Guild Wars 2

Of having a matchmaking system where it takes 10 players within a pip range, and tosses them into a game without any manipulation?

InsidiousWaffle.7086:

Of having a matchmaking system where it takes 10 players within a pip range, and tosses them into a game without any manipulation?

This would be a much more fair system.

However, it’s impossible to apply to GW2: not enough players, by a long shot. And the numbers are going down every day apparently.

I’m not sure what you mean, the system already finds 10 players within 2 minutes, so I dont think that will be a problem

InsidiousWaffle.7086:

I’m not sure what you mean, the system already finds 10 players within 2 minutes, so I dont think that will be a problem

Yeah, by widening more and more the pips-range from which to choose.

Example, the other day I played a match in duoq, my team was a double duoq+soloq of all sapphires (I was at T4, 15 pips total), enemy team had a duoq (ruby+sapph) and 3 soloq (2 rubies + sapphire). So, at best, 3 of their players were +10 pips from me.

So, in order to shorten the queue times, the system widens the accettable range from which to choose. If you limit this range to just 3 pips (0/+1/-1), queue times will skyrocket, we don’t have enough players.

Have you, moreover, noticed how many times you get the same people over and over in your matches, even for 3-4 consecutive games?

Not, enough, players.

actually, this is created by the fact that it finds two sets of players, each group having it’s own mmr baseline. If you eliminated it, you could probably find players more easily

If I’m understanding this correctly I think it would not be a good idea.

Providing equal chances does not result in equal outcomes.

Roulette:
Player 1 betting on Red
Player 2 betting on Black

Both players have equal chance of success.
However the outcome will not be equal and will be dictated by luck.

The element of time in an even chance game can make this equitable in the long run but standard deviation ruins it’s chances over a short/fixed term. And the ‘long run’ can be a long kitten time!

In a competitive environment, an uncomplicated sytem to provide real data as to a player’s performance and relative standing should be obtainable.

What this system has done is muddy the mathematical waters by trying to shoe-horn an unnecessary ‘Ranking’ template over what could otherwise be a robust, accurate, numerical representation of skill – solo queue vs premade caveats apply :-)

InsidiousWaffle.7086:

Of having a matchmaking system where it takes 10 players within a pip range, and tosses them into a game without any manipulation?

should happen

Dirtyrascal.1023:

If I’m understanding this correctly I think it would not be a good idea.

Providing equal chances does not result in equal outcomes.

Roulette:
Player 1 betting on Red
Player 2 betting on Black

Both players have equal chance of success.
However the outcome will not be equal and will be dictated by luck.

The element of time in an even chance game can make this equitable in the long run but standard deviation ruins it’s chances over a short/fixed term. And the ‘long run’ can be a long kitten time!

In a competitive environment, an uncomplicated sytem to provide real data as to a player’s performance and relative standing should be obtainable.

What this system has done is muddy the mathematical waters by trying to shoe-horn an unnecessary ‘Ranking’ template over what could otherwise be a robust, accurate, numerical representation of skill – solo queue vs premade caveats apply :-)

This system removes all variables except the player (over time). This should naturally give you around a 50/50 win rate. If you are losing the majority of games, the constant could be definitively shown as you, not matchmaking, and the same goes for winning.

The best part is, there is no bias(manipulating the players on each team to create a good matchup), and no victimization(because matchups are random). This would generally make the community happier, because they can’t blame arenanet, and they can only be upset with other players.

InsidiousWaffle.7086:

sorry, but you clearly have bias about the topic, because you are on the leading edge of things. That kind of invalidates you opinion on matchmaking changes, because you are directly benefiting from the current system

What part of what I said do you disagree with: that I don’t want to be partied with droolers, or that me losing sometimes will make me stay in Amber for longer and thus spend more time beating up on bad players?

I wasn’t aware that any of that was opinion.

I don’t wan’t
^
Opinion

anyways, if you are a good player, you should still have trouble progressing through ranks. Progression should require effort, some might even say grind, to get the best rank, and there should be prestige associated with it.

Galandil.9641:

Yeah, by widening more and more the pips-range from which to choose.

Example, the other day I played a match in duoq, my team was a double duoq+soloq of all sapphires (I was at T4, 15 pips total), enemy team had a duoq (ruby+sapph) and 3 soloq (2 rubies + sapphire). So, at best, 3 of their players were +10 pips from me.

So, in order to shorten the queue times, the system widens the accettable range from which to choose. If you limit this range to just 3 pips (0/+1/-1), queue times will skyrocket, we don’t have enough players.

Have you, moreover, noticed how many times you get the same people over and over in your matches, even for 3-4 consecutive games?

Not, enough, players.

But where was the person you were duo-queuing with at?

Pakkazull.6894:

But where was the person you were duo-queuing with at?

He was at T3.