Points Lost Should Scale Based On How Badly You Lost

[Chase.8415]

The issue with this style of matchmaking is that you don't feel rewarded for working hard and overcoming a close match, it wouldn't be really fun spending 12 minutes only to gain +5 rating. It's bad enough you need to be in top 1% just to even exit out of gold, it would be practically impossible to do it with this change.

The way ELO systems work is by taking points away from other players, if a close match awards -5 then that means the other team got +5

Edited November 4, 2023 by Chase.8415

[JusticeRetroHunter.7684]

On 11/2/2023 at 6:51 PM, Eddie.9143 said:

losing is losing, whether it's by an inch or a mile. handout generation post

The point of this post has nothing to do with free hand outs. It’s to lesson the incentive and impact of griefers/AFKer of games. 

Edited November 4, 2023 by JusticeRetroHunter.7684

[Endo.1652]

20 hours ago, Gotejjeken.1267 said:

Sure, but if you are bunkering side node you are not allowing the enemy to have it and possibly drawing their attention away from other nodes.  It's why far node camping works so well as a lot of people will feed.

If you still manage to lose while doing that, then I don't see why you shouldn't get credit for the top stats.  

As in, I've never seen anyone in an aisle getting top stats unless the entire team is spawn camping the other.  Usually, even if you 'farm' it you have to be on a build that can generate a lot of kills, healing, bunker, etc.

Even if it is opposite as you say, where someone is fighting a bunker on a node all game--at least they are doing something.  If you AFK or rage you wouldn't be getting any top stat and so should lose more points.

I think you read my conment wrong

[Saiyan.1704]

10 hours ago, Chase.8415 said:

The way ELO systems work is by taking points away from other players, if a close match awards -5 then that means the other team got +5

Except that the #1 biggest issue with games like GW2 and Overwatch is that these games are so incredibly team dependent to the point where there will be extreme outliers in rating accuracy. A prime example is how Placements work. I can literally end up in Silver3 due to a bad dice roll even though I'm farming people in that low elo after Placements.

Games like GW2, you would need to accumulate an excruciating amount of games for this elo to be anywhere near accurate... compared to say, a game like chess where a players rating is accurately determined by who they beat.

If people tryhard and lose the majority of their matches with a very close game of 450red, 500blue but are losing full points.... compared to some one who has a historically worse win/lose margin but in the same rating then why punish the players who are actually trying to win??

IF this promotes AFK's and I doubt it will, then their rating will be eons worse than people who don't AFK because their point spread will be larger when they afk. Meaning they will receive less points in a period of 155 game compared to people who decide not to afk.

I don't see a negative here but please elaborate if you disagree.

[Gotejjeken.1267]

2 hours ago, Endorphin.9147 said:

I think you read my conment wrong

Possibly, but in my experience, I've yet to see a game where I carry and don't get top stats--if I don't get top stats then the rest of the team was playing equally or better to me. 

The only thing that comes close in recent memory is a thief going far and getting literally all top stats as they kept home feeding--but the rest of us didn't have to do anything so I consider it warranted.  

That was an obvious win though--trying to think of a time where we lost and I felt I played well but didn't get any top stat, or felt someone else played better and they didn't get top stats on that loss--may have happened but I guess I'm not in high enough tier to see the win-trading / purposeful side node bunker farming for stats?

[magickthief.6492]

On 11/2/2023 at 4:51 PM, Eddie.9143 said:

losing is losing, whether it's by an inch or a mile. handout generation post

Sure thing Toretto

[Naqser.4653]

I've spent most my life on other game forums and just recently picked GW2 back up again after some switch sessions ages ago. Did play GW a lot.
In those years on other forums a topic I've been participating in threads regarding Skill Ranking. I'm assuming this OP is about the Rating players get or lose after a PvP match based on the outcome of the match.

Skill Rating systems are meant to be a tool for comparing players' relative skills to each other, or, ability to win inside the frame of the game.
The closer the Skill Ratings the more even the matches should be, with more matches played for each player the more the lifetime curve flattens out unless the player makes huge improvements.
In matches, the Skill Rating basically already gives us an outcome of the match on a probability scale, depending on systems used of course. There's always a chance that a worse player will win against a better player fair and square, no matter how small that chance is.
Dismiss it as anecdotal, it weighs extremely little;
I played in a 1v1 Quake 4 tournament years ago on a large Lan party, two tiered elimination system, first opponent was below my skill and I won. Second opponent was extremely difficult for me and I managed to win 25 kills to 24 on overtime, it was extremely intense. Third opponent trashed me hard, 25-0 for him, I went to the second tier and faced off against the opponent I beat 25-24. Only this time I got majorly trashed even faster than in my third match, 25-0 in less time than the match before. This person then went on to win the entire tournament. Probably some 0.01% chance that I'd win against that person, and I did win one time, the other 9999 matches he'd win.

The system expects a winner to a certain percentage, close matches means that the teams are more even, big score differences that the teams aren't even.
Where rating comes in is in the match up, if the ratings are way different then the game will "expect" that the higher rated team wins, and the lower rated team loses, yes?
What I don't agree with your system is that if a lower rated team loses against a higher rated team, then the lower team, expected to lose, will get "punished" more for doing poorly when they're expected to do so. If you also award the same amount to the winning team, when they're expected to win big, then there's a risk of instability in the system and no one will be rated properly.

What this system lacks is "relativity".
If let's say a team rated at 1500 would win 500-10 against a team rated at 1000, with the expected outcome that the higher rated team wins, then the loser get punished far more than an even match up of 1500vs1500, where the outcome is 500-476, and the statistical outcome is 50/50. Why should the punishment be higher when you're expected to lose?
So what I'm seeing is that table only being a representation of how the higher rated team would lose Rating based on their performance against teams of equal rating to them. The more you lose against a team of equal rating the more rating you lose. Though as I'm writing this I'm wondering if you're expecting equal rating, which would make it "fair".

Another thing with ratings is that getting one takes time. The placement matches are generally just a good pointer to where you belong, but won't put you on the absolute rating you actually are. You can have bad matches, or even streaks of bad matches, then good ones, and streaks of them, but it's the average that counts, that's your capability. You could be rated 1200, but you may also perform on a 1250 level at times when you're in a stride, or when you're not doing so good you may even be on a 1100 level. For those cases you don't want a fast system, that too will be unstable.

This however leads me to another issue.
I play a fair amount of Halo Infinite, and that game has matchmaking issues. It's short wait times for some of the most ridiculous player matchups I've seen, you can have players with single digit amount of matches played match against players having 3k+ matches. One thing you can see is each team's rating after the match, and while there are outliers, most of the matches have been close to equal, even when there's been a massive point difference. The system picks whatever is needed to mathematically get an Average Ratings for both teams that are equal, but it doesn't work in practice because some players will run laps around the lower skilled players. I'm assuming that's what GW2 does, get players together and make two teams which have on somewhat equal average ratings.
When there's a big difference in skill between the players in the match, it's a player population issue, more so when wait times are long.
That's most likely where GW2 currently is, with a PvP population issue.
There's unfortunately not going to be an easy mathematical fix to this, no algorithm is going to make players magically appear to make even matches.
Far longer wait times with no expanding skill criteria is the only solution if the player population doesn't grow, and then people have to wish others in their skill range are available at the same time, and be patient.

On the AFK issue, that happens, and overall they should go down to the bottom, they make their team lose more than they make it win. They do end up in non-low rating matches because there's no a healthy population to keep them away. Ideally the reports could function faster as I ran into a serial AFKer and someone who had played several games with said AFKer a day before.

[Bunbury.8472]

54 minutes ago, Naqser.4653 said:

Another thing with ratings is that getting one takes time. The placement matches are generally just a good pointer to where you belong, but won't put you on the absolute rating you actually are. You can have bad matches, or even streaks of bad matches, then good ones, and streaks of them, but it's the average that counts, that's your capability. You could be rated 1200, but you may also perform on a 1250 level at times when you're in a stride, or when you're not doing so good you may even be on a 1100 level. For those cases you don't want a fast system, that too will be unstable.

consider that 1200 rating, gold 1 is what anet give new accounts in pvp

which is why the rating system has such issues, because gold 1 is both brand new players & vets trying to climb to plat. you can't have a functional skill rating system, bronze silver gold plat, & put new players in the second highest bracket and then expect it to be able to make good games. if the bracket system does not have the least skilled & least experienced players in the lowest bracket & puts them in a high bracket with experienced players it will not be able to make good games.

not to mention the issues this causes with player retention. expecting new players to get farmed all the way down to their rank is perhaps one of the most moronic things I have ever seen in a game. no wonder the pop is dead, anet basically designed the system to farm new players into quitting buy giving them an inflated skill rating and making them play with far better and more experienced players. dumb as heck

[Chase.8415]

9 hours ago, Bunbury.8472 said:

consider that 1200 rating, gold 1 is what anet give new accounts in pvp

which is why the rating system has such issues, because gold 1 is both brand new players & vets trying to climb to plat. you can't have a functional skill rating system, bronze silver gold plat, & put new players in the second highest bracket and then expect it to be able to make good games. if the bracket system does not have the least skilled & least experienced players in the lowest bracket & puts them in a high bracket with experienced players it will not be able to make good games.

not to mention the issues this causes with player retention. expecting new players to get farmed all the way down to their rank is perhaps one of the most moronic things I have ever seen in a game. no wonder the pop is dead, anet basically designed the system to farm new players into quitting buy giving them an inflated skill rating and making them play with far better and more experienced players. dumb as heck

Literally every PvP game that has elo based rating has this issue, it is expected that anyone participating in ranked PvP will lose a lot of games until they reach the ratings they belong. Give me one game that doesn't have this problem

[JusticeRetroHunter.7684]

20 hours ago, Naqser.4653 said:

I've spent most my life on other game forums and just recently picked GW2 back up again after some switch sessions ages ago. Did play GW a lot.
In those years on other forums a topic I've been participating in threads regarding Skill Ranking. I'm assuming this OP is about the Rating players get or lose after a PvP match based on the outcome of the match.

Skill Rating systems are meant to be a tool for comparing players' relative skills to each other, or, ability to win inside the frame of the game.
The closer the Skill Ratings the more even the matches should be, with more matches played for each player the more the lifetime curve flattens out unless the player makes huge improvements.
In matches, the Skill Rating basically already gives us an outcome of the match on a probability scale, depending on systems used of course. There's always a chance that a worse player will win against a better player fair and square, no matter how small that chance is.
Dismiss it as anecdotal, it weighs extremely little;
I played in a 1v1 Quake 4 tournament years ago on a large Lan party, two tiered elimination system, first opponent was below my skill and I won. Second opponent was extremely difficult for me and I managed to win 25 kills to 24 on overtime, it was extremely intense. Third opponent trashed me hard, 25-0 for him, I went to the second tier and faced off against the opponent I beat 25-24. Only this time I got majorly trashed even faster than in my third match, 25-0 in less time than the match before. This person then went on to win the entire tournament. Probably some 0.01% chance that I'd win against that person, and I did win one time, the other 9999 matches he'd win.

The system expects a winner to a certain percentage, close matches means that the teams are more even, big score differences that the teams aren't even.
Where rating comes in is in the match up, if the ratings are way different then the game will "expect" that the higher rated team wins, and the lower rated team loses, yes?
What I don't agree with your system is that if a lower rated team loses against a higher rated team, then the lower team, expected to lose, will get "punished" more for doing poorly when they're expected to do so. If you also award the same amount to the winning team, when they're expected to win big, then there's a risk of instability in the system and no one will be rated properly.

What this system lacks is "relativity".
If let's say a team rated at 1500 would win 500-10 against a team rated at 1000, with the expected outcome that the higher rated team wins, then the loser get punished far more than an even match up of 1500vs1500, where the outcome is 500-476, and the statistical outcome is 50/50. Why should the punishment be higher when you're expected to lose?
So what I'm seeing is that table only being a representation of how the higher rated team would lose Rating based on their performance against teams of equal rating to them. The more you lose against a team of equal rating the more rating you lose. Though as I'm writing this I'm wondering if you're expecting equal rating, which would make it "fair".

Another thing with ratings is that getting one takes time. The placement matches are generally just a good pointer to where you belong, but won't put you on the absolute rating you actually are. You can have bad matches, or even streaks of bad matches, then good ones, and streaks of them, but it's the average that counts, that's your capability. You could be rated 1200, but you may also perform on a 1250 level at times when you're in a stride, or when you're not doing so good you may even be on a 1100 level. For those cases you don't want a fast system, that too will be unstable.

This however leads me to another issue.
I play a fair amount of Halo Infinite, and that game has matchmaking issues. It's short wait times for some of the most ridiculous player matchups I've seen, you can have players with single digit amount of matches played match against players having 3k+ matches. One thing you can see is each team's rating after the match, and while there are outliers, most of the matches have been close to equal, even when there's been a massive point difference. The system picks whatever is needed to mathematically get an Average Ratings for both teams that are equal, but it doesn't work in practice because some players will run laps around the lower skilled players. I'm assuming that's what GW2 does, get players together and make two teams which have on somewhat equal average ratings.
When there's a big difference in skill between the players in the match, it's a player population issue, more so when wait times are long.
That's most likely where GW2 currently is, with a PvP population issue.
There's unfortunately not going to be an easy mathematical fix to this, no algorithm is going to make players magically appear to make even matches.
Far longer wait times with no expanding skill criteria is the only solution if the player population doesn't grow, and then people have to wish others in their skill range are available at the same time, and be patient.

On the AFK issue, that happens, and overall they should go down to the bottom, they make their team lose more than they make it win. They do end up in non-low rating matches because there's no a healthy population to keep them away. Ideally the reports could function faster as I ran into a serial AFKer and someone who had played several games with said AFKer a day before.

made some comments a long time ago, that explain that this isn’t a problem with population, it’s a problem with elo (the algorithm) and that low population only magnifies the observation of the problem.

similiar to the YouTube algorithm, generally the same  principles apply and we can as this question: what counts as a high quality video? according to the YouTube algorithm, a high quality video is determined by practically two overwhelming factors: the title/thumbnail and people watching the video all the way through. You would think that when we watch videos you would expect it to be the utopia of art pieces or something but instead, what you see are skibidi toilet ripoffs and low efforts shitposts , shock and awe, fake news…conspiracy videos and so on. What do they all have in common? Click bait title, clickbait thumbnail, and clickbait content long enough for you to watch the whole thing without realizing it said nothing at all. Half of all social media content especially on Twitter (now called X) is photoshopped fake, made up entirely by AI, or repeating news that happened years ago as if it happened today.

You could ask a similar question; what counts as player skill? Well according to Elo only one parameter: winning or losing games… but is that really what skill is? Could it be the same type of thing that is happening in YT algorithm is happening in elo algorythm…where the algorithm is determining something that isn’t “skill” but only what the algorithm thinks is skill.

in chess, the algorithm makes sense. But in a multiplayer it becomes less obvious why it would make sense. it’s no longer just your win/loss being weighed for rating…your rating is now correlated to 10 random strangers every game, gives you a number based on the outcome of that game and then using only this number puts you in with another 10 strangers…and that is a continual process between every player so what’s happening is a correlation soup: your win and loss or “skill” is now highly correlated to everyone in the game therefor making the distribution of player skill to be random…so what is win/loss actually influenced by? It’s influenced the most by who is on your team, and who is on the other team in this correlation soup and not really your true skill…

and in a cycle it uses this random correlation soup it’s calling skill, to give you more wins and losses, to give you more correlation soup…and so on. 

let’s roll with more analogies because I like to do that. Psychopass the anime is a great show mainly about this problem: You have a system called Sybll that judges criminals based off a single number called a “crime coefficient” and the show basically goes into all the flaws such a system would have. People cheating the system, innocents being “criminals” and criminals getting away with murder, and so on. It’s a near perfect one to one analog to elo and other systems built like this: players receiving a rating and being judged by that rating alone and nothing else.

Ultimately the core problem with elo, in a multiplayer game is that the question “what is player skill?” is not truly defined by whether you only win or lose in a multiplayer game. Player skill is defined by a long and complex number of different events and experiences in every game that you’ve ever played, and such a notion is not calculable by any computer. 

like the sybll system in psychopass…generally it will do an okay job at generalizing the population…that the overall good outweighs its massive failures. Like how winning and losing games is a “good enough” metric to determine someone’s skill. But that’s the thing…it’s really not good enough and the flaws become apparent when you see the big picture for what it’s actually doing.

TLDR; In chess, the rating makes sense because that number “skill rating” truly belongs to you and you alone. In multiplayer that number does not only belong to you it belongs to every other player in the game you’ve played with. This creates a correlation soup that pulls everyone’s rating into the same zone (the middle of a bell curve) aka elo hell…and so you wind up with ameteurs with the same rating as veterans, and the overall frustration of matchmaking comes from this lack of attention to detail.

Edited November 5, 2023 by JusticeRetroHunter.7684

[Dickinson.7368]

Points gained/lost should be based on personal performance only.
And no, as long as its not full premade, its not a team game/effort.

[Bunbury.8472]

14 hours ago, Chase.8415 said:

Literally every PvP game that has elo based rating has this issue, it is expected that anyone participating in ranked PvP will lose a lot of games until they reach the ratings they belong. Give me one game that doesn't have this problem

I actually can't name a single other game where you start ranked in the second highest bracket (legend is a meme). this is the equivalent of starting dota 2 players in Ancient or Divine, it's insane. new players should be able to start at the bottom of the ranked ladder and climb like any sane ranked system. expecting a new player to play with the top 250 players in the mode get farmed constantly and then still want to play is bonkers.

[Kuya.6495]

I actually like the idea. Would be nice to have like a beta week (kinda like wvw betas) to test drive this and see how well it works before permanently implementing it.

[Multicolorhipster.9751]

22 hours ago, Bunbury.8472 said:

I actually can't name a single other game where you start ranked in the second highest bracket (legend is a meme).

Even ignoring legend, there are technically 10 divisions, not 5. After bronze each rating is broken up into 3 divisions and g1 falls directly in the middle being the 5th division one would be in starting from bronze.

The difference between g1 and p1 is 300 rating or roughly ~30 consecutive wins without losing a single match and having to win at least 2 matches to make up for every loss, and if people are being matched against others 300 rating higher than they are then that is a failure on the part of matchmaking. 300 rating would be 17% of the max possible rating you could even earn if you willfully choose to ignore legend, so close to 1/5th of the entire climb just going from g1 to p1.

22 hours ago, Bunbury.8472 said:

this is the equivalent of starting dota 2 players in Ancient or Divine, it's insane. new players should be able to start at the bottom of the ranked ladder and climb like any sane ranked system. expecting a new player to play with the top 250 players in the mode get farmed constantly and then still want to play is bonkers.

Just because the standard for top 250 is well into gold now doesn't mean that gold-rated players have gotten magically better, it's because the plat population has dwindled away to almost nothing. Vet players cannot stand the real issues like duoq, afks, match manipulation, and the terrible RNG rating system that OP is suggesting edits to, so they leave and then there is absolutely no buffer between the wintrading cheats actively trying to snipe those newer players and those newer players.

Not like there was before the plats left either, beause duoq lets cheaters pick and choose their effective MMR, so new pvpers are going to go against top players regardless so long as it exists unrestricted. There's a real issue. Where players start after placements is a nonissue.

[Deadmoose.6594]

On 11/4/2023 at 1:23 AM, Chase.8415 said:

The issue with this style of matchmaking is that you don't feel rewarded for working hard and overcoming a close match, it wouldn't be really fun spending 12 minutes only to gain +5 rating. It's bad enough you need to be in top 1% just to even exit out of gold, it would be practically impossible to do it with this change.

The way ELO systems work is by taking points away from other players, if a close match awards -5 then that means the other team got +5

If you read my post I don't say anything about changing winning points, just losing

On 11/4/2023 at 5:42 PM, Bunbury.8472 said:

consider that 1200 rating, gold 1 is what anet give new accounts in pvp

which is why the rating system has such issues, because gold 1 is both brand new players & vets trying to climb to plat. you can't have a functional skill rating system, bronze silver gold plat, & put new players in the second highest bracket and then expect it to be able to make good games. if the bracket system does not have the least skilled & least experienced players in the lowest bracket & puts them in a high bracket with experienced players it will not be able to make good games.

not to mention the issues this causes with player retention. expecting new players to get farmed all the way down to their rank is perhaps one of the most moronic things I have ever seen in a game. no wonder the pop is dead, anet basically designed the system to farm new players into quitting buy giving them an inflated skill rating and making them play with far better and more experienced players. dumb as heck

Really good point here, solving this problem might help retention a bit

On 11/4/2023 at 4:37 PM, Naqser.4653 said:

What I don't agree with your system is that if a lower rated team loses against a higher rated team, then the lower team, expected to lose, will get "punished" more for doing poorly when they're expected to do so. If you also award the same amount to the winning team, when they're expected to win big, then there's a risk of instability in the system and no one will be rated properly.

What this system lacks is "relativity".
If let's say a team rated at 1500 would win 500-10 against a team rated at 1000, with the expected outcome that the higher rated team wins, then the loser get punished far more than an even match up of 1500vs1500, where the outcome is 500-476, and the statistical outcome is 50/50. Why should the punishment be higher when you're expected to lose?
So what I'm seeing is that table only being a representation of how the higher rated team would lose Rating based on their performance against teams of equal rating to them. The more you lose against a team of equal rating the more rating you lose. Though as I'm writing this I'm wondering if you're expecting equal rating, which would make it "fair".

Another thing with ratings is that getting one takes time. The placement matches are generally just a good pointer to where you belong, but won't put you on the absolute rating you actually are. You can have bad matches, or even streaks of bad matches, then good ones, and streaks of them, but it's the average that counts, that's your capability. You could be rated 1200, but you may also perform on a 1250 level at times when you're in a stride, or when you're not doing so good you may even be on a 1100 level. For those cases you don't want a fast system, that too will be unstable.

Yes the lower rated team will get punished more for losing against the higher rated team, which would help to put them in their actual rank. If a silver team goes up against a gold team, the silver team is expected to lose and get demoted. However, crazy things happen and if they do their best with my system their rating won't take as much of a hit. You're also forgetting several variables, teams are composed of a mix of players of various ratings and skill level. Certain classes and specs have advantages over others. There's a lot at play. People afk, people duo, people win trade. As it stands right now, you can get unlucky on a team with someone afk and you're pretty much guaranteed to lose that match. This happened to me recently but we 4v5'd and won the match. Should you go on a massive losing streak because you got unlucky several times in a row?

On 11/4/2023 at 12:34 PM, Saiyan.1704 said:

Except that the #1 biggest issue with games like GW2 and Overwatch is that these games are so incredibly team dependent to the point where there will be extreme outliers in rating accuracy. A prime example is how Placements work. I can literally end up in Silver3 due to a bad dice roll even though I'm farming people in that low elo after Placements.

Games like GW2, you would need to accumulate an excruciating amount of games for this elo to be anywhere near accurate... compared to say, a game like chess where a players rating is accurately determined by who they beat.

If people tryhard and lose the majority of their matches with a very close game of 450red, 500blue but are losing full points.... compared to some one who has a historically worse win/lose margin but in the same rating then why punish the players who are actually trying to win??

IF this promotes AFK's and I doubt it will, then their rating will be eons worse than people who don't AFK because their point spread will be larger when they afk. Meaning they will receive less points in a period of 155 game compared to people who decide not to afk.

Very well said, I don't even have anything to add to this but I just want to point out how good of a reply this is

 

[Khalisto.5780]

are you sure you want that? I think you're not going to be very happy when you get that plat duo back to back and the games are 500-50

 

[Deadmoose.6594]

9 hours ago, Khalisto.5780 said:

are you sure you want that? I think you're not going to be very happy when you get that plat duo back to back and the games are 500-50

 

This already happens, so not sure what your point is. At least now people might try harder to not lose as bad, and there will be less of a chance of afks.

[Khalisto.5780]

8 hours ago, Deadmoose.6594 said:

This already happens, so not sure what your point is. At least now people might try harder to not lose as bad, and there will be less of a chance of afks.

but you losing -7 to -9 in those matches, and no afkers don't care about their rating, you giving them tools to hurt you even more

"oh look, that dude that got pissed last game, let's make him lose another 29" but this time the match was completely winnable

what we should be doing is make a thread on reddit and try to upload videos from afkers there to make anet look bad for not banning players, cuz we can't do it here

[Sahne.6950]

Correct me if i am wrong! i think:

Points lost stand in correlation to the difference in MMR between the two teams.    Nothing more nothing less.

How many points you will loose on a loss and how many you will gain when winning is determined the second you all hit accept. 

 

Anything else can and will promote selfish gameplay.  

If you loose 21 points on a loss, your team was expected to win this game according to the MMR... but you still lost.  

If you only gained 7 points for a win....  you were supposed to win that game by a landslide. 

 

The current low population makes you experience those extremes more than usual.

Edited November 8, 2023 by Sahne.6950

[Sahne.6950]

41 minutes ago, Khalisto.5780 said:

i always come to this posts trying to stack as many confused emotes as possible,

let me help you with that.

[Dawdler.8521]

3 hours ago, Sahne.6950 said:

If you only gained 7 points for a win....  you were supposed to win that game by a landslide. 

If the matchmaker is supposed to make a decent match, why should you ever be expected to win by a landslide? 

If that’s the case I want to be in a team of the shittiest low ranks always matched against the best players in the world.

I will assume the matchmaker is supposed to make me loose by a landslide so logically if we can get just 1 point or more that should be net positive score and a rise in ranks!

[Eddie.9143]

On 11/4/2023 at 1:04 PM, magickthief.6492 said:

Sure thing Toretto

Lol glad someone caught it 

[shion.2084]

I agree that rank loss or gain should be related to how close the match was.  A tiered system, within 50 pts its less of a rank loss than if you lose by 450.  Thresholds can be workedd out, but basically if in the current system your expect to take say a -15 loss, perhaps your only take a. third of that loss if you are within 50 pts of each other.

The issue with this of course is that you don't want folks to just grind games to have high score, so you can't give the winning team their full amount either if the game is close.  Otherwise what would happen is just playing more games would have you risking less every time than you stood to win.

 

[Sahne.6950]

22 hours ago, Dawdler.8521 said:

If the matchmaker is supposed to make a decent match, why should you ever be expected to win by a landslide?

On 11/8/2023 at 11:34 AM, Sahne.6950 said:

The current low population