forum

multiplayer team vs.

posted
Total Posts
11
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +0
Topic Starter
yukihyo
i've noticed in team vs. if all but one person are on say the red team and then just one person's on the blue team it's impossible i think it would be nice if there was some way the depending on how many people are on whatever team that there's some divide the score among how many people there are with the highest possible score of the song like say you could get 200k on a song then there's 4 people on each team then each person gets 50k each and if the SS they get 200k (plus whatever spinner changes are made to final score)
XPJ38
So if I understood correctly (please use some punctuation in the future, I had to re-read you multiple times to get your point :? ), you want to decide which team wins by doing averages of the teams players' scores and compare them?
VoidnOwO
:oops:
Topic Starter
yukihyo
Sorry, I'm not the best for puctuation, but yes, i mean like 1v4 the team of 4 people would have a score x0.25
YGOkid8
i dunno, from what i've seen and done, i only ever did 1v4 or something cause i reckoned i could beat the 4 by myself :)

well, i suppose as an option? might be useful
Oinari-sama
How do people get into 1:4 1:5 situations in the first place =.=

What's the intention for setting up those odd teams? Showing off? I hate to be the party spoiler but a simple [Team Score]/[Number of Team Players] doesn't give a "fair" handicap. Here's why:

Disclaimer - If you don't like maths/statistics you should probably skip the spoiler box and jump to TL;DR!

Simplified Statistical Reasoning
Each player have their strengths and weakness. For the sake of this exercise we'll assume that every player in the game has exactly the same ability. I will use an over-simplified example to explain why 1-vs-many match is not fair to start with and an "average" will not help to balance the issue.

For any given song, the player will have a chance for doing well, as wel as for doing bad. I'll only stick to 1 variable for simplicity's sake: chance to MISS. Since osu! calculates scores based on the combo, chance to MISS will give a fair simplification on how well the players are doing.

Recall that we've assumed that all the players in the game have exactly the same ability. If we assume that each player has a 25% chance to MISS in a 1-vs-4 scenario (I know, 25% is quite high but that's for making the calculations easier to see):

For the 1-man-team, the team's chance to MISS is 25%.

The same cannot be said for the 4-man team. Assuming that the player's ability to FC/MISS is independent on how other's are doing, the chance for each of the 5 man team to MISS is still 25%, but as a team they got 4 "tries" at missing, ie:

Chance for at least one of the 4-man-team member to MISS = 25% x 4 = 100%

In other words, one of the member on the 4-man-team is gauranteed to miss, statistically anyway. This is like the classic "throw-a-dice" problem. Statistically speaking, if you throw a dice 60 times, you have 1/6 chance of getting a "3" so your Expectation Value is 10 times in a 60 throw scenario.

Starting to see the problem now? The 1 man team is expected to miss 1 out of every 4 games, while someone on the 4 man team is expected to miss every game. This gives an edge for the 1 man team, since the averge score for the 4 man team is expected to be lower than the 1 man team, 75% of the time.

I know this is an over-simplified statistical reasoning and irl you can have a gosu vs 4 noobs going on. But what's "fair" in that match up to start with? I personally do not see a need to implement something extra for the reason stated in OP.

TL;DR: In "1 vs many" games, "averaging" the scores doesn't really give a "fair" comparison between both teams. It's best to have as evenly matched teams as possible. I can see the "average scores" be used for a 3v4 or 2v3 in a crude way though.
YGOkid8
^ has a point, but i think yukihyo was referring to when members are of similar skill... which kinda ruins the point i suppose

what about a handicap system rather than a score multiplier?
Topic Starter
yukihyo
i was just thinking it's a bit unfair if it ends up being like 5 against 1 i've had it happen to me, i tried telling someone to switch they wouldn't so my main thought is it's unfair that you pretty much can't win unless they all suck but a handicap i suppose that could work too depending, what kind of handicap
Harappy

Oinari-sama wrote:

Chance for at least one of the 4-man-team member to MISS = 25% x 4 = 100%
Just to point out, the chance for at least one of the 4-man-team member to miss = 1 - (75% x 75% x 75% x 75%) = 68.36%
At the same time, as long as one of the 4 man team manages the same combo as the 1 man army, it is certain that the 1 man will lose.
I have only found team vs fun when both sides are more or less even, in numbers and strength.

And the only times I have found 1 vs many situations happening are when someone tries to flex his muscles. There's been many times when a 'pro' forces everyone to go up against him, only to ragequit when his @$$ gets owned.
If you are stuck with no teammates and no one is willing to swap, you just got unlucky. Maybe try to find a room with less kids.
jemhuntr
meh. i support this.
Full Tablet
A system I think would be more interesting:

If the number of players in each team is uneven, discard the score of the worst players in the bigger team so the scores for each team include the same number of players.

Example: In a 2vs6 (Blue vs Red) Team Versus. Blue's score would be the sum of the 2 players in that team, and Red's score would be the sum of the highest 2 scores in the team.
Please sign in to reply.

New reply