Anyway the fact of the matter is, unless you live in australia, don't play on rookie servers if your presence determines the outcome of the game.
This isn't at all true. I'm not going to just not play because the only open server has bad players on it. I am in the US. Someone, probably you, made this claim that there are always open servers earlier in this thread. Right after reading that, I fired up the game and had two rookie friendly servers as the only two options available to me. It really made me laugh. I'm not going to play on a pub with 150 ping or skip playing. If this is what my server list looks like, it isn't going to be better for others.
I did leave a bad pub to join a different one when a spot became available though.
I really don't see how people are seeing this huge problem though. To me, it sure seems like the good players avoid rookie friendly servers. I get really good scores on them and get called a cheater. When I join non rookie friendly servers, I have as many really bad rounds as good rounds.
I recommend that you take a step back and look at this thread objectively. No one is stating that good players should not be allowed in rookie servers, the issue here is that new players frequently have no environment that they are comfortable playing in.
Your nitty gritty details and assumptions only highlight the issue with a lot of good players, that they are unwilling to sacrifice their ego/opinion for the sake of the game's future.
It would be great if you could prove that my friends simply 'don't like the game' but that is not what I have been told, and I prefer not to assume based on my own individual perspective.
Waiting an hour for matchmaking, what a terrific idea.
Soft matchmaking could work well even with NS2's small playerbase. Something like this:
a) The server browser shows the average skill rating of the players currently on each server, and
b) The ready room (or tab playerlist) shows the average skill rating of the players currently on each team.
People could still join any server and team just like now. But with the extra information, people would be able to make better balanced matches all by themselves without needing the overhead (and waiting times) of forced matchmaking.
If you want a challenge, join a server with a high skill rating. If you want an easy game, join a low one. Once you're in a server, stack the teams any way you'd like - but if the playerlist shows that one team is higher-skilled, expect the other team to have a tough time. And if the skill difference is huge, expect a stomping, or just F4 and shuffle the teams to get a closer (and more fun) match.
This would be a pretty simple change (the hardest part is generating the player rating, but UWE has all the stats they need to do this), but even just this much would go a long way towards getting more games that were more fun because they were more even.
That COULD actually work. But how do you work out someones skill rating? You can't use k/d that doesn't really mean someone is bad at the game, maybe they suicide for RT's all game. Can't really use the elo system, which was designed for 1v1 play, since games like 12v12 would throw it way out of whack.
I recommend that you take a step back and look at this thread objectively. No one is stating that good players should not be allowed in rookie servers, the issue here is that new players frequently have no environment that they are comfortable playing in.
Your nitty gritty details and assumptions only highlight the issue with a lot of good players, that they are unwilling to sacrifice their ego/opinion for the sake of the game's future.
It would be great if you could prove that my friends simply 'don't like the game' but that is not what I have been told, and I prefer not to assume based on my own individual perspective.
You have stated clearly that you think it should be rookie only.
You state there is a problem, but you can't show the problem. This doesn't make any sense.
You have stated clearly that you think it should be rookie only.
You state there is a problem, but you can't show the problem. This doesn't make any sense.
I'm going to stop responding to you. My suggestion in the first post is not the only solution and you would know that if you read the whole thread.
The problem is shown in this thread by various people who have posted their experience, as well as me posting mine. If you need screenshots to see that a problem is viable you will miss 50% of the problems that occur in your life.
To which you respond to read more than one post... lol...
There is a problem, I just can't show it! ...
Now lets explain why your friends will never like this game. They want a rookie onlt experience. This will mean that they never learn to play the game and have to stick to this rookie only thing. The difference between rookies and average players is pretty huge. It all comes from movement, positioning, and learning to aim at stuff that moves fast. If you are only playing other terribpe players, you won't learn to not run on the ground as a skulk. You also will only be aiming at floor skulks, which means you don't learn to aim. What good does this do for your friends? They will still get run over every time they try playing on non rookie only servers because they aren't learning how to play on them.
Tjis game takes a lot of time to learn and get good at. You will never achieve this by playing with others who don't know what they are doing. This is much different than most modern games. It isn't for everyone.
That COULD actually work. But how do you work out someones skill rating? Can't really use the elo system, which was designed for 1v1 play, since games like 12v12 would throw it way out of whack.
The basic concept behind Elo is sound, and it can be extended to team games as well. You end up with something like Microsoft's TrueSkill.
The exact details don't matter all that much. The fundamental concept is to calculate a rating using a person's wins and losses, taking into consideration the strength of the players they won and lost against, such that the rating can be used to accurately predict their chance of winning against another player. This has been researched a lot and the basics are well-understood. NS2 introduces one wrinkle that's already been handled before (team games) and two more that probably haven't been handled before (asymmetric sides, and highly-differentiated roles i.e. comms vs. field players). But any competent and interested statistician could develop a reasonably useful system given access to UWE's raw data.
And any such system would be a vast improvement over what we have now, which is nothing, which all-too-often leads to pissed-off players and ruined games (for low-skill and high-skill players alike).
Matchmaking gives me a headache, if it was implemented it should be optional
Re optional: agreed.
Re headache: what don't you like?
I don't like the flaws that exist in it, such as L4D2, Modern Warfare 3. console type matchmaking. I think that an ELO rating or something would be handy, but that's a lot of work for something with little benefit. I also do not like how difficult it is to stay in a good server, with plenty of regulars. I have a ton of friends that play on my reg server, which is the NS2 Geriatric Ward OLDF.net server. It's got amazing adminning, reserve slot if you play there often, and they're all solid guys. To top it off, many good players play there such as Ekrizon, w0k, Boom, and others, and I enjoy their company. Matchmaking would potentially put a dent in that.
Matchmaking gives me a headache, if it was implemented it should be optional
Re optional: agreed.
Re headache: what don't you like?
I don't like the flaws that exist in it, such as L4D2, Modern Warfare 3. console type matchmaking. I think that an ELO rating or something would be handy, but that's a lot of work for something with little benefit. I also do not like how difficult it is to stay in a good server, with plenty of regulars. I have a ton of friends that play on my reg server, which is the NS2 Geriatric Ward OLDF.net server. It's got amazing adminning, reserve slot if you play there often, and they're all solid guys. To top it off, many good players play there such as Ekrizon, w0k, Boom, and others, and I enjoy their company. Matchmaking would potentially put a dent in that.
I like the idea of ELO but I don't think there's enough players to bother with it. There's a certain old-school charm to the open public server/client model that NS2 uses that, like you say, would have a damper put on it if there was a matchmaking system alongside it.
Dealing with people who leave early or are otherwise obnoxious is also a huge pain in the ass with matchmaking games. It's pretty much why I can't play MOBAs despite all of my friends being Dota 2 fanatics (and it's not for lack of trying to like it, either). It might be the genre itself, but I think the matchmaking (specifically in that you can't do anything about a player leaving to ruin a game) certainly hurts a lot as well.
Dealing with people who leave early or are otherwise obnoxious is also a huge pain in the ass with matchmaking games. It's pretty much why I can't play MOBAs despite all of my friends being Dota 2 fanatics (and it's not for lack of trying to like it, either). It might be the genre itself, but I think the matchmaking (specifically in that you can't do anything about a player leaving to ruin a game) certainly hurts a lot as well.
NS2's drop-in drop-out style of gameplay (for pub games, obviously not for comp matches) is a real strength. A soft matchmaking system like the kind I described here would preserve that type of gameplay. Basically nothing would change, except that people could play more closely-matched games if they wanted to.
I like the idea of ELO but I don't think there's enough players to bother with it. There's a certain old-school charm to the open public server/client model that NS2 uses that, like you say, would have a damper put on it if there was a matchmaking system alongside it.
Dealing with people who leave early or are otherwise obnoxious is also a huge pain in the ass with matchmaking games. It's pretty much why I can't play MOBAs despite all of my friends being Dota 2 fanatics (and it's not for lack of trying to like it, either). It might be the genre itself, but I think the matchmaking (specifically in that you can't do anything about a player leaving to ruin a game) certainly hurts a lot as well.
How exactly does a system without matchmaking solve the two issues you mentioned? In NS2 there are a lot of obnoxious people, as well as a lot of people who concede early. This is not related to the matchmaking system in any way. Matchmaking has very few drawbacks, and one of the huge positives is that every single new player is put into an environment with other new players so they can learn together. Experienced players are put into an environment where they are challenged and therefore encouraged to pursue their goals such as e-sports, etc. DotA 2 also provides the tools necessary to learn without having to have a pro shove knowledge down your throat, you can learn at your own pace and still become a pro if you put in the time and effort. Obviously this is much more welcoming, which means players are more likely to play again.
I know there are a lot of nostalgic old-school gamers who think that the die hard approach is the right way to game. That's great, but even though I adapt that approach, I'm not obnoxious enough to think that others who don't have the same approach are wrong. Not everyone plays games for the same reasons, and the goal as a developer should be to get as many people to play your game as possible, especially when you owe investors on their investments. Even as I player I feel obligated to help keep players in the game, which is why I created this thread, so I could hopefully get some insight on a problem that caused my friends to leave the game. Or at least that is what they claim, and I have no reason to doubt that, as I have seen it myself.
I do appreciate all of the on-topic responses that did provide insight into the situation, they will help me convince my friends to play again. That's called being constructive. Thank you!
I like the idea of ELO but I don't think there's enough players to bother with it. There's a certain old-school charm to the open public server/client model that NS2 uses that, like you say, would have a damper put on it if there was a matchmaking system alongside it.
Dealing with people who leave early or are otherwise obnoxious is also a huge pain in the ass with matchmaking games. It's pretty much why I can't play MOBAs despite all of my friends being Dota 2 fanatics (and it's not for lack of trying to like it, either). It might be the genre itself, but I think the matchmaking (specifically in that you can't do anything about a player leaving to ruin a game) certainly hurts a lot as well.
How exactly does a system without matchmaking solve the two issues you mentioned? In NS2 there are a lot of obnoxious people, as well as a lot of people who concede early. This is not related to the matchmaking system in any way. Matchmaking has very few drawbacks, and one of the huge positives is that every single new player is put into an environment with other new players so they can learn together. Experienced players are put into an environment where they are challenged and therefore encouraged to pursue their goals such as e-sports, etc. DotA 2 also provides the tools necessary to learn without having to have a pro shove knowledge down your throat, you can learn at your own pace and still become a pro if you put in the time and effort. Obviously this is much more welcoming, which means players are more likely to play again.
I know there are a lot of nostalgic old-school gamers who think that the die hard approach is the right way to game. That's great, but even though I adapt that approach, I'm not obnoxious enough to think that others who don't have the same approach are wrong. Not everyone plays games for the same reasons, and the goal as a developer should be to get as many people to play your game as possible, especially when you owe investors on their investments. Even as I player I feel obligated to help keep players in the game, which is why I created this thread, so I could hopefully get some insight on a problem that caused my friends to leave the game. Or at least that is what they claim, and I have no reason to doubt that, as I have seen it myself.
I do appreciate all of the on-topic responses that did provide insight into the situation, they will help me convince my friends to play again. That's called being constructive. Thank you!
I didn't mean to dump on the idea at all. I was actually a little surprised when I saw there were no plans to formalize team sizes and use a matchmaking system in the alpha and beta. The issue I see now is that we have what we have, and if you want to add matchmaking you have to either replace the existing system (which is a non-starter for a lot of people) or craft a system that can play nice with the existing system. That's not a trivial challenge and there's a lot of ways it can go haywire. Do you have some servers that are matchmaking only and other servers that are drop-in/drop-out? Do you allow players to join and leave servers with matchmaking enabled as freely as they wish? If so, how do you account for skill discrepancy in matched games when the players you're matching against leave and are replaced by someone with a different rating? What metrics do you use to rate individual players; score/time, average/lifetime K:D, average/lifetime W:L, resource efficiency for points scored, etc.? Do you allow rating metrics to be observed by players or keep them hidden? How do you account for the difference between an individual's performance and the performance of their team; after all, NS2 is a team game and a great player is sometimes dragged down by his team and a mediocre team is sometimes carried to victory by a great player.
Again, I'm not trying to rain on the parade here, I'm just trying to put it into context. Systems like this are very difficult to create both from a design perspective as well as a technical perspective, and that challenge is amplified immensely when you're trying to add it on top of an existing system that is completely different.
It seems a lot of good players simply lack the balls to go againt each others and prefer to kill noobs and casual players. Thus the game audience will never get very large.
It sadden me but everytime I show the game to someone the comment: "wow these 3 guys are making all the kills" worm it's way out. The enthousiast for the game kinda get a solid cold shower.
The issue I see now is that we have what we have, and if you want to add matchmaking you have to either replace the existing system (which is a non-starter for a lot of people) or craft a system that can play nice with the existing system.
I suggested a simple matchmaking mechanism that would play nice with the existing system in this post.
That's not a trivial challenge and there's a lot of ways it can go haywire.
My suggested scheme would be non-trivial, but still entirely doable (and not very expensive, I think).
----
To answer your specific questions:
>Do you have some servers that are matchmaking only and other servers that are drop-in/drop-out?
My suggested scheme uses "soft" matchmaking and is different from matchmaking in other games. All games would still be drop-in/drop-out just like they are today, but players would be able to find games with teammates and opponents of their own skill level if they so choose. So all servers would be matchmaking, and all servers would be drop-in/drop-out.
>Do you allow players to join and leave servers with matchmaking enabled as freely as they wish?
Yes.
>If so, how do you account for skill discrepancy in matched games when the players you're matching against leave and are replaced by someone with a different rating?
My suggested scheme doesn't guarantee skill-matched games. Instead, it informs everyone of the current skill balance (or lack thereof). So if you're playing in a closely matched game, and a low-skill player leaves and is replaced by a high-skill player, then oh well, the game is no longer closely matched. But at least everyone playing would be able to see just how badly matched the game has become - quantitatively, rather than just "hey what happened why are we losing so badly all of a sudden".
>What metrics do you use to rate individual players; score/time, average/lifetime K:D, average/lifetime W:L, resource efficiency for points scored, etc.?
Something similar to Elo or TrueSkill, i.e. a rating derived strictly from wins, losses, and the strength of the opponents won or lost against. Any other metrics are beside the point and would introduce unwanted biases or perverse incentives.
>Do you allow rating metrics to be observed by players or keep them hidden?
I've gone back and forth on this. Hiding them is nice, because it keeps players from treating their rating as a type of score and thus trying to game the system to boost it. The system can't actually be gamed, but most people don't understand that, so hiding scores prevents a lot of pointless posturing and drama. But in the technical implementations I've considered, people's scores would be available to server operators if they tried hard to get them, which means that people's scores would eventually become public knowledge anyway, so maybe it's better not to try to hide them at all in the first place.
>How do you account for the difference between an individual's performance and the performance of their team; after all, NS2 is a team game and a great player is sometimes dragged down by his team and a mediocre team is sometimes carried to victory by a great player.
A rating scheme similar to TrueSkill would handle team play quite well. Your rating would reflect the degree to which your presence on a team increases that team's chances of winning a game. In other words, it measures your contribution to a team's victory without needing to measure the thousands of individual components that such a contribution might consist of.
Again, I'm not trying to rain on the parade here, I'm just trying to put it into context. Systems like this are very difficult to create both from a design perspective as well as a technical perspective, and that challenge is amplified immensely when you're trying to add it on top of an existing system that is completely different.
Difficult, yes. But doable. Very, very doable, and I would even go so far as to say well worth doing.
It’s going to happen pro players need to have fun in some way too. However you can tell what kind of player they are when they post videos of themselves owning new players just to show how good they really are. Fortunately those kinds of players and the servers I play on tend to get banned.
Ppl dont always play pub, but when they do, they don't want to spent 20mins trying to fill up an empty server or trying to find a open slot on an "appropriate" server.
Neiher rookies nor "pros" do this.
Ppl just join the first server with good performance and ping that is populated and has an open slot most of the time.
I dont want to defend bastards that actively try to make rookie servers ragequit, but dunno i cant quite believe that this is a common thing.
It seems a lot of good players simply lack the balls to go againt each others and prefer to kill noobs and casual players. Thus the game audience will never get very large.
It sadden me but everytime I show the game to someone the comment: "wow these 3 guys are making all the kills" worm it's way out. The enthousiast for the game kinda get a solid cold shower.
Casual players who aren't bad should do what then? Not play at all? And the decent players should wait hours for a gather that may or not be good right?
Can't see the argument for matchmaking when the playbase is so small and loses it's new players so quickly.
From what I've noticed when a player have better skills (and kills) then most and doesn't win he quit. That's what happen most of the time from my experience. So my idea of some players actively looking for an easy kill looks accurate.
A lot of good players who aren't affraid of a fight will stick to the same servers and those are not uncommon. There are a lot of good players around. It should be pretty easy for them to get a fair game.
Difficult, yes. But doable. Very, very doable, and I would even go so far as to say well worth doing.
I don't see what the point of that would be, at all. So a system that informs us that the game is imbalanced in terms of 'skill' (how accurate can these measurements be..? sometimes a game just goes lopsided, have you ever played dota? How's the ELO work out in that? I had more games that were better before ELO was introduced.) They are going to create this all just to inform people that there is a skill differential, that they could have logically deduced from the outcome of the game. No thanks, I don't want any sort of matchmaking.
I often am on rookie friendly servers because honestly they aren't rookie only servers they are just supposed to be places where they aren't going to be bothered for sucking and stuff, it's hard enough finding a server with good ping that has the right amount of players and isn't full, let alone sitting around waiting for some special server you're supposed to find if you're not horrible. On these servers more often the issue is one team is just filled with people who have absolutely no clue, and so they get squashed.
Comments
Re optional: agreed.
Re headache: what don't you like?
This isn't at all true. I'm not going to just not play because the only open server has bad players on it. I am in the US. Someone, probably you, made this claim that there are always open servers earlier in this thread. Right after reading that, I fired up the game and had two rookie friendly servers as the only two options available to me. It really made me laugh. I'm not going to play on a pub with 150 ping or skip playing. If this is what my server list looks like, it isn't going to be better for others.
I did leave a bad pub to join a different one when a spot became available though.
I really don't see how people are seeing this huge problem though. To me, it sure seems like the good players avoid rookie friendly servers. I get really good scores on them and get called a cheater. When I join non rookie friendly servers, I have as many really bad rounds as good rounds.
I recommend that you take a step back and look at this thread objectively. No one is stating that good players should not be allowed in rookie servers, the issue here is that new players frequently have no environment that they are comfortable playing in.
Your nitty gritty details and assumptions only highlight the issue with a lot of good players, that they are unwilling to sacrifice their ego/opinion for the sake of the game's future.
It would be great if you could prove that my friends simply 'don't like the game' but that is not what I have been told, and I prefer not to assume based on my own individual perspective.
That COULD actually work. But how do you work out someones skill rating? You can't use k/d that doesn't really mean someone is bad at the game, maybe they suicide for RT's all game. Can't really use the elo system, which was designed for 1v1 play, since games like 12v12 would throw it way out of whack.
You have stated clearly that you think it should be rookie only.
You state there is a problem, but you can't show the problem. This doesn't make any sense.
I'm going to stop responding to you. My suggestion in the first post is not the only solution and you would know that if you read the whole thread.
The problem is shown in this thread by various people who have posted their experience, as well as me posting mine. If you need screenshots to see that a problem is viable you will miss 50% of the problems that occur in your life.
To which you respond to read more than one post... lol...
There is a problem, I just can't show it! ...
Now lets explain why your friends will never like this game. They want a rookie onlt experience. This will mean that they never learn to play the game and have to stick to this rookie only thing. The difference between rookies and average players is pretty huge. It all comes from movement, positioning, and learning to aim at stuff that moves fast. If you are only playing other terribpe players, you won't learn to not run on the ground as a skulk. You also will only be aiming at floor skulks, which means you don't learn to aim. What good does this do for your friends? They will still get run over every time they try playing on non rookie only servers because they aren't learning how to play on them.
Tjis game takes a lot of time to learn and get good at. You will never achieve this by playing with others who don't know what they are doing. This is much different than most modern games. It isn't for everyone.
The basic concept behind Elo is sound, and it can be extended to team games as well. You end up with something like Microsoft's TrueSkill.
The exact details don't matter all that much. The fundamental concept is to calculate a rating using a person's wins and losses, taking into consideration the strength of the players they won and lost against, such that the rating can be used to accurately predict their chance of winning against another player. This has been researched a lot and the basics are well-understood. NS2 introduces one wrinkle that's already been handled before (team games) and two more that probably haven't been handled before (asymmetric sides, and highly-differentiated roles i.e. comms vs. field players). But any competent and interested statistician could develop a reasonably useful system given access to UWE's raw data.
And any such system would be a vast improvement over what we have now, which is nothing, which all-too-often leads to pissed-off players and ruined games (for low-skill and high-skill players alike).
I don't like the flaws that exist in it, such as L4D2, Modern Warfare 3. console type matchmaking. I think that an ELO rating or something would be handy, but that's a lot of work for something with little benefit. I also do not like how difficult it is to stay in a good server, with plenty of regulars. I have a ton of friends that play on my reg server, which is the NS2 Geriatric Ward OLDF.net server. It's got amazing adminning, reserve slot if you play there often, and they're all solid guys. To top it off, many good players play there such as Ekrizon, w0k, Boom, and others, and I enjoy their company. Matchmaking would potentially put a dent in that.
Consistent pings(And low ones, for me, because of proximity).
I don't like that it's a 19 player server though.
It's actually 18 players in practice, and happens to be my favorite level of players. (16-18 is the sweet spot).
Geriatric Ward has best ping for me, and usually fairly skilled players. My favorite server so far
I like the idea of ELO but I don't think there's enough players to bother with it. There's a certain old-school charm to the open public server/client model that NS2 uses that, like you say, would have a damper put on it if there was a matchmaking system alongside it.
Dealing with people who leave early or are otherwise obnoxious is also a huge pain in the ass with matchmaking games. It's pretty much why I can't play MOBAs despite all of my friends being Dota 2 fanatics (and it's not for lack of trying to like it, either). It might be the genre itself, but I think the matchmaking (specifically in that you can't do anything about a player leaving to ruin a game) certainly hurts a lot as well.
NS2's drop-in drop-out style of gameplay (for pub games, obviously not for comp matches) is a real strength. A soft matchmaking system like the kind I described here would preserve that type of gameplay. Basically nothing would change, except that people could play more closely-matched games if they wanted to.
How exactly does a system without matchmaking solve the two issues you mentioned? In NS2 there are a lot of obnoxious people, as well as a lot of people who concede early. This is not related to the matchmaking system in any way. Matchmaking has very few drawbacks, and one of the huge positives is that every single new player is put into an environment with other new players so they can learn together. Experienced players are put into an environment where they are challenged and therefore encouraged to pursue their goals such as e-sports, etc. DotA 2 also provides the tools necessary to learn without having to have a pro shove knowledge down your throat, you can learn at your own pace and still become a pro if you put in the time and effort. Obviously this is much more welcoming, which means players are more likely to play again.
I know there are a lot of nostalgic old-school gamers who think that the die hard approach is the right way to game. That's great, but even though I adapt that approach, I'm not obnoxious enough to think that others who don't have the same approach are wrong. Not everyone plays games for the same reasons, and the goal as a developer should be to get as many people to play your game as possible, especially when you owe investors on their investments. Even as I player I feel obligated to help keep players in the game, which is why I created this thread, so I could hopefully get some insight on a problem that caused my friends to leave the game. Or at least that is what they claim, and I have no reason to doubt that, as I have seen it myself.
I do appreciate all of the on-topic responses that did provide insight into the situation, they will help me convince my friends to play again. That's called being constructive. Thank you!
I ping 35 there.
It really is a great server, the average player skill seems higher there than on the average pub server.
Again, I'm not trying to rain on the parade here, I'm just trying to put it into context. Systems like this are very difficult to create both from a design perspective as well as a technical perspective, and that challenge is amplified immensely when you're trying to add it on top of an existing system that is completely different.
It sadden me but everytime I show the game to someone the comment: "wow these 3 guys are making all the kills" worm it's way out. The enthousiast for the game kinda get a solid cold shower.
My suggested scheme would be non-trivial, but still entirely doable (and not very expensive, I think).
----
To answer your specific questions:
>Do you have some servers that are matchmaking only and other servers that are drop-in/drop-out?
My suggested scheme uses "soft" matchmaking and is different from matchmaking in other games. All games would still be drop-in/drop-out just like they are today, but players would be able to find games with teammates and opponents of their own skill level if they so choose. So all servers would be matchmaking, and all servers would be drop-in/drop-out.
>Do you allow players to join and leave servers with matchmaking enabled as freely as they wish?
Yes.
>If so, how do you account for skill discrepancy in matched games when the players you're matching against leave and are replaced by someone with a different rating?
My suggested scheme doesn't guarantee skill-matched games. Instead, it informs everyone of the current skill balance (or lack thereof). So if you're playing in a closely matched game, and a low-skill player leaves and is replaced by a high-skill player, then oh well, the game is no longer closely matched. But at least everyone playing would be able to see just how badly matched the game has become - quantitatively, rather than just "hey what happened why are we losing so badly all of a sudden".
>What metrics do you use to rate individual players; score/time, average/lifetime K:D, average/lifetime W:L, resource efficiency for points scored, etc.?
Something similar to Elo or TrueSkill, i.e. a rating derived strictly from wins, losses, and the strength of the opponents won or lost against. Any other metrics are beside the point and would introduce unwanted biases or perverse incentives.
>Do you allow rating metrics to be observed by players or keep them hidden?
I've gone back and forth on this. Hiding them is nice, because it keeps players from treating their rating as a type of score and thus trying to game the system to boost it. The system can't actually be gamed, but most people don't understand that, so hiding scores prevents a lot of pointless posturing and drama. But in the technical implementations I've considered, people's scores would be available to server operators if they tried hard to get them, which means that people's scores would eventually become public knowledge anyway, so maybe it's better not to try to hide them at all in the first place.
>How do you account for the difference between an individual's performance and the performance of their team; after all, NS2 is a team game and a great player is sometimes dragged down by his team and a mediocre team is sometimes carried to victory by a great player.
A rating scheme similar to TrueSkill would handle team play quite well. Your rating would reflect the degree to which your presence on a team increases that team's chances of winning a game. In other words, it measures your contribution to a team's victory without needing to measure the thousands of individual components that such a contribution might consist of.
Difficult, yes. But doable. Very, very doable, and I would even go so far as to say well worth doing.
It's easy enought to find a regular team at higher level and other players should be allowed to learn at their own pace if they want to.
Neiher rookies nor "pros" do this.
Ppl just join the first server with good performance and ping that is populated and has an open slot most of the time.
I dont want to defend bastards that actively try to make rookie servers ragequit, but dunno i cant quite believe that this is a common thing.
And they could add server option so that on rookie servers a player who does like 10k / 1d will be kicked. something like that.
Also add a balancing system that forces people on teams so that people with high K/D cannot stack on the server.
Casual players who aren't bad should do what then? Not play at all? And the decent players should wait hours for a gather that may or not be good right?
Can't see the argument for matchmaking when the playbase is so small and loses it's new players so quickly.
A lot of good players who aren't affraid of a fight will stick to the same servers and those are not uncommon. There are a lot of good players around. It should be pretty easy for them to get a fair game.
I don't see what the point of that would be, at all. So a system that informs us that the game is imbalanced in terms of 'skill' (how accurate can these measurements be..? sometimes a game just goes lopsided, have you ever played dota? How's the ELO work out in that? I had more games that were better before ELO was introduced.) They are going to create this all just to inform people that there is a skill differential, that they could have logically deduced from the outcome of the game. No thanks, I don't want any sort of matchmaking.
I often am on rookie friendly servers because honestly they aren't rookie only servers they are just supposed to be places where they aren't going to be bothered for sucking and stuff, it's hard enough finding a server with good ping that has the right amount of players and isn't full, let alone sitting around waiting for some special server you're supposed to find if you're not horrible. On these servers more often the issue is one team is just filled with people who have absolutely no clue, and so they get squashed.