.

people need to stop thinking bs they spew out is gospel. if you dont want to discuss, then wtf are you doing here?
good riddance.