<!--quoteo(post=2036997:date=Nov 28 2012, 11:07 PM:name=sedek)--><div class='quotetop'>QUOTE (sedek @ Nov 28 2012, 11:07 PM) <a href="index.php?act=findpost&pid=2036997"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->You misunderstand. Mechanically, players who are more experienced, players who have more time invested and thus are more skilled, those players are more accurate. Over time, if you are improving as a player, your total accuracy will go up. Outside of pre-scheduled matches, people generally don't play gorge one hundred percent of the time, and even assuming they do, rolling the hydra accuracy into the player's totals will balance that out (similarly, weighting Lerk spikes and shotgun blasts differently will balance the outcome). Extended suppressing fire is behavior that, in NS2, doesn't make any sense, and as such, counts against your accuracy and thus your potential skill rating, while short bursts on rare occasions to keep a skulk's snout out won't negatively affect the overall accuracy of a skilled player, they will make up for it over time.<!--QuoteEnd--></div><!--QuoteEEnd-->
Says who? You can't just handwave "good players are accurate and we can perfectly compensate for support classes and playstyles with a few gimmicks" and expect us to buy it. You have no evidence supporting any sort of global unifying trend amongst "skilled" players beyond your own anecdotes, and every method of measurement of success you expound misses so many major variables (accuracy with different weapons, playstyle, variation in amount of effort put into playing, how different classes treat accuracy to name a few) that they are as useless as selecting random numbers. Additionally, your extreme disregard to these major issues shows how little you care or understand about the issues involved in globally determining something as vague as skill for a highly tactical game like this.
<!--quoteo(post=2037005:date=Nov 28 2012, 09:20 PM:name=Techercizer)--><div class='quotetop'>QUOTE (Techercizer @ Nov 28 2012, 09:20 PM) <a href="index.php?act=findpost&pid=2037005"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Says who? You can't just handwave "good players are accurate and we can perfectly compensate for support classes and playstyles with a few gimmicks" and expect us to buy it. You have no evidence supporting any sort of global unifying trend amongst "skilled" players beyond your own anecdotes, and every method of measurement of success you expound misses so many major variables (accuracy with different weapons, playstyle, variation in amount of effort put into playing, how different classes treat accuracy to name a few) that they are as useless as selecting random numbers. Additionally, your extreme disregard to these major issues shows how little you care or understand about the issues involved in globally determining something as vague as skill for a highly tactical game like this.<!--QuoteEnd--></div><!--QuoteEEnd-->
I don't see what's difficult to grasp about the idea that players who have more skill have better aim across the board. They are more familiar with weapon handling and melee mechanics simply by nature of having more time to get used to them. I addressed the concern about differing fire rates and spread patterns in that some weapons may be weighted differently, and quite frankly, the effort put into playing is something no matchmaking system that I know of deals with. If you're going to play in a ranked mode, you are expected to bring your A game, and choosing to bring your C game is going to get you quickly rematched with people who's A game is your C game until such time as you decide to push yourself back up to where you were. This also helps mitigate (not eliminate) skilled players purposefully preying on new or inexperienced players, as they will have to maintain their lower skill level or the system will quickly kick them back up the food chain where they are performing. Play style, unfortunately, isn't really a thing in NS2. This game doesn't have much depth as it stands. Either you're a good player or you're not, there isn't much middle ground so far A supporting player will still need to defend themselves, and thus will contribute to their accuracy rating.
3rd party servers that are possibly running mods means this will never happen
Match making is for games where the company hosts everything and then has access to better statistical tracking where rank and ratings might mean something
Although it would be amusing to think that in a perfect match making system that all the cheaters would be forces to play with each other on one server<!--QuoteEnd--></div><!--QuoteEEnd-->
everyone in grnadmaster is cheating?
<!--quoteo(post=2036857:date=Nov 28 2012, 04:43 PM:name=wiry)--><div class='quotetop'>QUOTE (wiry @ Nov 28 2012, 04:43 PM) <a href="index.php?act=findpost&pid=2036857"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The same ones that are hosting servers now preferably. Also, don't forget that UWE are having a couple of official servers up. There is no reason what so ever why there suddenly would be no servers.
Also, if the public chaos is appealing just mod it, people would still play.<!--QuoteEnd--></div><!--QuoteEEnd-->
maybe you make me a baby? :3
Whenever a new player takes part into gather they are amazed by the depth of gameplay this game has to offer, eventhough they get destroyed the first few times.
<!--quoteo(post=2036865:date=Nov 28 2012, 04:58 PM:name=ScardyBob)--><div class='quotetop'>QUOTE (ScardyBob @ Nov 28 2012, 04:58 PM) <a href="index.php?act=findpost&pid=2036865"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->In my experience, winrates in pubs are not random, but highly related to the skill level of each side (most specifically to how well the marine team can shoot). The best approach I've seen for a NS2 Elo-style ranking system would be: - Percentage of losing teams summed individual Elo-rating is transferred to the winning team - The Elo-points won or lost are distributed among the teams by their score*
*This is predicated on the score system being updated such that things such as healing/reparing and assists give points.<!--QuoteEnd--></div><!--QuoteEEnd-->
even simplier, you gain ranking points when you win a round and you lose ranking points if you leave or lose a round based around the amount of ranking points the opposition has. At the beginning everyone would have like 1000 points and the better players should have a positive impact on their team and make them win more often, as they build up points they matched against other players who built around the same amount of points and they should be expected to be around their own skill level.
Also the incentive to gain ranking points by winning a round will motivate players to be less selfish in their play (which is the problem with your system, you gain ranking by paying attention to your points.)
<!--quoteo(post=2036875:date=Nov 28 2012, 08:15 PM:name=biz)--><div class='quotetop'>QUOTE (biz @ Nov 28 2012, 08:15 PM) <a href="index.php?act=findpost&pid=2036875"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->i think the scoring system is good enough to distribute points among players, at least for the winning team. i see the better players getting higher scores nearly all the time in pubs<!--QuoteEnd--></div><!--QuoteEEnd--> You've obviously never played gorge, and always got that last hit on a cc.
I'd like to have an opt in ladder system, just because I like to have more supporting features in the game, but I don't think you could build the game around it like StarCraft 2 is built around ladder. It just wouldn't work as intended. We don't have the player numbers to consistently pair players of equal skill at all times, and it would be a game that would be very hard to consistently rank players (most ELO systems do it by wins, but simply being a good player doesn't guarantee wins in NS2)
There are a couple of points that bother me about matchmaking systems, and some really good benefits, too.
I tried to play the co-op TF2 thing a few weeks back with my mates. There were 4 of us, and after 15 minutes of trying to get a game, we gave up and played planetside instead. How any system with apparently that many servers running can fail to put you in a game is beyond me. GRO is better than this, but still not too great for time spent looking at the screen waiting to play. This is my biggest bugbear with such systems.
With that said, the ability for a bunch of mates who play 2-3 times a week only (we're getting old, having kids, busy jobs, etc etc) to jump into a queue together to guarantee that they'll end up on the same side in a match is great. This is one thing that GRO gets right (although the inability to invite people into your game is rubbish there!). We're not pros, we're casual players who have limited gaming time in a week: some weeks it might be 1 or 2 hours, and others it might be more like 6 - 8. We're not stacking servers to pwn the other team (though we sure as hell play to win and mostly do in pubs (>80% for sure), we're trying to play with our mates, because we love the tactical/teamwork aspects of the game. I for one would MUCH rather be up against a bunch of decent players and have good, well balanced games. But I'd also like to have the opportunity to play marines once in a while - you just can't get in with everyone humping that portal most of the time!
Once the round is ready to start, I'd like to see players/groups of players split up (retaining those groups where possible) to even out the teams based on something very simple:
A combination (with whatever weighting works empirically...) of the following, easy to access metrics: 1) Time spent playing the game 2) Uses a mic? (have a checkbox that players have to tick to say, "yes I use a mic" would be simple - who cares if it's abusable by a few, we're after a crude, fast cheap and effective system here!) 3) desire to command (could include marine vs alien distinction here - again a checkbox for the player to tick would do)
There's no point in overcomplicating a system, trying to include some sort of ranking for players - the simplest and fastest (and cheapest) way is to do it on the fly on the server. It won't be perfect, but no system will be. Trying to divide those non-communicators and newbies out between the teams is in my view the best way to balance out the teams.
Comments
Says who? You can't just handwave "good players are accurate and we can perfectly compensate for support classes and playstyles with a few gimmicks" and expect us to buy it. You have no evidence supporting any sort of global unifying trend amongst "skilled" players beyond your own anecdotes, and every method of measurement of success you expound misses so many major variables (accuracy with different weapons, playstyle, variation in amount of effort put into playing, how different classes treat accuracy to name a few) that they are as useless as selecting random numbers. Additionally, your extreme disregard to these major issues shows how little you care or understand about the issues involved in globally determining something as vague as skill for a highly tactical game like this.
I don't see what's difficult to grasp about the idea that players who have more skill have better aim across the board. They are more familiar with weapon handling and melee mechanics simply by nature of having more time to get used to them. I addressed the concern about differing fire rates and spread patterns in that some weapons may be weighted differently, and quite frankly, the effort put into playing is something no matchmaking system that I know of deals with. If you're going to play in a ranked mode, you are expected to bring your A game, and choosing to bring your C game is going to get you quickly rematched with people who's A game is your C game until such time as you decide to push yourself back up to where you were. This also helps mitigate (not eliminate) skilled players purposefully preying on new or inexperienced players, as they will have to maintain their lower skill level or the system will quickly kick them back up the food chain where they are performing. Play style, unfortunately, isn't really a thing in NS2. This game doesn't have much depth as it stands. Either you're a good player or you're not, there isn't much middle ground so far A supporting player will still need to defend themselves, and thus will contribute to their accuracy rating.
3rd party servers that are possibly running mods means this will never happen
Match making is for games where the company hosts everything and then has access to better statistical tracking where rank and ratings might mean something
Although it would be amusing to think that in a perfect match making system that all the cheaters would be forces to play with each other on one server<!--QuoteEnd--></div><!--QuoteEEnd-->
everyone in grnadmaster is cheating?
<!--quoteo(post=2036857:date=Nov 28 2012, 04:43 PM:name=wiry)--><div class='quotetop'>QUOTE (wiry @ Nov 28 2012, 04:43 PM) <a href="index.php?act=findpost&pid=2036857"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The same ones that are hosting servers now preferably. Also, don't forget that UWE are having a couple of official servers up. There is no reason what so ever why there suddenly would be no servers.
Also, if the public chaos is appealing just mod it, people would still play.<!--QuoteEnd--></div><!--QuoteEEnd-->
maybe you make me a baby? :3
Whenever a new player takes part into gather they are amazed by the depth of gameplay this game has to offer, eventhough they get destroyed the first few times.
<!--quoteo(post=2036865:date=Nov 28 2012, 04:58 PM:name=ScardyBob)--><div class='quotetop'>QUOTE (ScardyBob @ Nov 28 2012, 04:58 PM) <a href="index.php?act=findpost&pid=2036865"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->In my experience, winrates in pubs are not random, but highly related to the skill level of each side (most specifically to how well the marine team can shoot). The best approach I've seen for a NS2 Elo-style ranking system would be:
- Percentage of losing teams summed individual Elo-rating is transferred to the winning team
- The Elo-points won or lost are distributed among the teams by their score*
*This is predicated on the score system being updated such that things such as healing/reparing and assists give points.<!--QuoteEnd--></div><!--QuoteEEnd-->
even simplier, you gain ranking points when you win a round and you lose ranking points if you leave or lose a round based around the amount of ranking points the opposition has. At the beginning everyone would have like 1000 points and the better players should have a positive impact on their team and make them win more often, as they build up points they matched against other players who built around the same amount of points and they should be expected to be around their own skill level.
Also the incentive to gain ranking points by winning a round will motivate players to be less selfish in their play (which is the problem with your system, you gain ranking by paying attention to your points.)
You've obviously never played gorge, and always got that last hit on a cc.
I tried to play the co-op TF2 thing a few weeks back with my mates. There were 4 of us, and after 15 minutes of trying to get a game, we gave up and played planetside instead. How any system with apparently that many servers running can fail to put you in a game is beyond me. GRO is better than this, but still not too great for time spent looking at the screen waiting to play. This is my biggest bugbear with such systems.
With that said, the ability for a bunch of mates who play 2-3 times a week only (we're getting old, having kids, busy jobs, etc etc) to jump into a queue together to guarantee that they'll end up on the same side in a match is great. This is one thing that GRO gets right (although the inability to invite people into your game is rubbish there!).
We're not pros, we're casual players who have limited gaming time in a week: some weeks it might be 1 or 2 hours, and others it might be more like 6 - 8. We're not stacking servers to pwn the other team (though we sure as hell play to win and mostly do in pubs (>80% for sure), we're trying to play with our mates, because we love the tactical/teamwork aspects of the game. I for one would MUCH rather be up against a bunch of decent players and have good, well balanced games. But I'd also like to have the opportunity to play marines once in a while - you just can't get in with everyone humping that portal most of the time!
Once the round is ready to start, I'd like to see players/groups of players split up (retaining those groups where possible) to even out the teams based on something very simple:
A combination (with whatever weighting works empirically...) of the following, easy to access metrics:
1) Time spent playing the game
2) Uses a mic? (have a checkbox that players have to tick to say, "yes I use a mic" would be simple - who cares if it's abusable by a few, we're after a crude, fast cheap and effective system here!)
3) desire to command (could include marine vs alien distinction here - again a checkbox for the player to tick would do)
There's no point in overcomplicating a system, trying to include some sort of ranking for players - the simplest and fastest (and cheapest) way is to do it on the fly on the server. It won't be perfect, but no system will be. Trying to divide those non-communicators and newbies out between the teams is in my view the best way to balance out the teams.
Just my 2p
Roo