Jump to content
  • Sign Up

Matchmaking algorithm too rigged,


Bast.7253

Recommended Posts

@Namless.4028 said:@"rwolf.9571" in gold3/plat1 your impact on your matches is very low since you get matched regulary with plat3/legend tier players

What do you mean by "impact"? As in less of a statistic?

I get matched with gold 3 or lower more than those in Plat 1 or higher. I had a player thinking I was speed hacking the other day on my Sword Weaver the other day, until I linked 4 of my high mobility skills too him.

Link to comment
Share on other sites

  • Replies 203
  • Created
  • Last Reply

Top Posters In This Topic

  • ArenaNet Staff

@"cptaylor.2670" said:

@Ben Phongluangtham.1065 said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

Im curious about what the average rating difference in a match is when there is a legend tier player in it? This season or last season or whatever

The sample of games with a legend player is pretty low, so I extended it down to 1700 (plat 3+). Here's the data including last season and this season:Average skill rating difference between teams: 14.16Average standard deviation difference between teams: 13.55Average rating difference in a match: 189.71Percent of games with average skill rating difference >50: 0.4%

There's a significant increase in rating range at this level, but as seen the skill rating mean and standard deviation differences between teams are pretty similar. The rating range is always going to be higher at the edges of the rating curve as a tradeoff with keeping reasonable queue times , but it doesn't stop the matcher from making fair teams which is the most important.

As some others have pointed out, the average game doesn't always tell the whole story. So I also grabbed all matches from last and this season and paired it down to the set of games with a rating range over 200 (these account for about 10% of all matches). Here's the data for that set:Average skill rating difference between teams: 19.83Average standard deviation difference between teams: 33.67Average rating difference in a match: 279.42Percent of games with average skill rating difference >50: 7.5%

While the skill rating and standard deviation differences are a bit higher in this set, these numbers aren't too bad overall.

I'm not trying to say that the matcher is perfect, but the vast majority of games are pretty balanced.

I should probably be able to understand what this means, but it's not making a whole lot of sense to me.

Average skill rating difference between teams being the average of the ratings of the 5 person on one team being a 20 point difference between the other team?Deviation difference between teams? Not sure what this is.

Rating difference in a match between the lowest rated player and the highest? Is this on the same team or just everyone in both teams? Either way that's a pretty huge difference.

Percent of games with average skill rating difference? Not sure what this is either, but it's weird that it jumps from .4% to 7.5% just in one season? I know that's still less than 10% but it's huge in comparison to the previous. This might not mean anything given that I'm not entirely sure what it is to begin with. lol

@"cptaylor.2670" said:

@Ben Phongluangtham.1065 said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

Im curious about what the average rating difference in a match is when there is a legend tier player in it? This season or last season or whatever

The sample of games with a legend player is pretty low, so I extended it down to 1700 (plat 3+). Here's the data including last season and this season:Average skill rating difference between teams: 14.16Average standard deviation difference between teams: 13.55Average rating difference in a match: 189.71Percent of games with average skill rating difference >50: 0.4%

There's a significant increase in rating range at this level, but as seen the skill rating mean and standard deviation differences between teams are pretty similar. The rating range is always going to be higher at the edges of the rating curve as a tradeoff with keeping reasonable queue times , but it doesn't stop the matcher from making fair teams which is the most important.

As some others have pointed out, the average game doesn't always tell the whole story. So I also grabbed all matches from last and this season and paired it down to the set of games with a rating range over 200 (these account for about 10% of all matches). Here's the data for that set:Average skill rating difference between teams: 19.83Average standard deviation difference between teams: 33.67Average rating difference in a match: 279.42Percent of games with average skill rating difference >50: 7.5%

While the skill rating and standard deviation differences are a bit higher in this set, these numbers aren't too bad overall.

I'm not trying to say that the matcher is perfect, but the vast majority of games are pretty balanced.

I should probably be able to understand what this means, but it's not making a whole lot of sense to me.

Average skill rating difference between teams being the average of the ratings of the 5 person on one team being a 20 point difference between the other team?

Yeah. Average skill rating of team A is within 20 points of the average skill rating of team B. On average.

Deviation difference between teams? Not sure what this is.

Standard deviation is the average of how far is each team member away from the average skill rating. We want the standard deviation from the mean rating for team A to be close to that of Team B. Generally, it should mean that they have a similar spread of skill ratings on each team.

Rating difference in a match between the lowest rated player and the highest? Is this on the same team or just everyone in both teams? Either way that's a pretty huge difference.

Yeah, it's a high number at high ratings. However, as evidenced by the previous data, both team are generally dealing with the same type of spread. So it's usually an even match, from a skill rating stand point.

Edit: also keep in mind that the 280 variance number was obtained when only looking at matches where the rating difference was over 200. This was to show that even though the variance between the lowest and highest player in the match was high, the teams are still usually evenly matched.

Link to comment
Share on other sites

  • ArenaNet Staff

@Rufo.3716 said:I've pretty much come down to ranked as being a farm fest and having fun. I seriously just got out of 2 matches as a silver player (been high gold for pretty much ever) got 8 points and 10 points.

Another question I have is, is MMR and ranked rating the same, or are there 2 different rating for ranked and unranked matchmaking. I seem like I play with higher level players in unranked vs. ranked.

Your unranked and ranked ratings are different.

Link to comment
Share on other sites

Is there a possiblity to have it looked at. Every description matches my experiences with PvP, getting worse and worse every season.

I just barely made it out a 6 match loosing streak, loosing 25+ rating per match.. it seems that they are all unfair balanced, as matches i lost end with a score of <250 vs 500getting teammates getting steamrolled outside spawn, or simply dieing the moment they touch the point. the player rating differences are too high. i feel like get stuck in a team of bronze/silver players, while we have matches against teams with plat rating.

as many state: i am not a great pvp player, but i know i am good enough to fight plat matches, as long as i don't have 4 different team mates that aren't capable of doing so.,,

Link to comment
Share on other sites

I've been playing now for about 20 ranked matches and lost 17(!). While you could always argue about my personal skill etc. It's simply impossible that in 17 matches the other team can be constantly better. I've a friend with the same problem. She won last season easily and now had no chance at all. When we joined the queue together we always had someone afk or playing Leeroy Jenkins. I don't wanna say that I'm a very good player like a friend..... But still I should've won statistically at least one game in the last 17 matches. I won the 3 first matches when I couldnt even play my profession good as now. Then I lose 17(?).

It's my impression that because of the fact that we just start "now" in this season - that we get paired up with people that dont even try to read. Everytime I try to talk about strategy or other things there's simply no response.

Now that's nothing that Anet can fix concerning not talking players; though it would be nice if Anet could fix the algorityhm. I've read in the past in the forum that Anet employee's can see the amount of matches I've lost on my account.

Me and my Guildmate don't seem to be the only ones with that problem. I think it has to do with the fact that we started later this season and now somehow get paired up with people that go afk don't talk etc. What I mean: Due to the advanced time in the season we've worst starting conditions.

Link to comment
Share on other sites

@"Ben Phongluangtham.1065" said:

Rating difference in a match between the lowest rated player and the highest? Is this on the same team or just everyone in both teams? Either way that's a pretty huge difference.

Yeah, it's a high number at high ratings. However, as evidenced by the previous data, both team are generally dealing with the same type of spread. So it's usually an even match, from a skill rating stand point.

Edit: also keep in mind that the 280 variance number was obtained when only looking at matches where the rating difference was over 200. This was to show that even though the variance between the lowest and highest player in the match was high, the teams are still usually evenly matched.

I'd like to highlight here, that while NUMERICALLY the spread is relatively even, the idea that "both teams are dealing with the same type of spread" is akin to "both teams have to deal with the same enormous problem, so we think that's ok." I'm sorry but that is just confirmation for me that there is NO INTENT to improve the problem from the foundation up but simply bandaid it relentlessly. That's lazy in my eyes. This is probably the only game on the market that I can think of that specifically targets the players that are most invested in the game mode, and punishes them for it. That's unreasonable.

Link to comment
Share on other sites

@"Ben Phongluangtham.1065" said:Edit: also keep in mind that the 280 variance number was obtained when only looking at matches where the rating difference was over 200. This was to show that even though the variance between the lowest and highest player in the match was high, the teams are still usually evenly matched.

Ok Ben, I believe this statement right here sheds a bit of light on one of the bigger roots of this problem. I'm sure you devs have heard me explain this numerous times in different threads but I am going to do it again:

  • 10 mmr is high - 1 mmr is low
  • 10x players and only 10x players sitting in a que, they have mmrs of - 10, 4, 4, 4, 3, 3, 2, 2, 2, 2
  • System attempts a perfect average party vs. average party being RED 10, 2, 2, 2, 2 = 18 vs. BLUE 4, 4, 4, 3, 3 = 18
  • Looks perfect on paper but is completely unfair in actual conquest application
  • What happens here is that the elevated base mmr & rating of the highest player has fished in every lower 2 player to put on his team, to weight him against a team of averages. The problem here is that the high mmr player can viciously defend every node he is at, kill everyone, and never die, but while he is doing that, his 2s are being crunched on the other two nodes by enemy 4s and 3s. Since the game is about winning by holding 2 nodes, not winning from a single player's performance on 1 node, this situation generally ends up as a swift GG loss for the higher rated player.
  • This is where forum stories come from about "my teammates are exploding on contact and the other team is clearly better than my team, I cannot carry matches like this." <- this is happening in matches that could be labeled as "perfect average party rating vs. average party rating." This is because only balancing average party vs. average party is not adequate. It needs to be balancing for both average party vs. average party, and lessening the marginal difference in ratings between team mates so the above example doesn't happen so often. This is also where forum stories come from that depict examples of "how much lower my team rating was than the opposing team" when in fact, it was a perfectly averaged rating vs. rating. It just felt like it was a 200 rating difference because the 10 can't carry his 2s against 4s, and was only ever able to hold 1 node his color, the entire match. <- That is a broken algorithm function and I understand the critical importance behind keeping que times as low as possible but it may be time to consider elongating the que times in interests of better match making so the above example isn't happening so often. Most veteran players who are the pumping heart & blood of the spvp community, would rather wait and have 2 or 3 actually balanced matches a day, rather than have 20 super random matches that aren't balanced at all.

But all of the above still however, does not explain how or why some given 1500 plat player will be given 10x matches in a row where he is plat1/gold3/gold3/gold2/gold2 vs. plat3/plat2/plat1/plat1/plat1 or some otherwise clearly lopsided split, and for those 10 games, never once be put on the high team favored to win, despite there clearly being other plat 1s out there who are. This IS happening and it happens too often to be a coincidence. For those of us who play in the top 250, it is easy to see when it is happening because everyone in the game is on the leaderboards and you can easily view their ratings. It is very visible when some algorithm function is directing you towards a lose streak, when game after game on a fresh streak, a player takes screenshots and/or add contacts to view ratings afterwards, and can see that most of or even none of their teammates are on the leaderboards, but then every player on the enemy team is playing between plat 3 and plat 1, sometimes maybe even 1800+. I mean, can you explain what is happening there? This kind of strange streak happens to me personally about two or three times a season, every season, like clockwork. It isn't random, it's very predictable. It always happens when I'm approaching around 65% or 70% win rate then BAM a lose streak that always lasts long enough to make sure I go back down to around 49% or 51% win rate. Then it lets up and it lets me play normally again. Then I start hitting around 65% win and BAM it begins happening again, with the type of matches I was describing, clearly lopsided splits where my team is being funneled to lose for many matches in a row.

I wanted you to know, incase my posts come off in the wrong way, that I am sincerely just trying to give productive feedback. You know, I had decided to walk away from gaming for awhile in interests of other activities, but for the purposes of this thread, I may come back to play 50 -100 matches with screenshots/player rating evidence of them all to post. With all the assumed lack of evidence going on around here, I think it may help.

Link to comment
Share on other sites

@Crinn.7864 said:The rating systems distributes across a curve with a mean of 1200. Gold starts at 1200, so a player that is placing Gold is merely slightly above average. Your rating placement is done with the exact same glicko2 algorithm that rates you after placements. Placements is merely a mechanic for hiding your rating change for the first 10 matches, the reason the first ten are hidden is because glicko2 has enormously high deviation in early matches, and devs don't want players flipping out over the rating swings during early matches.Thank you for quoting the rating algorithm but I still don't understand how (and if) individual skill and game performance are calculated when defining a player's rating.

@Cal Cohen.3527 said:Average rating difference in a match: 279.42That explains why I 've been getting teammates with absolutely no clue of the mode.

@Trevor Boyer.6524 said:Ok Ben, I believe this statement right here sheds a bit of light on one of the bigger roots of this problem. I'm sure you devs have heard me explain this numerous times in different threads but I am going to do it again:

  • 10 mmr is high - 1 mmr is low
  • 10x players and only 10x players sitting in a que, they have mmrs of - 10, 4, 4, 4, 3, 3, 2, 2, 2, 2
  • System attempts a perfect average party vs. average party being RED 10, 2, 2, 2, 2 = 18 vs. BLUE 4, 4, 4, 3, 3 = 18
  • Looks perfect on paper but is completely unfair in actual conquest application
  • What happens here is that the elevated base mmr & rating of the highest player has fished in every lower 2 player to put on his team, to weight him against a team of averages. The problem here is that the high mmr player can viciously defend every node he is at, kill everyone, and never die, but while he is doing that, his 2s are being crunched on the other two nodes by enemy 4s and 3s. Since the game is about winning by holding 2 nodes, not winning from a single player's performance on 1 node, this situation generally ends up as a swift GG loss for the higher rated player.

And the above explains what happens in reality and nicely describes the 'blowout matches' which are the most annoying given that you normally expect a decently balanced fight.

Link to comment
Share on other sites

@Dreddo.9865 said:

@Cal Cohen.3527 said:Average rating difference in a match: 279.42That explains why I 've been getting teammates with absolutely
no clue
of the mode.Have you actually read the whole post or you just got instantly triggered (like many others) by seeing a high rating difference?

  • 10x players and only 10x players sitting in a que, they have mmrs of - 10, 4, 4, 4, 3, 3, 2, 2, 2, 2
  • System attempts a perfect average party vs. average party being RED 10, 2, 2, 2, 2 = 18 vs. BLUE 4, 4, 4, 3, 3 = 18
  • Looks perfect on paper but is completely unfair in actual conquest application
  • What happens here is that the elevated base mmr & rating of the highest player has fished in every lower 2 player to put on his team, to weight him against a team of averages. The problem here is that the high mmr player can viciously defend every node he is at, kill everyone, and never die, but while he is doing that, his 2s are being crunched on the other two nodes by enemy 4s and 3s. Since the game is about winning by holding 2 nodes, not winning from a single player's performance on 1 node, this situation generally ends up as a swift GG loss for the higher rated player.Based on your logic, the top100 players must be completely random as matchmaking is random and people are getting uncarriable games.

For some reason though, you can see the same names in each and every season (more or less) in top100. How would you explain that? Always getting lucky?

Link to comment
Share on other sites

@Trevor Boyer.6524That's precisely it, I kept seeing the same posts of yours through the forums but everyone just ignores them.I find it hard to believe that pvp devs are really incapable of understanding this concept that a single pro player cannot carry a game with 3 objectives spread out over the map.This is precisely the reason why so many matches are steamrolls.Either that or as ben said, we just need to stop "tilting".

Link to comment
Share on other sites

@Nappa.1904 said:@Trevor Boyer.6524That's precisely it, I kept seeing the same posts of yours through the forums but everyone just ignores them.Because most people understand how the matchmaking works.I find it hard to believe that pvp devs are really incapable of understanding this concept that a single pro player cannot carry a game with 3 objectives spread out over the map.Untrue. While there are unwinnable games, the numbers of these are simply low. Most pro players can do it, that's why they are always on top (logical).This is precisely the reason why so many matches are steamrolls.As it has been explained by devs like a million times, steamrolls are not the reason of matchmaking working badly. You can have 10 players of exactly the same rating, and the match could still end up 500-50. There are so many variables you have to take into account, how can you not understand this? (team composition, maps, carried players and soooo on)

Personally I've been low plat t3 constantly for the last few seasons regardless of the meta. I was placed in silver, gold, platinum, everywhere by placement matches, but regardless of that I quickly easily carried the games until plat t1.

This is also true from the other side, I got placed once very high plat t3, and quickly dropped to high t2. So I have a feeling that's exactly my skill level, seems to be working for me.

Also, there are people who open threads every season about how impossible it is to climb and carry in lower divisions, even though it means they are just not good enough (not counting the rare examples like we have seen in the statistics). It's the sad truth. If you are good enough, you will climb. You will lose unwinnable games here and there, but eventually you will get to your level

Link to comment
Share on other sites

@Trevor Boyer.6524 except, if this is true:

@Ben Phongluangtham.1065 said:

Standard deviation is the average of how far is each team member away from the average skill rating. We want the standard deviation from the mean rating for team A to be close to that of Team B. Generally, it should mean that they have a similar spread of skill ratings on each team.

your rating spread statement shouldn't be applicable in the first place.

Link to comment
Share on other sites

@Trevor Boyer.6524 said:

@Ben Phongluangtham.1065 said:Edit: also keep in mind that the 280 variance number was obtained when only looking at matches where the rating difference was over 200. This was to show that even though the variance between the lowest and highest player in the match was high, the teams are still usually evenly matched.

Ok Ben, I believe this statement right here sheds a bit of light on one of the bigger roots of this problem. I'm sure you devs have heard me explain this numerous times in different threads but I am going to do it again:
  • 10 mmr is high - 1 mmr is low
  • 10x players and only 10x players sitting in a que, they have mmrs of - 10, 4, 4, 4, 3, 3, 2, 2, 2, 2
  • System attempts a perfect average party vs. average party being RED 10, 2, 2, 2, 2 = 18 vs. BLUE 4, 4, 4, 3, 3 = 18
  • Looks perfect on paper but is completely unfair in actual conquest application

Which is why there's also a standard deviation.Red:Mean: 3.6S.D: 2.6Blue:Mean: 3.6S.D: 0.5That's over 5 times a difference in standard deviation.

Link to comment
Share on other sites

@"Exatherion.3075" said:I've been playing now for about 20 ranked matches and lost 17(!). While you could always argue about my personal skill etc. It's simply impossible that in 17 matches the other team can be constantly better. I've a friend with the same problem. She won last season easily and now had no chance at all. When we joined the queue together we always had someone afk or playing Leeroy Jenkins. I don't wanna say that I'm a very good player like a friend..... But still I should've won statistically at least one game in the last 17 matches. I won the 3 first matches when I couldnt even play my profession good as now. Then I lose 17(?).

It's my impression that because of the fact that we just start "now" in this season - that we get paired up with people that dont even try to read. Everytime I try to talk about strategy or other things there's simply no response.

Now that's nothing that Anet can fix concerning not talking players; though it would be nice if Anet could fix the algorityhm. I've read in the past in the forum that Anet employee's can see the amount of matches I've lost on my account.

Me and my Guildmate don't seem to be the only ones with that problem. I think it has to do with the fact that we started later this season and now somehow get paired up with people that go afk don't talk etc. What I mean: Due to the advanced time in the season we've worst starting conditions.

The first three matches are those that have the most impact on your rating (+80 to +50 rating in one match as gw2 efficiency shows) since you won those games you might be higher ratet as you deserve.Also you got probably tilted after such a loosing streak and took more risks than you would do normally (so you probably played worse than in the first 3 games)I wont deny that bad teammates are also a factor, but not the only reason for your loosing streak.

Link to comment
Share on other sites

@Cal Cohen.3527 said:

@Ben Phongluangtham.1065 said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

Im curious about what the average rating difference in a match is when there is a legend tier player in it? This season or last season or whatever

The sample of games with a legend player is pretty low, so I extended it down to 1700 (plat 3+). Here's the data including last season and this season:Average skill rating difference between teams: 14.16Average standard deviation difference between teams: 13.55Average rating difference in a match: 189.71Percent of games with average skill rating difference >50: 0.4%

There's a significant increase in rating range at this level, but as seen the skill rating mean and standard deviation differences between teams are pretty similar. The rating range is always going to be higher at the edges of the rating curve as a tradeoff with keeping reasonable queue times , but it doesn't stop the matcher from making fair teams which is the most important.

As some others have pointed out, the average game doesn't always tell the whole story. So I also grabbed all matches from last and this season and paired it down to the set of games with a rating range over 200 (these account for about 10% of all matches). Here's the data for that set:Average skill rating difference between teams: 19.83Average standard deviation difference between teams: 33.67Average rating difference in a match: 279.42Percent of games with average skill rating difference >50: 7.5%

While the skill rating and standard deviation differences are a bit higher in this set, these numbers aren't too bad overall.

I'm not trying to say that the matcher is perfect, but the vast majority of games are pretty balanced.

Thanks for the response, i was genuinely curious. I dont doubt the balance of the games. The only problem i really have with the games is that sometimes your influence in the game feels meaningless since regardless of how well you do your team can still fuck you over by losing while outnumbering heavily.

Link to comment
Share on other sites

@Frostball.9108 said:

@Ben Phongluangtham.1065 said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

Im curious about what the average rating difference in a match is when there is a legend tier player in it? This season or last season or whatever

The sample of games with a legend player is pretty low, so I extended it down to 1700 (plat 3+). Here's the data including last season and this season:Average skill rating difference between teams: 14.16Average standard deviation difference between teams: 13.55Average rating difference in a match: 189.71Percent of games with average skill rating difference >50: 0.4%

There's a significant increase in rating range at this level, but as seen the skill rating mean and standard deviation differences between teams are pretty similar. The rating range is always going to be higher at the edges of the rating curve as a tradeoff with keeping reasonable queue times , but it doesn't stop the matcher from making fair teams which is the most important.

As some others have pointed out, the average game doesn't always tell the whole story. So I also grabbed all matches from last and this season and paired it down to the set of games with a rating range over 200 (these account for about 10% of all matches). Here's the data for that set:Average skill rating difference between teams: 19.83Average standard deviation difference between teams: 33.67Average rating difference in a match: 279.42Percent of games with average skill rating difference >50: 7.5%

While the skill rating and standard deviation differences are a bit higher in this set, these numbers aren't too bad overall.

I'm not trying to say that the matcher is perfect, but the vast majority of games are pretty balanced.

Thanks for the response, i was genuinely curious. I dont doubt the balance of the games. The only problem i really have with the games is that sometimes your influence in the game feels meaningless since regardless of how well you do your team can still kitten you over by losing while outnumbering heavily.

Yeah, I think (as in pure hypothetical) the main problem isn't with the matchmaker, it's with the fact that you can considerably screw over your team with very small mistakes. And there's no way to carry your team out of that by compensating by playing better. The problem with PvP in GW2 is more that carrying pretty much isn't possible and playing at your best is very possible. Which means you can't compensate those mistakes that will inevitably happen.

As in in one wrong movement and you can put your whole team in a bad position. vs as one player you can't really undo or even compensate that mistake. All you can do is try to play best as a team, and never giving up on trying to play the best possible way in the hopes the other team makes a wrong move instead.

Link to comment
Share on other sites

@rank eleven monk.9502 said:Have you actually read the whole post or you just got instantly triggered (like many others) by seeing a high rating difference?When you have such big differences in rating it is very natural for unbalanced games to occur, because it is not 1v1 or even 2v2 but 5v5. An experienced player will almost always identify the weak link of the enemy and perma having him out of the game resulting in a 4v5 at best - if your team happens to have more than one of these 'low rating players' it could become even 3v5 and so on. It's not rocket science, such big differences are not healthy for the game balance.

Link to comment
Share on other sites

@"Exatherion.3075" said:I've been playing now for about 20 ranked matches and lost 17(!). While you could always argue about my personal skill etc. It's simply impossible that in 17 matches the other team can be constantly better. I've a friend with the same problem. She won last season easily and now had no chance at all. When we joined the queue together we always had someone afk or playing Leeroy Jenkins. I don't wanna say that I'm a very good player like a friend..... But still I should've won statistically at least one game in the last 17 matches. I won the 3 first matches when I couldnt even play my profession good as now. Then I lose 17(?).

It's my impression that because of the fact that we just start "now" in this season - that we get paired up with people that dont even try to read. Everytime I try to talk about strategy or other things there's simply no response.

Now that's nothing that Anet can fix concerning not talking players; though it would be nice if Anet could fix the algorityhm. I've read in the past in the forum that Anet employee's can see the amount of matches I've lost on my account.

Me and my Guildmate don't seem to be the only ones with that problem. I think it has to do with the fact that we started later this season and now somehow get paired up with people that go afk don't talk etc. What I mean: Due to the advanced time in the season we've worst starting conditions.

I feel you x100. It's frustrating man... I have had many many games where I am explaining what people should do and they completely ignore my requests and get steamrolled somewhere. So many people just throw games on purpose and because of that, you end up losing. Since no one else cares about winning and I lose all the time, I am starting to feel the same way and I don't care as much anymore. I don't even try to tell people what to do because they literally don't listen. Why even try in a match if many of your teammates just die over and over and don't even run away from fights they cannot win. At this point, because of the low population and the thrown matches, I am like everyone else who doesn't care and I just go for the rewards and don't try as hard as I used too. Sooooo many games I play teams who actually have teamwork and the team I am on is like a bunch of chickens with their head cut off.

It's like people don't understand how to play spvp at all... no rotations, no teamwork, afkers, people who dont care and throw games on purpose, and not to mention.... This Necro/Mesmer meta this season has ruined it for me and hundreds of other people playing ranked. It's not fair..... Why did ANET think creating the Scourge with massive AOES in a point capture game was fair? They are tanky, do insane damage, steal life from you, and also provide barriers or extra health to nearby teammates... like wtf?

I have hope for the Alliance WvW patch and after this season and getting my ascended gear in ranked spvp I have no interest in playing much of season 12. Oh and the toxcity... hahahaahah that's another thing that is horrible in ranked pvp. Heck in many games a person was literally yelling at people and saying stupid stuff to teammates and we won the game!!! Like wow bro... I can't be as good as you.... Bow to your godlike gaming abilities oh mighty one

But this is the issue: People don't care anymore. ANET needs to create a tutorial of some sort for SPvP to explain proper rotations and how the mode works in general with everything. A wall of text won't work because people don't want to read it. You need to make a video for new people or anyone to watch so people understand the gamemode.

Link to comment
Share on other sites

@Frostball.9108 said:

@"Ben Phongluangtham.1065" said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

Im curious about what the average rating difference in a match is when there is a legend tier player in it? This season or last season or whatever

The sample of games with a legend player is pretty low, so I extended it down to 1700 (plat 3+). Here's the data including last season and this season:Average skill rating difference between teams: 14.16Average standard deviation difference between teams: 13.55Average rating difference in a match: 189.71Percent of games with average skill rating difference >50: 0.4%

There's a significant increase in rating range at this level, but as seen the skill rating mean and standard deviation differences between teams are pretty similar. The rating range is always going to be higher at the edges of the rating curve as a tradeoff with keeping reasonable queue times , but it doesn't stop the matcher from making fair teams which is the most important.

As some others have pointed out, the average game doesn't always tell the whole story. So I also grabbed all matches from last and this season and paired it down to the set of games with a rating range over 200 (these account for about 10% of all matches). Here's the data for that set:Average skill rating difference between teams: 19.83Average standard deviation difference between teams: 33.67Average rating difference in a match: 279.42Percent of games with average skill rating difference >50: 7.5%

While the skill rating and standard deviation differences are a bit higher in this set, these numbers aren't too bad overall.

I'm not trying to say that the matcher is perfect, but the vast majority of games are pretty balanced.

Thanks for the response, i was genuinely curious. I dont doubt the balance of the games. The only problem i really have with the games is that sometimes your influence in the game feels meaningless since regardless of how well you do your team can still kitten you over by losing while outnumbering heavily.

The problem is the "carry" philosophy of the game. Dev think that a good or very good player can carry a group of numerous (2 ot 3 ) much lower players.So an average team of 1420 with (1700 + 1500 + 1300 + 1300 + 1300) can face a 1440 with (1500 + 1500 + 1400 + 1400 + 1400).

But not, the average lvl of team 2 will dominate most encounters against team 1, even if the 1700 player can kill any player of team 2 in 1v1. So it ends that team 2 will clean team 1 and then will jump at 3 v 1 vs the 1700 player of team 1.

But in the Anet stats, 1420 is ok vs a 1440. Systrem is working fine. Just 20 point of difference in this match. It's OK.

Link to comment
Share on other sites

in my modest opinion, this is not a fault of system being rigged, the fault is the system is not rigged enought. let me explain:

in this game mode not all classes roles have the same impact on game. imagine this escenario:

-2 teams of average 1500, all two teams have the same composition 1600, 1500, 1500, 1500, 1400-mirror comps-supose the mmr is acurate to "real skill"-team A their 1600 is a roamer thief and their 1400 a suport firebrand team B their 1400 is the roamer thief and 1600 their suport firebrand

A thief will destroy B thief in the 90% of fights making team A dominant in rotations/side objectives and having some time to +1 in bigger fights, the impact of superior skills of suporting character are less than the impact of superior skills of the roamer and this little advantage on big figths will be nullified by +1 .

this macht will end with a victory by a wide margin for team A

Link to comment
Share on other sites

@"Delweyn.1309" said:The problem is the "carry" philosophy of the game. Dev think that a good or very good player can carry a group of numerous (2 ot 3 ) much lower players.So an average team of 1420 with (1700 + 1500 + 1300 + 1300 + 1300) can face a 1440 with (1500 + 1500 + 1400 + 1400 + 1400).

But not, the average lvl of team 2 will dominate most encounters against team 1, even if the 1700 player can kill any player of team 2 in 1v1. So it ends that team 2 will clean team 1 and then will jump at 3 v 1 vs the 1700 player of team 1.

But in the Anet stats, 1420 is ok vs a 1440. Systrem is working fine. Just 20 point of difference in this match. It's OK.

Fine let's see how that works with Standard Deviation(S.D.):Team 1Mean: 1420S.D.: 144

Team 2:Mean: 1440S.D.: 48

That's nearly a difference of 100 opposed to the 13 on average and 39 in the higher tiers. But let's say you said 1600 or 1500 instead of the probably exaggerated 1700 order to make a point in team 1 (also to show how S.D. behaves when you're starting to make up random team ratings)

  • With 1600 instead of 1700:Mean: 1400S.D.: 120
  • With 1500 instead of 1700:Mean: 1380S.D.: 96

None of that matches up with what the devs have told us.

Link to comment
Share on other sites

@Dreddo.9865 said:

@Crinn.7864 said:The rating systems distributes across a curve with a mean of 1200. Gold starts at 1200, so a player that is placing Gold is merely slightly above average. Your rating placement is done with the exact same glicko2 algorithm that rates you after placements. Placements is merely a mechanic for hiding your rating change for the first 10 matches, the reason the first ten are hidden is because glicko2 has enormously high deviation in early matches, and devs don't want players flipping out over the rating swings during early matches.Thank you for quoting the rating algorithm but I still don't understand how (and if) individual skill and game performance are calculated when defining a player's rating.

It does not use individual performance. Glicko is a outcome based algorithm.

Link to comment
Share on other sites

@FrizzFreston.5290 said:

@"Delweyn.1309" said:The problem is the "carry" philosophy of the game. Dev think that a good or very good player can carry a group of numerous (2 ot 3 ) much lower players.So an average team of 1420 with (1700 + 1500 + 1300 + 1300 + 1300) can face a 1440 with (1500 + 1500 + 1400 + 1400 + 1400).

But not, the average lvl of team 2 will dominate most encounters against team 1, even if the 1700 player can kill any player of team 2 in 1v1. So it ends that team 2 will clean team 1 and then will jump at 3 v 1 vs the 1700 player of team 1.

But in the Anet stats, 1420 is ok vs a 1440. Systrem is working fine. Just 20 point of difference in this match. It's OK.

Fine let's see how that works with Standard Deviation(S.D.):Team 1Mean: 1420S.D.: 144

Team 2:Mean: 1440S.D.: 48

That's nearly a difference of 100 opposed to the 13 on average and 39 in the higher tiers. But let's say you said 1600 or 1500 instead of the probably exaggerated 1700 order to make a point in team 1 (also to show how S.D. behaves when you're starting to make up random team ratings)
  • With 1600 instead of 1700:Mean: 1400S.D.: 120
  • With 1500 instead of 1700:Mean: 1380S.D.: 96

None of that matches up with what the devs have told us.

Well, when you know that Sindrener got teamed with a silver player you know that I didn't exagerated the numbers.

And I would add that the numbers given are average. So if 90% of matches have a SD difference of -30, but 10% have SD difference of +100 then the global average number for SD would seems not so bad isn't it ?

Link to comment
Share on other sites

@FrizzFreston.5290 said:

@"Delweyn.1309" said:The problem is the "carry" philosophy of the game. Dev think that a good or very good player can carry a group of numerous (2 ot 3 ) much lower players.So an average team of 1420 with (1700 + 1500 + 1300 + 1300 + 1300) can face a 1440 with (1500 + 1500 + 1400 + 1400 + 1400).

But not, the average lvl of team 2 will dominate most encounters against team 1, even if the 1700 player can kill any player of team 2 in 1v1. So it ends that team 2 will clean team 1 and then will jump at 3 v 1 vs the 1700 player of team 1.

But in the Anet stats, 1420 is ok vs a 1440. Systrem is working fine. Just 20 point of difference in this match. It's OK.

Fine let's see how that works with Standard Deviation(S.D.):Team 1Mean: 1420S.D.: 144

Team 2:Mean: 1440S.D.: 48

That's nearly a difference of 100 opposed to the 13 on average and 39 in the higher tiers. But let's say you said 1600 or 1500 instead of the probably exaggerated 1700 order to make a point in team 1 (also to show how S.D. behaves when you're starting to make up random team ratings)
  • With 1600 instead of 1700:Mean: 1400S.D.: 120
  • With 1500 instead of 1700:Mean: 1380S.D.: 96

None of that matches up with what the devs have told us.

And that's why people are all over enabling ratings shown at the end of the match, because these kind of enormous deviation splits are happening very frequently or at least it feels like it, contrary to what the devs are saying. Again, when you play the top 250 and can see easily see where players you are facing or are with are rated, is when you start to notice something isn't quite right with what we're being told or maybe what people are assuming about the numbers we are being given.

Link to comment
Share on other sites

@ezd.6359 said:

@Ben Phongluangtham.1065 said:I asked for numbers for this season to be pulled since people are always interested:

Average skill rating difference between teams: 11.866Average standard deviation difference between teams: 11.643Average rating difference in a match: 98.203 (min rating vs max rating across all players in the match)

One thing to keep in mind that end score difference never means that the match didn't start off even. Scores tend to snowball in our game for a number of factors. Some due to map layout/mechanic design. Some due to human nature, as people tend to tilt or give up after getting behind by a certain number. Sometimes people play above or below their potential. That's just part of human performance.

So, if you like these numbers, why don't you show real numbers before/after match? If everything works as intended, it should not hurt anyone.Because he posted the average over thousands of games. Single games can be catastrophal. Like I am gold division and in enemy team was a God of PvP from r55 guild.We won and I got a lot more MMR than usual for a win, but that single god of pvp kept one shotting everyone with OP chrono build.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...