Jump to content
  • Sign Up

Ranks Should Be Based On Skill, Not Luck


Noe.8032

Recommended Posts

At the end of the match; your reward, and the amount of rank you gain or lose should be based on your placement within your own team. First second third fourth fifth, with a plus or negative modifier put on based on whether your team won or lost. That way, the main deciding factor in the game is how did YOU do. Not how did your TEAM do.

It gets really boring, win streaking one day then lose streaking the next. All while using the same build, same playstyle, in the same rank range. You're playing just as well every match, doing the thing that makes you a contributing valuable member of your team yet sometimes you win ten times in a row and sometimes you lose the whole day and all your progress gets stolen away. Because it's literally all random, did you get put on the team with more or less good players. You are just one player, there are nine others in the match. No matter how good you are the outcome is determined by the sum performance of all ten players. The only way you could consistently influence the match in your favour against the entire performance of the nine other players in the match is if you're using a build that's so broken it literally makes you unstoppable. Which shouldn't be possible anyways.

In a game with nine other players in every match; your individual choices are not the game changer. There are nine other people individually influencing the outcome. If your team isn't good, you lose; if your enemy team isn't good; you win. Regardless of how good you are at the game, if your team is that bad or the enemy team is that good it will override anything you can do. Yeah sometimes you will make the right choices at the right times and that sways a balanced match into your favour but that's only with a somewhat balanced match; which do not happen very often, and only some of the time will you even have that opportunity. The logic that you winning or losing is based off your choices in a game of ten people is impossible, because by that logic that means the other nine people also can make the game win or lose by their choices. Which means you only won because your enemies all made worse choices then your allies. So the outcome was still influenced by the sum of ten players; even when using the ridiculous insecure logic that your individual choices entirely determine the outcome and thus if you lose oh it's all your fault and those nine other players had nothing to do with it.

Being good at the game literally doesn't matter because there's so many other players in the match that even if you're bad, if you get put on the better team they can still easily carry you. While if you're good, being put on the worse team doesn't matter because you physically can't carry all of them at once against a better enemy team.

At this point the ranking system might as well be grading who is the luckiest, abuses the most scummy class pairings with the duo que; and or uses the most hot meta that is so broken it literally can win you the entire match over the nine other players in it. Yes, some people who read this will just try to make up excuses to perpetuate the illusion that the one out of ten players in the match is the sole one influencing the outcome as long as that player is specifically them. Since if they accept that winning or losing is out of their hands then they can't claim credit for all their wins and generate ego from it.

Doesn't change the fact that this system is fundamentally based off coin flipping whether you get the winning or losing team regardless of whether you yourself are a good player or not.

And after a thousand a five hundred games that's my biggest complaint about this whole system, is you are rewarded based on the coin flip and not your own performance. I don't especially mind going on random, asinine losing streaks. It's the fact that the game steals days worth of rank progress from me while it simultaneously is the one putting me on that losing streak. Just makes the entire process feel totally pointless and miserable because it honestly just is.

Link to comment
Share on other sites

@Aisling.5901 said:At the end of the match; your reward, and the amount of rank you gain or lose should be based on your placement within your own team. First second third fourth fifth, with a plus or negative modifier put on based on whether your team won or lost. That way, the main deciding factor in the game is how did YOU do. Not how did your TEAM do.

It gets really boring, win streaking one day then lose streaking the next. All while using the same build, same playstyle, in the same rank range. You're playing just as well every match, doing the thing that makes you a contributing valuable member of your team yet sometimes you win ten times in a row and sometimes you lose the whole day and all your progress gets stolen away. Because it's literally all random, did you get put on the team with more or less good players. You are just one player, there are nine others in the match. No matter how good you are the outcome is determined by the sum performance of all ten players. The only way you could consistently influence the match in your favour against the entire performance of the nine other players in the match is if you're using a build that's so broken it literally makes you unstoppable. Which shouldn't be possible anyways.

In a game with nine other players in every match; your individual choices are not the game changer. There are nine other people individually influencing the outcome. If your team isn't good, you lose; if your enemy team isn't good; you win. Regardless of how good you are at the game, if your team is that bad or the enemy team is that good it will override anything you can do. Yeah sometimes you will make the right choices at the right times and that sways a balanced match into your favour but that's only with a somewhat balanced match; which do not happen very often, and only some of the time will you even have that opportunity. The logic that you winning or losing is based off your choices in a game of ten people is impossible, because by that logic that means the other nine people also can make the game win or lose by their choices. Which means you only won because your enemies all made worse choices then your allies. So the outcome was still influenced by the sum of ten players; even when using the ridiculous insecure logic that your individual choices entirely determine the outcome and thus if you lose oh it's all your fault and those nine other players had nothing to do with it.

Being good at the game literally doesn't matter because there's so many other players in the match that even if you're bad, if you get put on the better team they can still easily carry you. While if you're good, being put on the worse team doesn't matter because you physically can't carry all of them at once against a better enemy team.

At this point the ranking system might as well be grading who is the luckiest, abuses the most scummy class pairings with the duo que; and or uses the most hot meta that is so broken it literally can win you the entire match over the nine other players in it. Yes, some people who read this will just try to make up excuses to perpetuate the illusion that the one out of ten players in the match is the sole one influencing the outcome as long as that player is specifically them. Since if they accept that winning or losing is out of their hands then they can't claim credit for all their wins and generate ego from it.

Doesn't change the fact that this system is fundamentally based off coin flipping whether you get the winning or losing team regardless of whether you yourself are a good player or not.

And after a thousand a five hundred games that's my biggest complaint about this whole system, is you are rewarded based on the coin flip and not your own performance. I don't especially mind going on random, asinine losing streaks. It's the fact that the game steals days worth of rank progress from me while it simultaneously is the one putting me on that losing streak. Just makes the entire process feel totally pointless and miserable because it honestly just is.

sounds good, doesnt work

Link to comment
Share on other sites

@"Aisling.5901" said:I don't especially mind going on random, asinine losing streaks. It's the fact that the game steals days worth of rank progress from me while it simultaneously is the one putting me on that losing streak. Just makes the entire process feel totally pointless and miserable because it honestly just is.**

It's interesting that you describe wins as "progress" and losses as "stolen progress".

It's like you think you should always be gaining rating, as though this is an RPG where as you go along you "level up" or advance a story-line. That's not how a competitive ladder works. You aren't entitled to "progress". If you've reached your appropriate rating, games will pretty much be 50/50, or as you would describe it "a coin toss".

If you go on a 10 win-streak one day, its almost to be expected that the next day you'll have a 10 loss-streak. Luck was with you one day, and against you the next. This is classic bias "when I'm winning, it's because I'm good, when I'm losing, it's because my team sucks and the game is stealing from me!".

Link to comment
Share on other sites

@"Ragnar.4257" said:

It's interesting that you describe wins as "progress" and losses as "stolen progress".

It's like you think you should always be gaining rating, as though this is an RPG where as you go along you "level up" or advance a story-line. That's not how a competitive ladder works. You aren't entitled to "progress". If you've reached your appropriate rating, games will pretty much be 50/50, or as you would describe it "a coin toss".

If you go on a 10 win-streak one day, its almost to be expected that the next day you'll have a 10 loss-streak. Luck was with you one day, and against you the next. This is classic bias "when I'm winning, it's because I'm good, when I'm losing, it's because my team sucks and the game is stealing from me!".

So your only attempt at proving me wrong is saying that when I reach my skill ceiling winning should be random, even though I win randomly at every rank level while almost always getting high placement on my team. Also you completely ignore the entire chunk of my post which points out that winning is not based on your individual skill, yet you are still talking like it does when it factually does not. So basically you aren't saying anything other then you don't want to agree with me.

Link to comment
Share on other sites

@Sigmoid.7082 said:By placement do you mean top stats?

I mean your placement on your team. Unless I am grossly mistaken, at the end of the game you are displayed as being higher or lower on your teams listing based on how well you performed overall. Best player is at the top, worst player is at the bottom.

If that is not the case then my point remains the same; you should be graded based off how much you contributed to the team overall. How much were you playing the objective, how many players did you kill while in range of the objective, how much did you contribute to team fights. Top stats only measures how much you did of something mathematically, it doesn't measure how productive what you did was to the game however,

Link to comment
Share on other sites

@Ragnar.4257 said:

@Sigmoid.7082 said:By placement do you mean top stats?

I am grossly mistaken

Thank you for openly proving that you have no constructive basis for your opinion other then because you don't want to accept reality, and thus all you can do is be obnoxious for the sake of contesting my point without having any objective counterpoints. I will not be acknowledging you again due to this.

Link to comment
Share on other sites

You are grossly mistaken (your own words), because, the display at the end of the game and the order of players on that display has literally nothing to do with your performance.

The reason that personal performance is not used as a metric for rating, is that there isn't any metric that can properly distinguish a good performance from a bad performance.

If you spend all game auto-attacking an enemy scrapper on a point that the scrapper owns, you'll probably get top damage and top offense. And yet, you have thrown the game for your team. And if the scrapper is auto-ing you back, and you keep healing it up, you'll probably get top healing as well. Still doesn't mean you did anything useful. You can point to your "3 top stats" as much as you want, you are the reason your team lost.

Link to comment
Share on other sites

@Aisling.5901 said:

@Sigmoid.7082 said:By placement do you mean top stats?

I mean your placement on your team. Unless I am grossly mistaken, at the end of the game you are displayed as being higher or lower on your teams listing based on how well you performed overall. Best player is at the top, worst player is at the bottom.

If that is not the case then my point remains the same; you should be graded based off how much you contributed to the team overall. How much were you playing the objective, how many players did you kill while in range of the objective, how much did you contribute to team fights. Top stats only measures how much you did of something mathematically, it doesn't measure how productive what you did was to the game however,

They should discover a way to properly gauge who contributed the most and who contributed the least, so that it can slightly alter the best player's rating gain or loss, as well as the worst player's gain & loss. But they shouldn't display this stat... it would only create QQ.

Link to comment
Share on other sites

@Aisling.5901 said:

@"Ragnar.4257" said:

It's interesting that you describe wins as "progress" and losses as "stolen progress".

It's like you think you should always be gaining rating, as though this is an RPG where as you go along you "level up" or advance a story-line. That's not how a competitive ladder works. You aren't entitled to "progress". If you've reached your appropriate rating, games will pretty much be 50/50, or as you would describe it "a coin toss".

If you go on a 10 win-streak one day, its almost to be expected that the next day you'll have a 10 loss-streak. Luck was with you one day, and against you the next. This is classic bias "when I'm winning, it's because I'm good, when I'm losing, it's because my team sucks and the game is stealing from me!".

So your only attempt at proving me wrong is saying that when I reach my skill ceiling winning should be random, even though I win randomly at every rank level while almost always getting high placement on my team. Also you completely ignore the entire chunk of my post which points out that winning is not based on your individual skill, yet you are still talking like it does when it factually does not. So basically you aren't saying anything other then you don't want to agree with me.

The simplest metric for measuring this is counting wins and losses and grading them based on the quality of your opponents. That is what the glicko algorithm does.

There are plenty of measurable things someone could do that don't help the team win. For example staying in a group fight too long after the outcome is decided is bad for the team, but you can pad your capture time, kills and damage by doing it.

Link to comment
Share on other sites

How the game should meassure your skill, based on what? TOP stats? Many times it was proven TOP stats is a bad idea, people would rather hunt the stats instead of keeping up with the game objectives. So how else would you meassure your skill? According your performance vs an enemy? This is a team based game right, you can go cheap side node build and win all your 1v1s, but your team is strugling to win any team fight due bad team comp and you can't help them much with your 1v1 build. But what if you played support FB to aid your 2 necros in the team fight? You could shift the game on your side. Just because you play strong 1v1 build and win most 1v1s, it doesn't mean you did the best for your team to win the game, it doesn't mean you were the best at all.

The game has personal rank system in team based game, already this is nonsence, it will NEVER ever be perfect. But how to make it as fair as possible? Perhaps same conditions for everyone? Same team comps with the exact same builds against each other? That would be boring as hell right. But the fact is, the game tries to redistribute classes equaly, but it doesn't care about specialization. It might be huge difference if you have core guard, dragonhunter or firebrand. Already this small fact can be game breaking and the game is not even considering that.

Yes, the current system doesn't make much of a sence, but you would hardly find any better, sadly.

Link to comment
Share on other sites

Good performance isn't exactly subjective, it can be quantified.But the amount of variations that can happen in a match is a mind-bogglingly huge number.Making a good algorithm for this purpose would likely be as big of a project as the making process of this entire game was. It would take years to code, and the cost would likely be in counted in the hundreds of thousands of dollars. Governments and militaries pay exorbitant amounts of money to have these kinds of scenario simulators that can predict outcomes of actions and weigh whether your response was the correct one.It's just not worth the time, money and effort.

Link to comment
Share on other sites

@Aisling.5901 said:blablablaYeah sometimes you will make the right choices at the right times and that sways a balanced match into your favour but that's only with a somewhat balanced match; which do not happen very often, and only some of the time will you even have that opportunity.blablabla

That's exactly how you gain rank though. You take the wins and losses that the game hands to you(with this population this is just unavoidable), but sometimes when your performance really matters, you gotta tryhard. Win that 2v1 while pushing far, dont let your team collapse at the midfight, keep the dumber enemies chasing you... whatever you need to win on your current rating.I remember winning matches just by going DE(back when it was actually capable of killing things), engaging midfight and then kiting the 2-3 enemies jumping me the second I left stealth. Ended the match with like 2 kills, we've still won. It was a dangerous and risky dance, but when I managed without f***ups, it was what really decided the outcome of the match.

Of course to do this, you need to play something that can really shine and overperform in some situations, for example your vanilla scourge won't ever carry a match like a decent support or sidenoder could.

If you can't make peace with this, wait for 2v2 to be released. Your personal performance will REALLY matter there.

Link to comment
Share on other sites

@Aisling.5901 said:

@"Sigmoid.7082" said:By placement do you mean top stats?

How much were you playing the objective, how many players did you kill while in range of the objective,

So if I farm two players outside their spawn the entire game. Did I perform poorly?This whole "fight on points" is a big sign of people not understanding the game mode.

Link to comment
Share on other sites

@"AngelLovesFredrik.6741" said:

@"Sigmoid.7082" said:By placement do you mean top stats?

How much were you playing the objective, how many players did you kill while in range of the objective,

So if I farm two players outside their spawn the entire game. Did I perform poorly?This whole "fight on points" is a big sign of people not understanding the game mode.

It's been ingrained so hard in people's mind people will just step onto the point no matter what/how/when. But that's what differenciates experienced players from those who tunnel vision. People lack global comprehension of the flow of the match, but rotations and map presence, contrary to many believers isn't a fixed strategy. Imo nothing demands as much experience as knowing where to be at any given time given your capabilities. It's like ppl fighting for tranq for 5 minutes when opponent is happily tri capping. Lack of adaptability is what kills a specie.

Link to comment
Share on other sites

pvp is a minigame. win.,lose, it's pointless . it's unbalanced and impossibile to balance not to mention match manipulation, afks and so on.

just do like i do : enter fight, enjoy the combats. sometimes you have good duels, sometimes there are nice rubles. most of the time is a 500 100 or the opposite. just relax get your easy daily and the bizantium chest at the end of the journey. that's all.

Link to comment
Share on other sites

@"Aisling.5901" said:

you should be graded based off how much you contributed to the team overall. How much were you playing the objective, how many players did you kill while in range of the objective, how much did you contribute to team fights. Top stats only measures how much you did of something mathematically, it doesn't measure how productive what you did was to the game however

Yeah, it would be nice if this could be achieved in a reasonable time and at reasonable cost. As someone else already pointed out, the mix/degree of different factors that go into making that judgment seems extremely complex, and I would agree that trying to achieve this kind of analysis is simply unjustifiable for an aging mmo (no matter how much we love it).

I'm sure that if you reflect on your 1500+ matches' worth of experience, you can think of numerous times when either you or an enemy performed extremely important roles in the win that just aren't captured that well by stats. Back when I still participated in ranked and spellbreaker was still a bit overtuned, I remember trying to kill this one spellbreaker with my teammates. He (she? they?) was just incredibly skilled at using the map to their advantage - they seemed to have strong grasp of the key weapon ranges (600, 900, and 1200), and managed to use regular movement and some weapon movement to "out of range" and LoS a few of us consistently without even burning a dodge. The only person on our team that could actually stick to the spellbreaker was our core guardian with sword 2 port and Judge's Intervention, and of course when our guard showed up the spellbreaker disengaged even faster and ran back to teammates that could peel the guard off him. I quickly found out that chasing this player was the absolute worst idea, so I adjusted... but 2 or 3 of my other teammates did not. As a result, the other team just ran roughshod over the rest of the map.

What that experience taught me is that someone can be incredibly effective at contributing to a win, but there's almost no set of metrics that could capture what that spellbreaker achieved. Avoiding damage consistently well without dodges - when damage is mitigated by being out of range or obstructed, how much of that is the target being smart, and how much of that is the attacker being stupid/untrained/not a good judge of how to time channels and cast times? How would the game measure that? You'd have to do something crazy like assign a unique ID to every single possible player position on the map, and post-match the algorithm would have to sift the data and figure out - based on objective data like attacker vs target position at start of attack cast, reaction time gap between attack cast/response (if any), movement (and possibly even camera) vectors, etc - I'm sure if you paid someone a whole lot to sit down and think it over, it could be done... but I'm not sure they could properly implement that system before the game shuts down. And the aforementioned ideas don't even touch on the other thing the spellbreaker was really good at, which was to guide pursuers right into teammates who could then kill them and turn the tide. When is it a good idea to pull your pursuers into a particular teammate? The game would have to make a set of assumptions about every possible build out there, factor in what combination of active abilities and passive traits were on cooldown for that teammate, etc. It's... it's just mind boggling when you actually try to think it out.

Also note that none of this accounts for the fact that a trollishly hard to kill player is perhaps only as good as the idiots chasing him. That includes both the stupidity of my teammates refusing to stop wasting time on the spellbreaker, as well as team comp (maybe none of us that round had what it took to stand up to whatever the spellbreaker was running). Yet another whole rabbit hole's worth of factors to consider.

So OP, what you're left with is unfortunately some form of "git gud" or at least "git better." Hopefully it's clear that I'm not just saying that to dismiss your desire for more precise individual performance rewards. But tying mmr to top stats (as many have mentioned all over the forums) is problematic because it would incentivize farming stats instead of good play. Trying to drill further down from top stats is, as I tried to demonstrate in this wall of text, even more problematic. I think the best you can hope for is just playing thousands more matches and finding your way to more victories.

EDIT: forgot to include this

I thought this was actually one of the best attempts at reworking the scoreboard, and @shadowpass.4236 rightly refrains from stepping into the unworkable morass of trying to measure metrics that have a very uncertain impact on victory.

Link to comment
Share on other sites

You want skill based rank? Switch ranked play back to team queue. No random factor there for you to blame!

Do you really think ANet can implement an alternative rating system which works? No other company has been successful with one yet. Plus, ANet has been working in Swiss for over a year, so don't expect it to be implemented any time soon.

Link to comment
Share on other sites

@Exedore.6320 said:You want skill based rank? Switch ranked play back to team queue. No random factor there for you to blame!

I think this hits a big part of the truth. Queuing with a full team means you should be judged on only win or loss.

I can see the OP's point, though, that queuing as an individual is many times a crapshoot. Some professions have it better than others, of course. My poor Thief is very dependent on the rest of my team's ability to win and hold points because I can't do much to help in that. I'm bouncing back and forth between decapping empty points and hunting for +1's opportunities. If I get stranded alone on a point, I have to run away at first sign of the enemy...

...And it's always an insanely beautiful sight to behold, even if a depressing one, whenever my team absolutely dominates the Mid fight at the start of a match, only for all 5 of our players to split up and go their own ways immediately after the last enemy dies... sometimes, even before we actually capture the point... so it's left completely empty and neutral. As a Thief, I'm forced with the choice of doing my actual job or to go fully capture Mid all alone, then try to defend it against the incoming enemy wave while my teammates get picked off one by one.

There's nothing fulfilling in that kind of gameplay. And to lose rank because of that, it's even more frustrating. Being a Thief player makes it even worse because the salt usually gets shoveled at me. At least some professions can bunker and hold a point against several enemies, while others can run right into the fray to impose their will. I'd be happy just to be able to 1v1 decently on point.

I'm not sure what a good, fair scoring solution would look like for solo-queue, but it does seem that there is some call for a different one compared to a 5-man queue. Perhaps just not allow the loss of rank/pips for a loss, but significantly raise the amount of gains for victory and also the requirements to hit higher ranks. Yeah, eventually, even the worst player could hit the top rank with enough grinding, but some still can now if they get lucky with their randomly chosen teammates. Or perhaps allow only the top rank(s) to have loss of pips/ranks so only the elite players can move up to the very tip-top, but other players can still get rank rewards and achievements. But yeah, every solution seems to have problems.

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...