The biggest design flaw with the leaderboards is that they show only individual statistics for a match type where win/loss ratio and our “ratings” are largely attributed to who we play with and if those players are good. Mathematically put: Five men are on a team and that means that only 20% of your personal rating can actually be attributed to your own skill level. The other 80% of your personal rating can be attributed and given credit to the four other people that you play with.
Simple fix for the leaderboards:
~ Add a new statistic that shows the four players that a user plays with the most.
This would significantly upgrade the community’s ability to read and gauge the skill level of a given user. We would be able to see if his rating is being carried by his team or if he is helping carry his team’s rating.
Team ratings are absolutely more accurate than individual ratings. Seems like the best implementation of your suggestion would be actual teams with ratings.
the issue with actual teams is that gw2 pvp community is very small, there is handful of actual teams and if requirement for actual team for pvp would be implemented, the queue times would be insanely long
tbh i am not quie sure how to make tpvp an actual team pvp and not another form of yoloq given the restriction of very small community
a bit off topic, but i really wish players could queue for soloq match with 1 friend if they wish for it (like in league)… i mean there IS way for it right now but it is very luck dependant and against the rules as far as i know :/
As of how to implement an actual team leaderboard:
*Any player can only be part of one team at the time
*Starting a game would require a minimum of three players from the same team, rest can be filled with subs
*The rating of the team represented is the one that will shift mmr
*A team can have a maximum of 7 players (2 subs and 5 regulars)
Team ratings are absolutely more accurate than individual ratings. Seems like the best implementation of your suggestion would be actual teams with ratings.
Team mmr for team queue would be really better, yes.
For solo queue is different but i think more stats in general, could help to identify player value. Not only on the leaderboard but at the end of the match too (basically a recap about what combat logs says with sum and other stuff about decap/cap,ress,kills/deaths etc. with every stats can help to identify a good performance in gw2 conquest).
Team ratings are absolutely more accurate than individual ratings. Seems like the best implementation of your suggestion would be actual teams with ratings.
Nice. Agreed 100%. Next step is to make the game actually reflect the views of Arenanet staff (you).
Forum Lord Chaith
Twitch.tv/chaithh
New Twitter: @chaithhh
Team ratings are absolutely more accurate than individual ratings. Seems like the best implementation of your suggestion would be actual teams with ratings.
Completely agree with this
I have brought this up in the forums before but it seemed to get no attention
This is why I proposed the quick fix to the already existing leaderboards
Though the 4-frequent player addition would only be a half step towards team ratings
Team Rating would certainly be preferable
Funny how when i que up in soloq i pretty much already know who’s gonna be in my team, or at least in my next match just looking in fl at who’s soloqueing and who’s already in, also sometimes waiting for ppl to get out from their run in order to be (More likely) teamed up with them …late night you pretty much get always same 10 ppl just rotating from one team to the other.
Team ratings are absolutely more accurate than individual ratings. Seems like the best implementation of your suggestion would be actual teams with ratings.
Which is yet another reason why you should introduced team-based ratings and leaderboards, associated with a guild tag. In the current system, a team’s rating can be artificially skewed by what the individual players get up to when they’re not playing in the team. Imagine that a team of 5 people play together, and do well and have a high win ratio. Four of those people only play solo queue when they’re not able to get a full team together, but the fifth plays in team queue even when his teammates are not online, joining solo. As a result of the fact that he plays with a lot of randoms, the fifth player’s win ratio and position in the team queue leaderboards are much lower than his teammates, even though he’s the same skill level as them, skewing the team’s average MMR and causing them to be matched with opponents of lower skill.
If there were team-based MMRs and leaderboards, these problems would be less exacerbated.
It’d probably be possible to weight wins with players of low rank more heavily… It shouldn’t take long to tack players to see how wins are on average distributed based on rank differences.
#1 of all #1 players as ranked by a fair and unbiased committee
