Swole, you tried to say that a normal distribution and a bell curve weren’t the same thing. You are objectively wrong. You not understanding does not make me incorrect.
I apologize for adding such a massive wall of text btw, carry on friends! There are some interesting tidbits here though.
This is super interesting. So you used to lose less points when losing to a team that was at your same mmr
Oh, you’re 100% trolling.
I said
and explained how it works.
Swole, you’re just objectively wrong here, man.
I haven’t but ill read through it. I’ve played a lot of classic pvp on the recent tbc and wotlk versions. That being said, i believe they changed the system to the more modern one for wotlk? We only had the traditional teams / starting at 1500 system for tbc.
anyway, my baseline thought on all of this is that the modern system is just not the same as what they used back then. back in the day, id be much more compelled to agree that it was close to a traditional elo system. the modern system that we have has been manipulated quite a bit tho
You explained normal distribution and I called it a bell curve and somehow I don’t know what I’m saying lmao
At any rate, whether you, Nas, Freest and etc all agree. I think you’re all doing solid work to get the conversation rolling, and at best improve the experience for everyone, and at worst highlight the poor system we currently have.
I appreciate all parties involved (:
This is just Nas trolling. There isn’t a real conversation going on.
No, you said I was WRONG for explaining a normal distribution and that I was “thinking of a bell curve”. They’re the same thing. You tried to say they weren’t.
Why are you lying and trying to backpedal?
You’re more than welcome to disagree, but nothing I’m doing is trolling.
He, along with the rest of us, don’t know how blizz is actually calculating it so all of his flinging around of elementary math is meaningless smoke and mirrors. Hes got no foundation on which to assert any of this outside of a
No, I said you’re wrong about your understanding of the MMR system. The MMR system does not function as a bell curve/normal distribution.
It does. It absolutely does.
another statement made with zero evidence. this is getting OLD
You’re free to prove that it does because it most definitely does not.
Man, we need that luduslabs guy back more than ever. By adjusting mmr to push players towards their desired rating over a large sum of games naturally causes a normal distribution. This is the basis of most ELO-based systems ASSUMING that everyone plays and keeps playing.
Man, you need to learn what the MMR system is and how it’s not a bell curve!
Explain to me with math why you think it’s not.
It’s up to you to prove that it is. I already know it’s not and I’m not doing your work for you.
You’re objectively wrong, man. Are there the same number of players at 1500 as at 2400? at 3k? No, you see increasingly small percentages at either end of the ladder.
Here’s one for DF S1
Ok so for those complaining about mmr if you are trying to compare it to blitz then blitz should be higher than Shuffle. If you are trying to compare it to 3s or 2s then you actually have a point.
According to check pvp Blitz has had 57k games played in last 24 hours, shuffle has had 215k Rounds (36k Lobbies), 3s 16k games and 2s 10k games.
Looking at this data then Blitz should have the highest ratings it mathematically makes sense.
Like I said above though shuffle should not have lower ratings than 3s or 2s based on that data. So if you want to say there is an issue there then I fully agree with you but if you are trying to say Shuffle should have as high of rated players as blitz then I do not agree.