Ranked systems do not work in team-based games.

[Smitty Wrbmnjnsn]

Ranking individuals based on the performance of the entire team creates a vicious cycle where in you have many players underranked, and winning heavily against the propperly ranked, and you players overranked, acting as a handicap to the properly ranked.

The rank system cannot be based on wins or losses, but the performance of each individual. If a team loses, the top 3 players should receive no penalty. If a team wins, the lowest 2 should receive no reward. Even with this, the problem remains when a crummy player can manage to get a single kill per round, to outdo his even crummier team, but it's a step in the right direction.

Also, there should be massive penalties for leaving a ranked game, and lessened ones for the team that suffers the leaving of said teammate.

[d0mw]

thats a pretty stupid idea, if you're good enough to outperform your rank you should win.
ranked systems like that are not perfect, but it makes sense.

theres 10 people in a match, and EVERYONE should be equally skilled.
so if you are better than the 9 other players.
your team has a higher % chance to win. its simple.

[WM[SW]]

i play ranked game solo.
after qualification i have plat2.

git gud.

[liberosis]

nope . they should have solo mmr and group mmr . that way you can show the person ability and how much he contribute to a team winning

[d0mw]

Originally posted by iDeadlyfury | Finding team:

nope . they should have solo mmr and group mmr . that way you can show the person ability and how much he contribute to a team winning

Im gonna quote what remorce said:
We need DuoQing in this game like league of legends used to have, a soloQ where you can bring like 1 buddy so you can team up with maximum of 1 other player, and a seperate Queue for full teams.
(i think remorce said that, if not it was some other youtuber, no idea :^) )

i think that would be a pretty neat idea aswell.

[izotron]

Originally posted by Smitty Wrbmnjnsn:

Ranking individuals based on the performance of the entire team creates a vicious cycle where in you have many players underranked, and winning heavily against the propperly ranked, and you players overranked, acting as a handicap to the properly ranked.

The rank system cannot be based on wins or losses, but the performance of each individual. If a team loses, the top 3 players should receive no penalty. If a team wins, the lowest 2 should receive no reward. Even with this, the problem remains when a crummy player can manage to get a single kill per round, to outdo his even crummier team, but it's a step in the right direction.

Also, there should be massive penalties for leaving a ranked game, and lessened ones for the team that suffers the leaving of said teammate.

Rank in team based games needs to be only for premade teams and the rank needs to be applied for that team not the individual, but that will never happen!

Btw, your idea makes no sanse with the current point system. I can refuse to play the game and still have the highest points in my team and so on. Btw you can be the most valuable player with oppositely 0 kills and you would have no points or the highest amount of points.

Your suggestion shows that you are new to the game.

Originally posted by iDeadlyfury | Finding team:

nope . they should have solo mmr and group mmr . that way you can show the person ability and how much he contribute to a team winning

Makes 0 sanse !

> that way you can show the person ability
by having 1v1 ranked

>how much he contribute to a team winning
Impossible to measure unless you have a judge spectating the game

And this is stupid

[-|Nur|-]

Do we have to get these topics every day?

Read a book on introductory statistics, then you'll understand why complaining about bad teammates or über-enemies is missing the point about an Elo-based system.

If a team loses, the top 3 players should receive no penalty. If a team wins, the lowest 2 should receive no reward.

One of the worst ideas I've ever seen presented on this message board. This would only encourage risky plays in the hopes of boosting those individual stats at the cost of team performance.

[Smitty Wrbmnjnsn]

Originally posted by -|Nur|-:

Do we have to get these topics every day?

Read a book on introductory statistics, then you'll understand why complaining about bad teammates or über-enemies is missing the point about an Elo-based system.

If a team loses, the top 3 players should receive no penalty. If a team wins, the lowest 2 should receive no reward.

One of the worst ideas I've ever seen presented on this message board. This would only encourage risky plays in the hopes of boosting those individual stats at the cost of team performance.

If you read any of those statistics books you're talking about, you'd know that the Elo system works for chess because it IS NOT a team based system.
But you're right, my system would force selfish play, and so it is stupid.

Originally posted by WMSW:

i play ranked game solo.
after qualification i have plat2.

git gud.

Like I said... overranked players become a draw on their properly ranked teams.

[-|Nur|-]

Originally posted by Smitty Wrbmnjnsn:

Originally posted by -|Nur|-:

Do we have to get these topics every day?

Read a book on introductory statistics, then you'll understand why complaining about bad teammates or über-enemies is missing the point about an Elo-based system.



One of the worst ideas I've ever seen presented on this message board. This would only encourage risky plays in the hopes of boosting those individual stats at the cost of team performance.

If you read any of those statistics books you're talking about, you'd know that the Elo system works for chess because it IS NOT a team based system.
But you're right, my system would force selfish play, and so it is stupid.

Makes no difference whether it's a team game or not (so long as the teams are randomized!) since an Elo system relies on large numbers for accuracy, and within those large numbers the expected effect of random variations in the composition of teams on your rank is expected to be zero.

[Smitty Wrbmnjnsn]

Originally posted by -|Nur|-:

Originally posted by Smitty Wrbmnjnsn:

If you read any of those statistics books you're talking about, you'd know that the Elo system works for chess because it IS NOT a team based system.
But you're right, my system would force selfish play, and so it is stupid.

Makes no difference whether it's a team game or not (so long as the teams are randomized!) since an Elo system relies on large numbers for accuracy, and within those large numbers the expected effect of random variations in the composition of teams on your rank is expected to be zero.

You're mistaken. Elo was designed for non-random, 1v1 games like chess, wherein the skill of each player is the direct cause of each win or loss, not the teammates, not the choice of pieces.

[-|Nur|-]

Originally posted by Smitty Wrbmnjnsn:

Originally posted by -|Nur|-:

Makes no difference whether it's a team game or not (so long as the teams are randomized!) since an Elo system relies on large numbers for accuracy, and within those large numbers the expected effect of random variations in the composition of teams on your rank is expected to be zero.

You're mistaken. Elo was designed for non-random, 1v1 games like chess, wherein the skill of each player is the direct cause of each win or loss, not the teammates, not the choice of pieces.

Does-not-matter. Even in 1-on-1 chess your Elo rating is considered wildly inaccurate if it's a result of merely a few games. Your rating becomes more accurate the more you play. And when you have a large number of games, the effect of random factors becomes negligible.

[Smitty Wrbmnjnsn]

Originally posted by -|Nur|-:

Originally posted by Smitty Wrbmnjnsn:

You're mistaken. Elo was designed for non-random, 1v1 games like chess, wherein the skill of each player is the direct cause of each win or loss, not the teammates, not the choice of pieces.

Does-not-matter. Even in 1-on-1 chess your Elo rating is considered wildly inaccurate if it's a result of merely a few games. Your rating becomes more accurate the more you play. And when you have a large number of games, the effect of random factors becomes negligible.

Yes, yes. I was not arguing the increased accuracy over the course of many games. I'm arguing the fact that teammates make it inaccurate ALWAYS.

[-|Nur|-]

That's not how "luck" works.

[Smitty Wrbmnjnsn]

Originally posted by -|Nur|-:

That's not how "luck" works.

What?

[Smitty Wrbmnjnsn]

Originally posted by -|Nur|-:

That's not how "luck" works.

I'd still like an explanation on how that applies to the current argument.

[Smitty Wrbmnjnsn]

Originally posted by Gimp:

I've seen people with 0 kills come out as top, even when I've got 7 kills. Apparently, it's a 'tactical shooter' without the shooting. ;)

Then again, the Battlefield games have a similar scoring system. A mad medic can rank up those points eith ease on those indoor maps.

I see why they give you so many points for non-kills. It's meant to encourage team play, but since the people that would care about points actually care about K/D, you still get selfish idiots.