Coming up with a better way to rank players
PvP games will always have a personal score (points)… why wouldn’t they?
The fact that rank is not a measure of skill is irrelevant. Time spent playing is time spent playing. Whether Player A with Rank 29 has developed keen and functional coordinated team play strategies or not is irrelevant. That is experience you earn while playing your tourneys.
The point of Rank is to illustrate to structured teams that applicants are not completely fresh out of the box. So, as a measure of time spent playing, it is working perfectly.
As it stands, if a team of all r50 players went up against a team of all r50 players, every player having just hit 50, the game would definitively broadcast which of those players are better than the others through things like personal score, team score, etc.
If Joe Blow farms hotjoin to get rank 50 and shows up in a tourney, he is going to get trashed, but after several matches, he is going to get better because he has already put in the hours needed to gain a commendable command of his class.
Tactics can be taught later.
Conversely, John Doe who has done nothing but TPvP is likely to get hammered and frustrated with the chaos that is hotjoin until having played several matches.
I thought at one point about suggesting a different rank for hotjoin vs tpvp, but in the end I decided that this is a bad idea.
At the end of the day, a skilled player is skill regardless of what their rank shows, and i don’t think anyone will argue that.
Rank, at the very least, illustrates some level of effort and commitment to the game, which is positive for everyone.
…
As for the leaderboards … <heavy sigh> I will need to put my thoughts in to more friendly words to talk about that.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
I agree with you on most parts regarding rank,glory and leaderboard.
Glory points, current game mode and current leaderboard don’t have the means to measure ‘individual’ skill, and with the current pvp game mode more or less focusing on teamwork to win above all else the only real way to measure rank is your ‘teams rank’. I guess if you are in the no.1 team in the world you can say you are amongst the top 10 players in the world for whatever class you are playing.
The only way to know for sure if your the best team in the world is to participate in organised tournaments – hotjoins, solo and group queue wont cut it for numerous reasons. Its way more prestige anyway to say you are part of a team that has won tournaments then to say you were rank X but never participated in serious pvp tournaments. So hopefully there will be lots of organised tournaments happening coz this can only be a good thing for everyone concerned.
On the idea of developing some kind of algorithm to determine a players metrics, that kind of sounds like sports science to me so you might want to look in that field.
the more complicated and convoluted the algorithms, the less accurate the representations of numbers will be.
K.I.S.S.
If Anet instituted a team registry, much like the WoW registry, then your “team” could conceivably rank up and make tournament matching more appropriate because only registered teams could play against registered teams.
SoloQ can stay as it is, just because a premade queues up does not make them a registered team.
Therein, time spent playing (Rank) still accurately reflects time spent playing, and team ranking would be the determinant for matchmaking in tournaments.
they already have the infrastructure for separate queues, or else they wouldn’t have had 2 team vs 8 team tournaments. why they haven’t just extended this principle is beyond my exceptional reasoning skills.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
PvP games will always have a personal score (points)… why wouldn’t they?
The fact that rank is not a measure of skill is irrelevant. Time spent playing is time spent playing. Whether Player A with Rank 29 has developed keen and functional coordinated team play strategies or not is irrelevant. That is experience you earn while playing your tourneys.
The point of Rank is to illustrate to structured teams that applicants are not completely fresh out of the box. So, as a measure of time spent playing, it is working perfectly.
As it stands, if a team of all r50 players went up against a team of all r50 players, every player having just hit 50, the game would definitively broadcast which of those players are better than the others through things like personal score, team score, etc.
If Joe Blow farms hotjoin to get rank 50 and shows up in a tourney, he is going to get trashed, but after several matches, he is going to get better because he has already put in the hours needed to gain a commendable command of his class.
Tactics can be taught later.
Conversely, John Doe who has done nothing but TPvP is likely to get hammered and frustrated with the chaos that is hotjoin until having played several matches.
I thought at one point about suggesting a different rank for hotjoin vs tpvp, but in the end I decided that this is a bad idea.
At the end of the day, a skilled player is skill regardless of what their rank shows, and i don’t think anyone will argue that.
Rank, at the very least, illustrates some level of effort and commitment to the game, which is positive for everyone.
…
As for the leaderboards … <heavy sigh> I will need to put my thoughts in to more friendly words to talk about that.
You completely missed the point of my entire post and then decided to add some snide remarks to display your clearly superior intelligence.
For one thing, I never said that we should get rid of PvP game points. How you managed to interpret that just… Blows my mind. My entire post is a plea to the community to find an algorithm to better rank players, not to get rid of PvP points. The two aren’t even remotely related.
The fact that rank is not a measure of skill is irrelevant… (irrelevant kitten)…So, as a measure of time spent playing, it is working perfectly.
But you’re still missing the point; I’m not looking for a measure of time spent, I’m looking for a measure of skill. Unfortunately, many people misconstrue rank on this forum to mean the same thing as skill, which is why I included rank as a faulty measure of skill in my OP.
As it stands, if a team of all r50 players went up against a team of all r50 players, every player having just hit 50, the game would definitively broadcast which of those players are better than the others through things like personal score, team score, etc.
… Except that we already discussed that personal scores are meaningless, and that team scores only show how good a person’s team is, and, in a broadcasted game, the chances of that team being random is… Well, zero. Furthermore, team score, with such a small sample, would only rank how good a team is relative to another team, and even then that ranking wouldn’t be particularly good anyways. So you suffer from both sampling bias and sample size errors.
Still, none of this has anything to do with ranking skill, which was the entire idea behind my post.
If Joe Blow farms hotjoin to get rank 50 and shows up in a tourney, he is going to get trashed, but after several matches, he is going to get better because he has already put in the hours needed to gain a commendable command of his class.
Which goes to show that rank and thus experience are not good measures of skill. Perhaps another player, Doe Deer, comes along and reaches rank 20 just through tourneys. Now Doe has both command of her class as well as the tactical and strategic knowledge required to be considered, by many, a “good player”, even though she has much less overall experience than Joe, who farmed hotjoin and didn’t do any tournies and thus has lacking tactical and strategic. In fact, Joe’s experiences may actually hurt him in the long run, because his farming hotjoin may have created some irreversible habits that are detrimental to his playstyle that Doe doesn’t have.
Conversely, John Doe who has done nothing but TPvP is likely to get hammered and frustrated with the chaos that is hotjoin until having played several matches.
Except that most teams don’t care about how much of a hotjoin hero you are, just how good you are in tourneys.
At the end of the day, a skilled player is skill regardless of what their rank shows, and i don’t think anyone will argue that.
Thank you for repeating my post…
Rank, at the very least, illustrates some level of effort and commitment to the game, which is positive for everyone.
But again, this isn’t a measure of skill. It isn’t even a measure of effort or commitment, given that there are other game modes in GW2 as well as many powerful, external game skills, like theorycrafting, strategizing, and team formation.
As for the leaderboards … <heavy sigh> I will need to put my thoughts in to more friendly words to talk about that.
Are you actually going to address what I wrote in my post this time?
I agree with you on most parts regarding rank,glory and leaderboard.
Glory points, current game mode and current leaderboard don’t have the means to measure ‘individual’ skill, and with the current pvp game mode more or less focusing on teamwork to win above all else the only real way to measure rank is your ‘teams rank’. I guess if you are in the no.1 team in the world you can say you are amongst the top 10 players in the world for whatever class you are playing.
The only way to know for sure if your the best team in the world is to participate in organised tournaments – hotjoins, solo and group queue wont cut it for numerous reasons. Its way more prestige anyway to say you are part of a team that has won tournaments then to say you were rank X but never participated in serious pvp tournaments. So hopefully there will be lots of organised tournaments happening coz this can only be a good thing for everyone concerned.
On the idea of developing some kind of algorithm to determine a players metrics, that kind of sounds like sports science to me so you might want to look in that field.
It is certainly true that having a good team in that kind of situation is very useful in determining skill rank; however, the issue comes up that there are millions of players, and you can’t rank each of them by having them play in a tournament with a control team of four people against another control team of 5 people to determine how good the person is. Not only is that impractical, but over time, your constants would no longer remain constant. It would also be deathly boring.
Also, even being the best team in the world doesn’t mean that you’re the best possible team in the world, which is why I’d like to look into ranking players based on different areas of play, like bunkering, support, etc.
On the idea of developing some kind of algorithm to determine a players metrics, that kind of sounds like sports science to me so you might want to look in that field.
This is essentially what I’m looking for, actually. I would like to be able to find good correlative equations to determine how good a player is; I just need to find a good set of two variables to measure and then get the actual statistics on this.
what snide comments?
this is actually one of my threads lacking in condescension and snark.
all i did was support the purpose of Rank as a measure of time spent playing, stating that it is working as intended.
what is broken is the match making system, which is tied to the leaderboards, a subject that i need to collect my thoughts on more deeply.
wtf is with people taking everything personally?
what contribution did you expect? glowing reviews of your awesomeness at pinpointing one of the problems in the game?
good for you.
how’s that for snide?
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
kittenkittenkitten
… Y’know, okay then.
your quoting skills are lack luster.
let me reiterate for you:
rank will never display actual skill level. ever.
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
the ranking system as it is now, functions as a report of how much time that player has spent playing pvp.
Why is that a bad thing?
all you will ever have on any stat sheet is a relational representation of time spent doing something.
i wasn’t being facetious or trolling in either of my posts, i was discussing the issues you take with the ranking system.
how does that not address what you want to talk about?
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
your quoting skills are lack luster.
let me reiterate for you:
rank will never display actual skill level. ever.
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
the ranking system as it is now, functions as a report of how much time that player has spent playing pvp.
Why is that a bad thing?
all you will ever have on any stat sheet is a relational representation of time spent doing something.
i wasn’t being facetious or trolling in either of my posts, i was discussing the issues you take with the ranking system.
how does that not address what you want to talk about?
Uh…
rank will never display actual skill level. ever.
Did you really not read my original post? At all? Any of it? Or did you just see the words “skill level” and “rank” and think “omg he must think that skill=rank lol”
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
And on what basis do you make this claim? Can one’s skill not be greater than another’s? Lesser than? Can’t a person’s skill level grow, or decrease? Isn’t skill an objective factor, and not a subjective one? How can you possibly state that there is no way to give a numerical rating of a person’s skill level, particularly when you haven’t even researched it yet?
the ranking system as it is now, functions as a report of how much time that player has spent playing pvp.
Why is that a bad thing?
Give me a single instance where I said it was a “bad thing”.
I never did say that, did I? I said that it didn’t serve the purpose people were using it for.
all you will ever have on any stat sheet is a relational representation of time spent doing something.
You make the claims, you better bring up the evidence.
i wasn’t being facetious or trolling in either of my posts, i was discussing the issues you take with the ranking system.
how does that not address what you want to talk about?
Because you never addressed how my idea doesn’t work, just how ideas that I didn’t have are actually completely false.
i agree the system is useless. the issue i have is that it is now a part of the game and people can in some way consider it useful information.
in reality it is relevant information to anet not other players. since what is important to people who run a mmo is time played per person.
adding a /playedpvp would have = the same thing pretty much and they could have saved thier time and effort for bug fixing.
i agree the system is useless. the issue i have is that it is now a part of the game and people can in some way consider it useful information.
in reality it is relevant information to anet not other players. since what is important to people who run a mmo is time played per person.
adding a /playedpvp would have = the same thing pretty much and they could have saved thier time and effort for bug fixing.
But if I’m looking to add a new Guard to my team, it’d be helpful to know which ones are the best that are free at any given moment.
Did you really not read my original post? At all? Any of it? Or did you just see the words “skill level” and “rank” and think “omg he must think that skill=rank lol”
your leap of logic is midichlorian.
of course i read your post, you want a ranking system that reflects skills. not that rank = skill.
i never said rank = skill, nor did i say that that is what you said.
i said that rank != skill.
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
And on what basis do you make this claim? Can one’s skill not be greater than another’s? Lesser than? Can’t a person’s skill level grow, or decrease? Isn’t skill an objective factor, and not a subjective one? How can you possibly state that there is no way to give a numerical rating of a person’s skill level, particularly when you haven’t even researched it yet?
because any numerical representation is a stripped down point of reference that divorces itself from context. this is why personal in-game score is, at the end of the day, irrelevant in a team match.
A Bunker with 0 points on a winning team did his job. How does 0 reflect skill? It doesn’t. It never will.
Conversely, a Roamer with 300 points on a losing team probably didn’t do his job. How does his 300 points reflect skill? It doesn’t. It never will.
the ranking system as it is now, functions as a report of how much time that player has spent playing pvp.
Why is that a bad thing?
Give me a single instance where I said it was a “bad thing”.
I never did say that, did I? I said that it didn’t serve the purpose people were using it for.
And my point is that people are expecting the wrong things from Rank. IE: You’re doing it wrong.
all you will ever have on any stat sheet is a relational representation of time spent doing something.
You make the claims, you better bring up the evidence.
Such as? What evidence is needed for a statement like that? Statistics are relational. That’s just what they are, a relation. They are valueless until manipulated and presented in a meaningful way. (MANIPULATE IS A VERB, NOT A NEGATIVE ACTION so don’t bother with some irrelevant argument about how stats aren’t manipulated. For example, tallying a RBI ratio is, technically, manipulation of statistics for meaningful presentation. Runs batted in over a period of time.) Statistics lack context.
i wasn’t being facetious or trolling in either of my posts, i was discussing the issues you take with the ranking system.
how does that not address what you want to talk about?
Because you never addressed how my idea doesn’t work, just how ideas that I didn’t have are actually completely false.
I never said your ideas don’t work. Not once. All I ever said is that rank, as it is, is working for what it is intended to do, and that the leaderboards and match making are borked.
you are the one jumping down my throat here, not the other way around.
All I am saying is that Rank will never be a measure of skill, because skill is immeasurable.
I never said “kitten you, stupid head, more stats is kitten bleeeheheh”
(this is me being facetious)
srsly.
more stats would be great, but you cannot expect them to measure skill.
even the reading of statistics is a subjective experience, because you have to reinterpret them, and everyone interprets differently, as this thread clearly showcases.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
(edited by nakoda.4213)
i agree the system is useless. the issue i have is that it is now a part of the game and people can in some way consider it useful information.
in reality it is relevant information to anet not other players. since what is important to people who run a mmo is time played per person.
adding a /playedpvp would have = the same thing pretty much and they could have saved thier time and effort for bug fixing.
But if I’m looking to add a new Guard to my team, it’d be helpful to know which ones are the best that are free at any given moment.
No matter what the statistical system, the only way to ever judge that appropriately is to watch that player play, and/or play with that player.
what do you think professional scouts are for in RL sports?
They start with the stats (rank) and then observe. Then they interpret the stats based upon their CONTEXTUAL OBSERVATION of reality. Then they decide to put the player on the list or not.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
Did you really not read my original post? At all? Any of it? Or did you just see the words “skill level” and “rank” and think “omg he must think that skill=rank lol”
your leap of logic is midichlorian.
of course i read your post, you want a ranking system that reflects skills. not that rank = skill.
i never said rank = skill, nor did i say that that is what you said.
i said that rank != skill.
So if you knew that that wasn’t what I said, why for the love of reason would you even tell me that rank != skill in your first comment? It was completely redundant and nonsensical if you did realize this. In fact, most of everything that you wrote was redundant if the whole time you knew that I had already explicitly stated it.
You try to make a case against statistics, and, while interesting, it fails on a few very basic level. For one thing, you state that “statistics lack context without manipulation” (for all intents and purposes), but you use “manipulate” as if it’s some kind of taboo word (and, of course, we’re speaking of the alternate definition of manipulate: "Handle or control (a tool, mechanism, etc.), typically in a skillful manner: “he manipulated the dials”, just from google). Why should manipulation of statistics, then, be considered a bad thing? This you do not explain; you simply state it as your reason why this project fails. Now, if you really are using the other definition of manipulation (to change, essentially), then we’re talking about two entirely different things. You’re talking about statistics being faulty in that they have to be altered in order to exist (which means that statistics, by definition, are all false because they must express facts in false manners), while I’m discussing statistics as being facts as long as they are unbiased estimators of a true value. Your definition automatically assumes an incorrect nature of all statistics, so I’m extremely doubtful that your definition is true, unless you’re about to consider everything in life a lie (given that everything that we see is only an estimation of what is true, as there are no objective truths, simply subjective ones based on previous assumptions- ironically, we can deduce this only via previous assumptions).
There is a further, more serious issue with your argument. We already showed that manipulation, in the context of something simply being “used”, doesn’t necessarily make something inaccurate. However, you also suffer from the simple fact that there is nothing that doesn’t require previously defined context. You stated that “statistics lack context”, but that’s only in the situation that none is given to them. In one of your examples, you mentioned RBI, so let’s look at that. So, for a statistic about RBI to mean anything, it has to be stated that the statistic is given in the context of it being an RBI measurement. Well, what is RBI? It stands for “run batted in”, which refers to baseball, which refers to the human game of… You get the idea. Thus, stating that statistics “lack context” is an issue that is detrimental to their accuracy (which, I presume, is the entire reason why you’re denying that they can be used to evaluate player skill levels in the first place) is no different from stating that nothing is accurate and thus no measurement of anything anywhere can be used to determine anything, because everything requires context.
And you did say that my ideas don’t work.
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
So my ideas being inaccurate isn’t the same as them being unable to work? I don’t think so; why measure something when you explicitly know that not only your statistic may be inaccurate, but actually uses a nonexistent scale to rank something?
You don’t even know what you’re saying.
more stats would be great, but you cannot expect them to measure skill.
even the reading of statistics is a subjective experience, because you have to reinterpret them, and everyone interprets differently, as this thread clearly showcases.
Does this change for anything? Everything is subjective. Yet, is subjectivity a bad thing? Not necessarily.
Furthermore, objects that can only exist as intangibles, such as numbers, can’t be misinterpreted if they are, by definition, objects of thought. Furthermore, numbers can only be interpreted one way as they are a measure of value relative to nothingness, which is the absence of something in a given context, and there is only one rational way to interpret nothingness- as an absence of existence, a term used to describe something that our sensory receptors can in some way experience.
i agree the system is useless. the issue i have is that it is now a part of the game and people can in some way consider it useful information.
in reality it is relevant information to anet not other players. since what is important to people who run a mmo is time played per person.
adding a /playedpvp would have = the same thing pretty much and they could have saved thier time and effort for bug fixing.
But if I’m looking to add a new Guard to my team, it’d be helpful to know which ones are the best that are free at any given moment.
No matter what the statistical system, the only way to ever judge that appropriately is to watch that player play, and/or play with that player.
what do you think professional scouts are for in RL sports?
They start with the stats (rank) and then observe. Then they interpret the stats based upon their CONTEXTUAL OBSERVATION of reality. Then they decide to put the player on the list or not.
So putting in scouts into a sport determines whether or not an unit of measurement is accurate? If ANet decided to replace the visibility of numbers like “power”, “precision”, or “health” and replaced them with subjective relative terms such as “powerful” or “weak”, would that mean that number cannot exist? Of course not, it just means that we don’t know what a particular numerical value is without researching it. Furthermore, watched “Moneyball”?
Also, your second part again assumes that there is a subjective value of numbers. In the situation that someone has to evaluate whether a player is “strong” or “weak”, I might agree with you if the categorical value is not based on a range of numbers (i.e. you define 1-10 as weak, 11-20 as mediocre, and 21-30 as strong), but instead based on personal influences such as bias (“oh, I can’t recall the last time I saw a twenty, this guy must be strong”), but this does not have to be the case. If we define strength as being relative to other players (which, by definition, it has to be, you can’t say that one player is stronger than the other without using relational terms between them), then we can simply run a numerical analysis on the skill of players based on the regression equations that we determine from our statistics that I mentioned in my response to Psycho. Thus, we get a numerical rank for players based on observed statistical facts that helped us to create both a regression equation as well as determine the value of some variable for a given player.
i clearly said that manipulate was NOT meant to be seen as a taboo thing, it is just a verb: to manipulate.
with that in mind, i didn’t bother reading the rest of what you said, because you haven’t bothered to read what i’ve said.
discussion over. good bye.
It’s funny that you say this, because then you went on to describe how apparently by being manipulated the statistic is inaccurate.
i clearly said that manipulate was NOT meant to be seen as a taboo thing, it is just a verb: to manipulate.
with that in mind, i didn’t bother reading the rest of what you said, because you haven’t bothered to read what i’ve said.
discussion over. good bye.
It’s funny that you say this, because then you went on to describe how apparently by being manipulated the statistic is inaccurate.
and where did i say that?
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
People higher ranked in the leaderboards are definitely better players. People who solo queue do not get into the top 1,000 without being better than most players. People who queue in teams do not get to top 100 without being better than most players. Obviously, there’s a difference in threshold between those who solo queue and those who queue in groups, but leaderboards do properly gauge consistently better performances in the long term.
(edited by Lopez.7369)
People higher ranked in the leaderboards are definitely better players. People who solo queue do not get into the top 1,000 without being better than most players. People who queue in teams do not get to top 100 without being better than most players. Obviously, there’s a difference in threshold between those who solo queue and those who queue in groups, but leaderboards do properly gauge consistently better performances in the long term.
Yes and no.
I see where you’re coming from, and I understand that some level of skill must factor in here in order to make a player find themselves in the top 1000 whilst solo queuing. Here’s the thing though.
This says absolutely nothing about play style or areas of strength and weakness. For example, perhaps I run a direct boon support build, but my allies don’t know how to accommodate me on the team. So, I lose match after match after match, even though my build is theoretically correct, and I could potentially be one of the strongest team players in the game. The issue is simply that my build doesn’t fit with others’ playstyles.
Or maybe I’m a thief with a GC build. Perhaps I got into the top 1000 because the majority of players don’t know how to counter my thief’s GC build, in spite of its theoretical inaccuracy. However, my build doesn’t require any special team cooperation, and I’m able to get very highly ranked because most tourney players aren’t pros, they’re just casual players.
So, yes, the leaderboard is some indicator of skill. Certainly, if I was evaluating players one by one, I would check the Leaderboard before I checked anywhere else. However, there are so many things that the leaderboard doesn’t factor into its measurements (other factors might be that a player buckles under pressure in a streamed tourney, is too command-oriented to be able to cooperate with players successfully, trying to force them to do things they have absolutely no desire to do; having a lack of build knowledge, having an inability to formulate good builds to fit the situation, being unable to work well over team chat, not being used to specific situations, like being on an all-thief team, etc.), and the leaderboard doesn’t evaluate certain skill areas, so I would suggest a different ranking system be created for this purpose.
i clearly said that manipulate was NOT meant to be seen as a taboo thing, it is just a verb: to manipulate.
with that in mind, i didn’t bother reading the rest of what you said, because you haven’t bothered to read what i’ve said.
discussion over. good bye.
It’s funny that you say this, because then you went on to describe how apparently by being manipulated the statistic is inaccurate.
and where did i say that?
no numerical tally system will ever accurately reflect skill, because skill is not numerical.
and part of your reasoning was that
all you will ever have on any stat sheet is a relational representation of time spent doing something.
To which, when asked for evidence, you replied that
You make the claims, you better bring up the evidence.
Such as? What evidence is needed for a statement like that? Statistics are relational. That’s just what they are, a relation. They are valueless until manipulated and presented in a meaningful way. (MANIPULATE IS A VERB, NOT A NEGATIVE ACTION so don’t bother with some irrelevant argument about how stats aren’t manipulated. For example, tallying a RBI ratio is, technically, manipulation of statistics for meaningful presentation. Runs batted in over a period of time.) Statistics lack context.
So, again, you are faced with a decision: choose between showing yourself to be redundant, misleading, and irrelevant to the conversation, or show yourself to be a lying. Either way, you said what you said, and it can only be interpreted one way.
(edited by Arganthium.5638)
that does not say that the manipulated stat is inaccurate.
i t says that statistics != skill. which is what we are debating here.
reading comprehension ftw.
and, again, statistics are a meaningless numerical relationships without manipulation to put them in to context and give them meaning.
this fact has not changed. there is an entire branch of science devoted to this very practice. it is called … statistics.
statistics mean nothing until you interpret them. they are just relational numbers.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
(edited by nakoda.4213)
You want an improvement over the current system? Do it by popular vote.
I fight people, sometimes they kill me, sometimes I kill them. Fun is had by all. [/quote]
what snide comments?
this is actually one of my threads lacking in condescension and snark.
all i did was support the purpose of Rank as a measure of time spent playing, stating that it is working as intended.
what is broken is the match making system, which is tied to the leaderboards, a subject that i need to collect my thoughts on more deeply.
wtf is with people taking everything personally?
what contribution did you expect? glowing reviews of your awesomeness at pinpointing one of the problems in the game?
good for you.
how’s that for snide?
Love, lol.
Ranks and points are fine, rank reflects time played, not how well that time was played or good you are at tpvp, but that’s what leaderboard is for. Points during matches don’t mean anything, and only noobs think they do, you don’t get any real advantage with things you buy with glory in this game so it’s just not a big deal. The only real problem atm is that the leaderboards between solo and premade queues aren’t split and the leaderboard does not have a proper decay feature. Those two things are elementary and need to be fixed before the leaderboards really reflects the players and teams, but adding a bunch of complex system measure individual players? Terrible use of development time.
Ranks and points are fine, rank reflects time played, not how well that time was played or good you are at tpvp, but that’s what leaderboard is for. Points during matches don’t mean anything, and only noobs think they do, you don’t get any real advantage with things you buy with glory in this game so it’s just not a big deal. The only real problem atm is that the leaderboards between solo and premade queues aren’t split and the leaderboard does not have a proper decay feature. Those two things are elementary and need to be fixed before the leaderboards really reflects the players and teams, but adding a bunch of complex system measure individual players? Terrible use of development time.
Joy, another one of these.
Did you not read my post? I know what rank and points reflect, and I explicitly stated that they do not reflect skill. Or are you being redundant?
Then you completely missed out on the point behind half of my leaderboard discussion: the idea is to rate players based on different areas of skill, not just an overall rank. And, again, as I explained in my OP and in one of my later comments, there are issues with using the leaderboard anyways. But I’m not sure I should even have to requote those.
You finish by asking why ANet would waste development time on this algorithm, but you fail to recognize that the point behind my post wasn’t to convince ANet to implement this system, it was for players to create the ranking system.
that does not say that the manipulated stat is inaccurate.
i t says that statistics != skill. which is what we are debating here.
reading comprehension ftw.
and, again, statistics are a meaningless numerical relationships without manipulation to put them in to context and give them meaning.
this fact has not changed. there is an entire branch of science devoted to this very practice. it is called … statistics.
statistics mean nothing until you interpret them. they are just relational numbers.
To which I asked you to bring up evidence why skill cannot be numerically represented, and you replied that statistics lack context and that they are manipulated as “evidence”. Now you’re just saying that manipulation doesn’t equal inaccuracy, but come back to your original argument of “well, stats can’t be represented numerically”, to which you used the aforementioned “evidence”. Circular logic?
I went into great depth about “statistics lacking context” later as well, but apparently you decided to skip my entire post. This is getting ridiculous; you still haven’t replied to any part of either of my arguments at all, and now you decide to make snide remarks like
this fact has not changed. there is an entire branch of science devoted to this very practice. it is called … statistics.
In spite of the fact that you never read what I actually wrote. Seriously?
(edited by Arganthium.5638)
please, oh great Arga, show me how you know that I never read what you wrote.
please.
it strikes me as funny that someone who agrees with me is “one of those”. did they not read what you wrote either?
ell oh ell.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
please, oh great Arga, show me how you know that I never read what you wrote.
please.
it strikes me as funny that someone who agrees with me is “one of those”. did they not read what you wrote either?
ell oh ell.
LOL… Dude, you deleted your own post. Cos honesty. Fortunately, I later quoted it, and here’s what you said:
i clearly said that manipulate was NOT meant to be seen as a taboo thing, it is just a verb: to manipulate.
with that in mind, i didn’t bother reading the rest of what you said, because you haven’t bothered to read what i’ve said.
discussion over. good bye.
Which was ridiculous, you had absolutely no idea what I was saying in the context of what I read, but you said it anyways. Now you delete your own post because you can’t face up to your own dishonesty.
You explicitly stated that you didn’t read what I said.
Furthermore, your arguments have been absolutely ridiculous. Your only actual relevant argument has been that statistics cannot numerically measure skill. I asked you why, and you replied by saying something obscure about them lacking context and having been “manipulated”. I then wrote an entire post in response to this, and you suddenly said that you weren’t going to bother reading what I wrote- a post which you later deleted. This is getting hilarious. Please, show us how much more kitten dishonesty you can pull off while still holding onto some remnant of dignity.
what is obscure about saying that statistics are meaningless without context, and that statistical manipulation is how they are made meaningful?
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
what is obscure about saying that statistics are meaningless without context, and that statistical manipulation is how they are made meaningful?
Why should I reply to this when you refuse to reply to almost all of the other things that I say? Maybe if you, y’know, read my post, and actually understood that everything lacks context until given one, and that there’s absolutely nothing about statistical manipulation (given the definition we’ve been using) that harms the accuracy of the statistic, making your mention of the subject completely irrelevant… Maybe then I’ll consider whether your questions actually have any value to them.
/facepalm.
thid is a placeholder until I get home and have a keyboard.
Benn E Violence :: 0/20/30/20/0
You kittens don’t even know what the prefix “meta” means.
I do remember a few months ago the devs were talking about expanding the end of game stats. It would be really cool if we could get some Sabermetric type stats like they use in baseball/basketball to more accurately define how much impact a certain player has on a game to game basis.
The rank-table would’ve been fine, but ye… they had to implement hotjoin to kitten up the whole system. GG
Indeed. The issue is multiple-fold – not only can you acquire rank from a very chaotic and zerg-promoting arena – you can also do it far better, in a much more timely fashion, and for a lot less effort.
This is why using rank in GW2 as an indication of anything but play time is silly. As a matter of fact, even that is questionable, because some proffessions can farm ranks really fast simply by dropping aoe on zergs.
In GW1, ranks were hard to obtain, and even when some ranks were shared by multiple arenas (e.g. RA and TA), the amount of rank points you could obtain from unorganized arena was limited: RA could get you a max of 1 point per 10 wins (afterwards the teamw as moved to TA), whereas in TA you would continue the streak until defeated (and a rank point was again granted ONCE every TEN wins). That did not only reflect play time, but overall team success, which depended largely on individual team member skill as well as overall team skill.
The gvg title used to be even harder to obtain, until it was watered down with the rest.
Here’s a trivia from GW wiki to indicate just how hard it used to be to obtain it…
CHAMPION
~ When this title was first introduced, the only victories that were considered for this title were those between guilds that had a minimum rating of 1,500. At that time, the only guilds meeting the rating requirement were those in the top fifteen or so of the entire guild ladder. The rating requirement was reduced to 1,200 in a later update, opening to the doors to players of lower-ranked guilds to start earning Champion victories. The February 22, 2013 update further reduced it to 1,050.
~The prefix for each tier’s title is the same as those for the Gladiator title track.
I don’t think it gets any more prestige than that, and it was insane. But it appears anet took a 180 degree turn on their past approach to titles, and instead chose to trivialize them alltogether, much like they trivialized grouping and guilds in general.
As to what system could give more precise data regarding player skill…it’s a tough nut to crack, especially because you seem to desire individual player rankings.
How can AI analyze matches and determine who is the carry and who the carried? What about the teams in which members roughly share the same level of skill? Can you base ranking on consistently winning all 1v1s? Or all team matches? Or consistently being on the right place at the right time? How could a program analyze such intricate data?
GW1 ladder back in the day was fairly accurate assessment of a guild’s competence, and the game also offered a fairly precise way to determine player skill in correlation to (past) ranks. Obviously it wasn’t perfect, and many other things influenced rank acquisition, but it was far better than what we have now. By mixing in hotjoin with the rest, as Emp already pointed out, GW2 has indeed spoiled it all.
I believe it is the community, the reputation (e-fame!) that usually give you the best clue when it comes to determining who’s any good, and who’s just getting carried. It has its own deficiencies, and isn’t always quite precise, but neither is the current system.
(edited by KarlaGrey.5903)
just say it directly… they messed it up with whole rank, glory and reward system, thats truth and none can say anything else…
and i wont repeate again what they should do to make it better, i gave a lot of suggestions in many threads and even made some new threads, none of them got any dev answer about ranks, glory and reward systems…
the only good thing is leaderboard with its own rating… anyway, solo vs team rating is a very subjective topic, i prefer solo, because i dont like to being fixed to 4 other people and being unable to play if 1 of them is not ready… and no, having more then 5 people in team isnt solution, none wants to play only when someone else is missing and be fifth wheel by a car… or even 2 people can be missing and we have the same problem again, 3 people still wants to play and they have friends to play with, but not in their team, what now? having even sixth wheel by a car? no…
(edited by MaXi.3642)
just say it directly… they messed it up with whole rank, glory and reward system, thats truth and none can say anything else…
and i wont repeate again what they should do to make it better, i gave a lot of suggestions in many threads and even made some new threads, none of them got any dev answer about ranks, glory and reward systems…
the only good thing is leaderboard with its own rating… anyway, solo vs team rating is a very subjective topic, i prefer solo, because i dont like to being fixed to 4 other people and being unable to play if 1 of them is not ready… and no, having more then 5 people in team isnt solution, none wants to play only when someone else is missing and be fifth wheel by a car… or even 2 people can be missing and we have the same problem again, 3 people still wants to play and they have friends to play with, but not in their team, what now? having even sixth wheel by a car? no…
The problem with individual ratings and not splitted solo/team-ladder is that good soloQ players are not rewarded for being good. On the other side you explained it perfectly well, with saying you don’t want to be bound to 4 other players.
In my opinion the leaderboards are completely useless. They were fancy the first weeks, but by now not accurate in any way. I do wanna soloq sometimes and being punished for that is just . . .I have no words for this. So the leaderboards failed to be an indicator of skill/being good. Again A-Net completely messed it up, by implementing a ladder into an incomplete system.
Ranks and points are fine, rank reflects time played, not how well that time was played or good you are at tpvp, but that’s what leaderboard is for. Points during matches don’t mean anything, and only noobs think they do, you don’t get any real advantage with things you buy with glory in this game so it’s just not a big deal. The only real problem atm is that the leaderboards between solo and premade queues aren’t split and the leaderboard does not have a proper decay feature. Those two things are elementary and need to be fixed before the leaderboards really reflects the players and teams, but adding a bunch of complex system measure individual players? Terrible use of development time.
Joy, another one of these.
Did you not read my post? I know what rank and points reflect, and I explicitly stated that they do not reflect skill. Or are you being redundant?
Then you completely missed out on the point behind half of my leaderboard discussion: the idea is to rate players based on different areas of skill, not just an overall rank. And, again, as I explained in my OP and in one of my later comments, there are issues with using the leaderboard anyways. But I’m not sure I should even have to requote those.
You finish by asking why ANet would waste development time on this algorithm, but you fail to recognize that the point behind my post wasn’t to convince ANet to implement this system, it was for players to create the ranking system.
You seem mad, my advice is to not make posts if you’ll get this upset when people disagree. Can’t be good for your health. I said rank is not a direct reflection of skill but is a reflection of time played which does matter and that they’re fine. I also said while points don’t reflect skill, that they are inconsequential and just for fun and therefore fine, which again, is a different take on it than what you presented. I also disagree that we need any additional rating system beyond the leaderboards, we just need the leaderboards to be improved, which is again, a different take on the issue than yours. I assumed you meant Anet would develop it because having players do it is such a terrible idea. Being that my post disagreed with all the main points of yours, I think you’re confused about what redundant means. It does not mean, “I don’t like your response therefore is redundant”. But it’s a big word, you probably just got it’s meaning mixed up. You can look it up here: http://dictionary.com/
(edited by Julius.1094)
it will shift to ranking Guilds/Teams again. A community leaderboards/ladder based on prized tourneys is the only way to go to make an honest representation on who are the top players/guilds in the game. Though the current leaderboards are quite representative of who are in the top currently (EU leaderboards has players from MiM and TP on the top, NA scene, i still have to see a tourney but good thing that Grouch, a reputable caster, is at the top too.)