Guild Wars 2

So, I suppose I have to preface this by saying that:
A) I know I’m not the best GW2 player in the world. I also know that many people claiming to be good are often bads.
B) Yes, I know leaderboards are just a vanity board to flex. It shouldn’t be what anyone is striving for.

Ok, now that I’ve got that out of the way, here is my question:

How does Symbolic being 11-0 make any sense? So, Symbolic won 11 games in a row and is number 1. He’s played no more games since, and is still number 1.

Does this make sense to you, and if so can you please explain it:
Symbolic: 11-0 Rank 1
stunningstyles: 115-32 Rank 19

I started off in the MMR basement and had to fight through game after game after game with level 1-10’s, AFKers, or no shows (for the first few days of Yolo Q I was a bit above .500) . Lately, I finally began getting competent players on BOTH sides (Kaypud, etc), and I am now winning way more often than losing.

Starting 2 days ago, I was 24-21 and at around 20%. Over the last few days in my limited play time, I went 11-0. I am now 35-21, and I’m at 55%.

So, how does him winning 11 games in row and me winning 11 in a row have THAT big of a difference. Should I be number 1 for my 11 game streak? No. Should I be top 100? No. But should I be higher than 55%? Yes. Should he be lower than 1? Yes.

How is there no weight in games played? It seems like this system is more apt to reward you for playing LESS than playing more and also reward you for playing against players with a inaccurate MMR, instead of rewarding players for beating players with a determined MMR.

Why would the ranking system not be a living rating, where each time it is configured, they adjust your MMR based on the CURRENT rank of past opponents. My guess is, if they did that alone, the person who is ranked 1 would be way down the list because he was playing against bad competition, but at the right time.
And I guess that is all I am asking. Remove from your head player names, my own ranking and all. How does the situation that I just mention, make sense for a rating system? You can start 11-0 playing against players that are undetermined, but later determined. You then play no games for a long period and stay on top, meanwhile there are other people on the boards (not me) that are consistently beating players that are KNOWN to be good.

EDIT: I added some of my responses from below to the OP to help clarify what I am asking.

he won against player with better raiting. You won against player with rating 20%, he won at the very first day as all had the same rating (100%). Wait some weeks and everything will be fine.

Why do you care about his rating?
We got the soloQ and now all I see is hate hate hate.

Lupanic.6502:

he won against player with better raiting. You won against player with rating 20%, he won at the very first day as all had the same rating (100%). Wait some weeks and everything will be fine.

Why do you care about his rating?
We got the soloQ and now all I see is hate hate hate.

Oh boy. I prefaced and closed to clarify just so this type of response wasn’t given, and yet it was the first one I received. Alright here we go:

There is NO way that in the first day or 2 of solo q that he was playing against people with better skill, because it was pure chaos. He was playing against the same people, some worse people, and some better people. It was a mixed bag of people because of the extreme volatility of MMR, especially at an MMR wipe.

Also, the part about why do I care about him: would you like me to remove his name and say “there is a human being at rank 1 that is 11-0.” Would this make you feel better? Now, if anyone out there can actually answer or discuss the question:

How does 11-0 at the start of MMR, when it’s pure chaos, get you rank 1, then 11-0 when things finally settle in get you 30% boost.

Lupa allready explained it…

You just had a bad start in the Game… if you had won your first 12 games you will be probably the 1st in the LB…. You lost of unlucky reassons and need to climb up again. (Will takes alot of 11-0 series…) until you are getting matched with even non ranked guys…

prozon.3561:

Lupa allready explained it…

You just had a bad start in the Game… if you had won your first 12 games you will be probably the 1st in the LB…. You lost of unlucky reassons and need to climb up again. (Will takes alot of 11-0 series…) until you are getting matched with even non ranked guys…

OK. Let’s remove me from the equation since that is becoming a barrier discussion.

Does this make sense to you, and if so can you please explain it:

Symbolic: 11-0 Rank 1
stunningstyles: 115-32 Rank 19

I would also like to know the collective rank of all the players that Symbolic has played because I am pretty certain that because of the lack of integrity of the values at the start of MMR, he was playing many scrubs, as well as competent players.

It would be like in college football.

Imagine if Anet College played their supposedly difficult schedule at the start of the season. Let’s say they went 6-0 playing against Alabama, Tennessee and Oregon and other teams

let’s say that as the college football season went on, you found that Alabama, Tennessee and Oregon were all in super down years and ended up being dropped from the ranks completely because they were determined, this year, they were terrible.

How would Anet college still be able to stay number 1 in the polls, while other people around them were playing and beating, and sometimes losing, to teams that are now 100% determined to be good.

Could be I’m misunderstanding all this, so take this with a grain of salt, but here’s what I believe people are trying to explain to you.

When symbolic beat those players, they were all “Higher ranked” – they weren’t better, they didn’t earn that rank, they were just at the “base rank” soloq assigns to players before they’ve played enough games to have a rank actually reflective of their W/L Ratio.

The players you beat had already lowered themselves in the ranking system by playing enough games for the system to rank them based on their W/L Ratio. The players you beat were worth less “points” (or whatever) than the players Symbolic beat, even though Symbolics players had ranks that were not necessarily reflective of their skill level.

Here’s an example -

Solo Q is released. Since Soloq has no metric to rank players, it assumes every single player is the #1 player in the system. Symbolic won 11 games, beating teams of the “#1 players in the system”. By the time you started beating those players, the system realized they were really much lower ranked on the leaderboards, and rewarded you accordingly.

evilapprentice.6379:

Could be I’m misunderstanding all this, so take this with a grain of salt, but here’s what I believe people are trying to explain to you.

When symbolic beat those players, they were all “Higher ranked” – they weren’t better, they didn’t earn that rank, they were just at the “base rank” for soloq, since there was no way to measure how good players were (Seeing as soloq just came out). The players you beat had already lowered themselves in the ranking system – your players had played enough games to have a rank actually reflective of their skill, and you earned points based on those wins.

See my above addition to my previous quote. The problem here isn’t that I’m playing against players that lowered themselves in the ranks. It’s that the players that HE beat have SINCE lowered themselves in the rank. How does the MMR not take into affect the rank of the people you beat during each time MMR is configured (hourly).

So, lets say player A beats player B, C, D, and E. but doesn’t play F-Z.

Let’s then say because he beat them all in succession, he is rank 1. Let’s say player B later on continues to climb in the ranks as well, but player C, D, E fall dramatically to where they are now determined to be at the bottom end of the scale.

Why would the ranking system not be a living rating, where each time it is configured, they adjust your MMR based on the CURRENT rank of past opponents. My guess is, if they did that alone, the person who is ranked 1 would be way down the list because he was playing against bad competition, but at the right time.

And I guess that is all I am asking. Remove from your head player names, my own ranking and all. How does the situation that I just mention, make sense for a rating system? You can start 11-0 playing against players that are undetermined, but later determined. You then play no games for a long period and stay on top, meanwhile there are other people on the boards (not me) that are consistently beating players that are KNOWN to be good.

It’s due to glicko. The first few games you play will have a bigger impact on your rating, so if you win them all, your rating will be huge. Basically, you should just ignore anyone under 50 games. By next month, you should be ignoring anyone with under 100. After 100 games played, the majority have been against strong opponents if you have a high rating, luck is less of a factor since the games where you have 4 or they have 4 are starting to balance each other out, and you’ve played the great majority of your games with the minimum rating deviation.

TBH, I think that in a few weeks they should raise the number of games required to appear on the leaderboards. 50 might be too high this early, few people have enough time to play an average of 2 games nearly every day, but eventually. The ratings just aren’t as accurate as they should be when very few games have been played. The standard glicko rating deviation seems too high for this size population.

Alternatively, instead of decay, just raise the minimum games required by 5 each month after you first appear on the leaderboards. 1 game required seems too few, it could take several years before you’ve played enough games to get an accurate read. Even 5 won’t give an accurate rating for several months, but it’s better than 1.

Mammoth.1975:

It’s due to glicko. The first few games you play will have a bigger impact on your rating, so if you win them all, your rating will be huge. Basically, you should just ignore anyone under 50 games. By next month, you should be ignoring anyone with under 100. After 100 games played, the majority have been against strong opponents if you have a high rating, luck is less of a factor since the games where you have 4 or they have 4 are starting to balance each other out, and you’ve played the great majority of your games with the minimum rating deviation.

TBH, I think that in a few weeks they should raise the number of games required to appear on the leaderboards. 50 might be too high this early, not everyone has enough time to play an average of 2 games every single day, but eventually. The ratings just aren’t as accurate as they should be when very few games have been played. The standard glicko rating deviation seems too high.

Alternatively, instead of decay, just raise the minimum games required by 5 each month after you first appear on the leaderboards. 1 game required seems too few, it could take several years before you’ve played enough games to get an accurate read. Even 5 won’t give an accurate rating for several months, but it’s better than 1.

I appreciate your response. I think your solution is a great one. I understand the downfalls of the glicko system, and people so far in the thread have been stating what I already know.

It’s like if I was asking what the best way to remove a wine stain from a carpet and the responses I get are:

Wine is red, and it is difficult to get up. You should try to remove it.
Why did you spill wine.
You are complaining about a stain when you shouldn’t be drinking wine on carpet.