Forums / Community / Matchmaking Feedback & Discussion

[Locked] Matchmaking and Ranking Manipulation Bans

OP ZaedynFel

  1. 1
  2. ...
  3. 6
  4. 7
  5. 8
  6. 9
  7. 10
  8. ...
  9. 11
SG FeLLoN wrote:
Actually I took a screenshot of the rank he has 3 wins out of 10 and a DIAMOND 3!!! WHATTTTTTT?? ??? ? HOW? ???
Was this his first season? We never reset MMR.

Also, depend on who he played in those 3 wins. Going 3 of 10 against Onyx will get you Diamond.
ZaedynFel wrote:
SG FeLLoN wrote:
Actually I took a screenshot of the rank he has 3 wins out of 10 and a DIAMOND 3!!! WHATTTTTTT?? ??? ? HOW? ???
Was this his first season? We never reset MMR.

Also, depend on who he played in those 3 wins. Going 3 of 10 against Onyx will get you Diamond.
No it was this season I have a screenshot of it
So as mentioned last night ...
I log into halo after a few days and I’m banned for 11 more.
I have never manipulated the gaming system or betrayed teammates or quit out of games. So I’m kinda titled at why I’m banned. I don’t know how I get this fixed or who to talk to but it’s an error l guarantee.
DaddyKT
DaddyKT wrote:
So as mentioned last night ...
I log into halo after a few days and I’m banned for 11 more.
I have never manipulated the gaming system or betrayed teammates or quit out of games. So I’m kinda titled at why I’m banned. I don’t know how I get this fixed or who to talk to but it’s an error l guarantee.
DaddyKT
You or someone with access to your console must have done something to earn a ban.

Try posting in the official forum for this kind of thing. https://www.halowaypoint.com/en-us/forums/e0d9938206f04a598ce84f2c8cadce02/topics/the-official-halo-5-why-was-i-banned-thread/19fa6556-e0b1-4248-b817-b3a99d0af5bf/postsRead through the FAQs at the top of the thread as there is lots of helpful info there.
ZaedynFel wrote:
reignotb wrote:
So what exactly characterizes as 'anomalous' behavior?
New account, plays 10 ranked matches, gets these kills:

3
2
0
0
3
1
4
1
2
0

Match 11:

23 kills
So what if I haven't played Halo 5's Ranked Multiplayer in a long time, and after getting back into it I finally start doing better on my 11th game, after being ranked?
Easily handled. The example given is a simple illustration for more complex stuff going on behind the scenes.
I'm just going to quit reading these forums 🤣... Everything I read, just makes me more, and more paranoid. I usually get 10 or more kills in a typical 4v4 game, but sometimes I get almost 20, or 25 depending on what's going on, people quitting out, etc. I feel like any average Halo Player could have drastically changing kills per game, with matchmaking as it stands today. Therefore, should I throttle my kills? Then I have 1 single friend left that plays Halo with me maybe once a month, he's a higher tiered Gold, he's never downranked, but now I'm even afraid to play with him. I have never put so much thought into Halo gameplay in my life....
What do we do with someone who clearly is griefing and doesn't want to play the game? what is the reporting for that?
This post has been edited by a moderator. Please do not post spam.
*Original post. Click at your own discretion.
Spoiler:
Show
ZaedynFel wrote:
reignotb wrote:
So what exactly characterizes as 'anomalous' behavior?
New account, plays 10 ranked matches, gets these kills:

3
2
0
0
3
1
4
1
2
0

Match 11:

23 kills
So what if I haven't played Halo 5's Ranked Multiplayer in a long time, and after getting back into it I finally start doing better on my 11th game, after being ranked?
Easily handled. The example given is a simple illustration for more complex stuff going on behind the scenes.
I'm just going to quit reading these forums 🤣... Everything I read, just makes me more, and more paranoid. I usually get 10 or more kills in a typical 4v4 game, but sometimes I get almost 20, or 25 depending on what's going on, people quitting out, etc. I feel like any average Halo Player could have drastically changing kills per game, with matchmaking as it stands today. Therefore, should I throttle my kills? Then I have 1 single friend left that plays Halo with me maybe once a month, he's a higher tiered Gold, he's never downranked, but now I'm even afraid to play with him. I have never put so much thought into Halo gameplay in my life....
Yeah, it's rubbish isn't it. If you get banned though, it won't be because matchmaking or the ban hammer is garbage. It's because you are cheating or aren't as good as you thought you were. Just ask the 343 reps on here. They'll tell you.
This post has been edited by a moderator. Please refrain from making posts that do not contribute to the topic at hand.
*Original post. Click at your own discretion.
Spoiler:
Show
Recently I found a bot in matchmaking. I wan't sure how they did this but would love to give it a go at making it myself. Would I get banned if I made a youtube video on it showing people how I would make it? I wouldn't use it online, it would be for fun. Would I get banned for this?
Boomy EU wrote:
Recently I found a bot in matchmaking. I wan't sure how they did this but would love to give it a go at making it myself. Would I get banned if I made a youtube video on it showing people how I would make it? I wouldn't use it online, it would be for fun. Would I get banned for this?
PM me the GT please and thanks.
Brok3nMik3 wrote:
Fix European servers...
in expandes sometimes it happens Is it lag ?
This post has been edited by a moderator. Please refrain from making non-constructive posts.
*Original post. Click at your own discretion.
Spoiler:
Show
This post has been edited by a moderator. Please refrain from making non-constructive posts.
*Original post. Click at your own discretion.
Spoiler:
Show
It doesn't make any sense that a team is penalized for an opponent quitting. If the full team wins the game, all four winning players go up +1 csr. However, If for some reason the team of four happens to lose to the team that had a teammate quit early, there is a loss of -30 csr for the team of four. By winning, the team with a quitter makes a gain of +30 for the win. This approach feels imbalanced and seems unfair in that the team with a quitter is catered to by maintaining csr or gaining whereas the full team can only go down in csr should they lose.(That seems like a penalty for the full team in the instance their opponent quits.)

A logical approach might maintain parameters that void any loss or gain of csr for both teams when any one individual quits whilst implementing a heftier penalty for the player who knowingly quits.
In the end, it doesn't seem fair to punish a full team when their opponent quits.
However, If for some reason the team of four happens to lose to the team that had a teammate quit early, there is a loss of -30 csr for the team of four. By winning, the team with a quitter makes a gain of +30 for the win.
This is because the team who is down a player is disadvantaged to the point where their chance of winning is much lower than the game would like. The system recognizes that in most scenarios where a team of 4 fights a team of 3, the team of 3 loses. So in the rare occurrence that a to3 actually beats a to4, the system recognizes that as the to3 performing much better than expected and deserving of a high CSR gain. If the to4 wins, that's normal and doesn't need to be rewarded any extra. A to4 beating a to3 isn't exactly impressive; it's expected.
A logical approach might maintain parameters that void any loss or gain of csr for both teams when any one individual quits whilst implementing a heftier penalty for the player who knowingly quits.
This would be easily abused. The moment a team of 4 decides a game isn't going their way, they have one person on that team quit and save the rest of the team from any rank loss. A party could go in with an alt or smurf account as the designated "quitter", so that their main account isn't negatively affected. Then they could just make a new account anytime their smurf gets banned, becoming an endless cycle of cheating the system anytime a game doesn't go their way.
Chimera30 wrote:
This would be easily abused. The moment a team of 4 decides a game isn't going their way, they have one person on that team quit and save the rest of the team from any rank loss. A party could go in with an alt or smurf account as the designated "quitter", so that their main account isn't negatively affected. Then they could just make a new account anytime their smurf gets banned, becoming an endless cycle of cheating the system anytime a game doesn't go their way.**So in the rare occurrence that a to3 actually beats a to4, the system recognizes that as the to3 performing much better than expected and deserving of a high CSR gain. If the to4 wins, that's normal and doesn't need to be rewarded any extra. A to4 beating a to3 isn't exactly impressive; it's expected.**
I am inclined to agree that a to4 over to3 victory is expected, mathematically speaking. However, I make my statement based on my over sixty days of Halo 5:Guardians arena experience coupled with opinions I've received from various experienced players. Not to mention a total of fifteen years of Halo multiplayer. More often than arithmetic would like to support, to3 over to4 victories do occur due to manipulation.

I'll provide two hypothetical scenarios for examples.

1.
Player one of Team A achieves an in game score of 12K, 5A, and 9D midgame.
Player two of Team A achieves an in game score of 9K, 8A, and 9D midgame.
Player three of Team A achieves an in game score of 10K, 7A, and 8D midgame.
Player four of Team A achieves an in game score of 0K, 3A, and 11D Midgame.
Team A has a score of 31
Team B has a score of 37 and is expected to win according to mathematical prediction.
At this point, Players One, Two, and Three can easily manipulate the system by harrassing an obviiously frustrated Player Four into quitting in order to secure a potential victory (As the game is not going their way with player four contributing to Team B's lead significantly). In this Scenario, a designated "Quitter" fails to exist as a variable and the ranking system is just as easily manipulated by a yet to be to3.
Often times, Player four of Team A, will quit out of sheer humiliation. Now given the statistics relative to Player One, Two, and Three, it would appear there is an astronomically greater chance of a "comeback victory" from Team A when player Four is manipulated out of the equation.

2.
On this occasion the ranked playlist in review is SWAT.
When a hypothetical quitter (A Player Four for example) removes themselves from the game,a massive imbalance occurs in the favor of the to3.
"Why?"might one ask?
Without shields Players are eliminated by a single headshot.
Considering all to3 players and to4 players have a similiar accuracy, you will be able to observe a simple pattern of addition occur when the point begin to add up.
Team A (to3) may only concede three points in a single rotation whereas Team B (to4) is able to concede four points. 3+3+3+3=12 compared to 4+4+4+4=16.
The reality here is that in SWAT, it is much easier for a to3 to even the odds after eliminating one player. In regular Slayer, shields make overtaking the opposing team near impossible. In shorr, in ranked Swat the system can simply defeat itself without even having to be manipulated .

If we're gonna consider there being a possibility of a 4stack having a designated "quitter" to manipulate the system being a reason to not balance the loss and gains of csr inn these situations, we must acknowledge the facts.

Teams can still easily manipulate the system in the exact manner in which you stated previously..

No matter what, there will be people who are going to manipulate the MMR,CSR ranking model.
I stand firm with many others that it does not seem nor feel fair to be punished for performing so well in a 4v4, that an opponent quits out of frustration, out of coercion, or out of simple will to manipulate the system to save the ranks of their teamates..

If Team A wins by a steak dinner measurement without any opponents quitting, they are rewarded in handsome csr gains.
If an opponent quits though, Team A is penalized with a gain of +1 CSR which may as well be nothing.

I believe a severe punishment ought to be implemented for quitters in the form of a 24HR or greater IP address ban depending on the number of times quit from a particular console in a given period of time. This may help reduce the rate of quitting and eliminate the potential for abuse by designated "quitter" alt or Smurf accounts as you said.
This post has been edited by a moderator. Please do not repost content a moderator has removed, repost a topic that has been locked, or post about forum moderation decisions. If you have a question or concern about a forum moderation decision, please private message the applicable moderator.
*Original post. Click at your own discretion.
Spoiler:
Show
This post has been edited by a moderator. Please do not bump.
*Original post. Click at your own discretion.
Spoiler:
Show
Chimera30 wrote:
This would be easily abused. The moment a team of 4 decides a game isn't going their way, they have one person on that team quit and save the rest of the team from any rank loss. A party could go in with an alt or smurf account as the designated "quitter", so that their main account isn't negatively affected. Then they could just make a new account anytime their smurf gets banned, becoming an endless cycle of cheating the system anytime a game doesn't go their way.**So in the rare occurrence that a to3 actually beats a to4, the system recognizes that as the to3 performing much better than expected and deserving of a high CSR gain. If the to4 wins, that's normal and doesn't need to be rewarded any extra. A to4 beating a to3 isn't exactly impressive; it's expected.**
I am inclined to agree that a to4 over to3 victory is expected, mathematically speaking. However, I make my statement based on my over sixty days of Halo 5:Guardians arena experience coupled with opinions I've received from various experienced players. Not to mention a total of fifteen years of Halo multiplayer. More often than arithmetic would like to support, to3 over to4 victories do occur due to manipulation.

I'll provide two hypothetical scenarios for examples.

1.
Player one of Team A achieves an in game score of 12K, 5A, and 9D midgame.
Player two of Team A achieves an in game score of 9K, 8A, and 9D midgame.
Player three of Team A achieves an in game score of 10K, 7A, and 8D midgame.
Player four of Team A achieves an in game score of 0K, 3A, and 11D Midgame.
Team A has a score of 31
Team B has a score of 37 and is expected to win according to mathematical prediction.
At this point, Players One, Two, and Three can easily manipulate the system by harrassing an obviiously frustrated Player Four into quitting in order to secure a potential victory (As the game is not going their way with player four contributing to Team B's lead significantly). In this Scenario, a designated "Quitter" fails to exist as a variable and the ranking system is just as easily manipulated by a yet to be to3.
Often times, Player four of Team A, will quit out of sheer humiliation. Now given the statistics relative to Player One, Two, and Three, it would appear there is an astronomically greater chance of a "comeback victory" from Team A when player Four is manipulated out of the equation.

2.
On this occasion the ranked playlist in review is SWAT.
When a hypothetical quitter (A Player Four for example) removes themselves from the game,a massive imbalance occurs in the favor of the to3.
"Why?"might one ask?
Without shields Players are eliminated by a single headshot.
Considering all to3 players and to4 players have a similiar accuracy, you will be able to observe a simple pattern of addition occur when the point begin to add up.
Team A (to3) may only concede three points in a single rotation whereas Team B (to4) is able to concede four points. 3+3+3+3=12 compared to 4+4+4+4=16.
The reality here is that in SWAT, it is much easier for a to3 to even the odds after eliminating one player. In regular Slayer, shields make overtaking the opposing team near impossible. In shorr, in ranked Swat the system can simply defeat itself without even having to be manipulated .

If we're gonna consider there being a possibility of a 4stack having a designated "quitter" to manipulate the system being a reason to not balance the loss and gains of csr inn these situations, we must acknowledge the facts.

Teams can still easily manipulate the system in the exact manner in which you stated previously..

No matter what, there will be people who are going to manipulate the MMR,CSR ranking model.
I stand firm with many others that it does not seem nor feel fair to be punished for performing so well in a 4v4, that an opponent quits out of frustration, out of coercion, or out of simple will to manipulate the system to save the ranks of their teamates..

If Team A wins by a steak dinner measurement without any opponents quitting, they are rewarded in handsome csr gains.
If an opponent quits though, Team A is penalized with a gain of +1 CSR which may as well be nothing.

I believe a severe punishment ought to be implemented for quitters in the form of a 24HR or greater IP address ban depending on the number of times quit from a particular console in a given period of time. This may help reduce the rate of quitting and eliminate the potential for abuse by designated "quitter" alt or Smurf accounts as you said.
Neither of your scenarios is supported by the data. In your first one, the to3 still loses 99%+ of the time, despite losing a bad teammate because, statistically, it is almost always still better to have a bad teammate than to be down a player. It does happen, but it is rare, and the CSR/MMR updates know exactly how rare it is, and adjust accordingly.

The second scenario is also rare, and not the rule.

This is based on observing more matches than your entire personal experience with the game and the anecdotes of those you know.

If we were to go back to giving the full win CSR for the situations where an opponent quits, we would also have to go back to giving the full loss CSR when a teammate quits. They have to be balanced to avoid abuse and rank inflation. We've simulated both outcomes and we currently have chosen the option that results in smaller losses rather than larger gains.
  1. 1
  2. ...
  3. 6
  4. 7
  5. 8
  6. 9
  7. 10
  8. ...
  9. 11