*** Official Dota 2 Thread ***

I think the rating certainly does give a general indication of how good you are but when your mmr is calculated by your teams oerfomance it cannot show how good you are individually.

Well it can, because over 100 games you will have both good and bad teams. If you are consistently better than everyone else at your MMR then your team will win more than lose and you will go up in MMR.

Being "stuck" at an MMR is a sign that you are suited for that MMR. I know it can be frustrating when you get horrible teams, but it does average out.
 
Every time I feel like playing DOTA I just come to this thread and it reminds me why I stopped.

Keep up the good work lads.
 
In other news I had my Dota 2 moment of fame the other night, SingSing was doing inhouses with people from his stream + Tobiwan and a few others. I managed to get into a caster slot and proceeded to get a lot of hate from Twitch chat, Tobiwan and Rime. :D
 
Well it can, because over 100 games you will have both good and bad teams. If you are consistently better than everyone else at your MMR then your team will win more than lose and you will go up in MMR.

Being "stuck" at an MMR is a sign that you are suited for that MMR. I know it can be frustrating when you get horrible teams, but it does average out.

Well you have just proven my point :confused:

It evens itself out you get good and bad teams. The game pushes you toward 50/50 so whilst I agree some players will excel and carry bad teams I cannot believe that everyone in 3k plus can solo win all their matches.
 
Well you have just proven my point :confused:

It evens itself out you get good and bad teams. The game pushes you toward 50/50 so whilst I agree some players will excel and carry bad teams I cannot believe that everyone in 3k plus can solo win all their matches.

The game does not push you towards 50/50, and you don't need to solo win. Over time, your team being good or bad will average out, meaning the deciding factor in your MMR is YOU.

I can't explain this any further, if you want to believe you are actually 4k and it's all your team-mates fault then fine, but it's not the case.

EDIT: http://www.playdota.com/forums/showthread.php?t=1398477

There's some proof for you. Obviously this is an extreme example but the principle is the same whether your "deserved" MMR is 2000 higher or 500 higher, eventually you will sway the matches enough to reach it. 122 wins 22 losses doesn't seem very 50% forced?
 
Last edited:
The game does not push you towards 50/50, and you don't need to solo win. Over time, your team being good or bad will average out, meaning the deciding factor in your MMR is YOU.

I can't explain this any further, if you want to believe you are actually 4k and it's all your team-mates fault then fine, but it's not the case.

Well it does. Valve even stated as much that win rates will normalise to 50%. But like you say this is over a lot of games when you find your correct MMR. I have lost 20 games in a row and similarly sometimes I have won 20 in a row. When things like this happen people like to see patterns and make excuses :)
 
Last edited:
I'm pretty lazy but it was in the press release when they released ranked match making. Maybe I will dig it up in a bit

Are you referring to this?

http://blog.dota2.com/2013/12/matchmaking/

If so, it doesn't say anywhere that win rates normalise to 50%. It says teams have 50% chance to win based on current MMR, that means if you are "better" than your MMR indicates, your team has more than 50% chance to win.

It says nothing about normalising win rates to 50%, just explains how RMM works.
 
If you are talking about this http://blog.dota2.com/2013/12/matchmaking/

The only thing they say about 50% is that its well balanced if each team has a 50% chance of winning.

That post on joinDota really does prove a lot of people wrong. It was also even funnier when a 3.5k player bought a 5k account because he thought teammates were holding him down and he just started feeding all the time and travelled back down the mmr ladder.
 
I think we are at cross purposes. I just said win rates will eventually normalise to 50%. Sometimes it does feels like you are stuck in ELO hell but like others have mentioned I think it is easier for carry/solo/offlane players to rise up rather than support players

edit: there are players with exceptional win rates 60% + but they are just that, the exception.
 
I think we are at cross purposes. I just said win rates will eventually normalise to 50%. Sometimes it does feels like you are stuck in ELO hell but like others have mentioned I think it is easier for carry/solo/offlane players to rise up rather than support players

Win rates normalise to 50% if you are at the correct MMR for your skill level. This is not the same as win rates being forced to normalise to 50%, which is what a lot of people think happens.
 
If you are talking about this http://blog.dota2.com/2013/12/matchmaking/

The only thing they say about 50% is that its well balanced if each team has a 50% chance of winning.

That post on joinDota really does prove a lot of people wrong. It was also even funnier when a 3.5k player bought a 5k account because he thought teammates were holding him down and he just started feeding all the time and travelled back down the mmr ladder.

If your talking about Zerrax he is holding steady at 50% win rate by playing hard support and attempting not to feed. He hasn't gone back down to his original rating as far as I know.

As for Juice, the guy is in the top 1% of Dota players playing nothing but high impact mid heroes. I don't believe that experiment proved anything conclusively at all.
 
It proves that if you are better than the ranking you are at you will rise. The problem with "support" players at 2-3k is they are so ridiculously passive that even if they have good game knowledge and keep ward vision up, don't feed etc, they don't actually have the game impact that a higher MMR support would have by roaming and ganking mid etc.
 
It proves that if you are better than the ranking you are at you will rise. The problem with "support" players at 2-3k is they are so ridiculously passive that even if they have good game knowledge and keep ward vision up, don't feed etc, they don't actually have the game impact that a higher MMR support would have by roaming and ganking mid etc.

The experiment was taken to the extremes by having a player of such a high caliber that even at the 5k+ MMR, he is still the best player.

Would be interesting to see how fast a 4k player could climb the 3k range :)
 
Back
Top Bottom