<!--quoteo(post=1914868:date=Mar 18 2012, 11:35 PM:name=Talesin)--><div class='quotetop'>QUOTE (Talesin @ Mar 18 2012, 11:35 PM) <a href="index.php?act=findpost&pid=1914868"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Or alternately, if people weren't hanging their performance on an unrelated metric like K/D, their score might improve. Again. Three Marines firing at one Skulk. Only one of the three gets the 'kill' point, while all three were contributing to the kill. There is no 'assist' counter in the flawed and <i>useless</i> K/D system.
If points are applied for most/all useful team-positive acts, it should actually provide a useful measurement, rather than who got luckiest at putting the last bullet into X enemy players, regardless of their upgrades or the quality of the fight. Except of course for the CoD kiddies harping on about K/D, who can't figure out why their score is abysmal when their entire focus is on K/D, instead of helping their team to win the game in ways other than finding things to shoot.
Player 1 kills 10 skulks and dies once. All 10 were last-shots while travelling in a large group. Player 2 kills 3 Skulks, 5 Fades, 2 Lerks and an Onos, and scares off a raiding party on a minibase, welds up a few RTs and follows the Comm's orders to scout out an area. All singlehanded. She dies 14 times over the course of the game.
K/D says that player 2 sucks, and player 1 waggles his K/D around like a typical pre-adolescent. When anyone with half a brain can see that player 1 was terribad, and player 2 was an incredibly useful asset to her team. K/D is WORTHLESS.<!--QuoteEnd--></div><!--QuoteEEnd-->
You basically just described the problem with any scoring system. Obviously score needs to be taken with a grain of salt. Ultimately the end result (W or L) is what matters from a 'skill' perspective. Any time you try and quantify a player's contribution toward winning, you're going to run into problems like what you describe in your post. You're not wrong, you're just missing the point. The problems you laid out aren't about KDRs but about scoring in general.
There is no completely accurate scoring method. Anyone trying to prove that something is the be-all-end-all has quite the burden of proof. I'm not sure that IE was trying to prove that KDR alone wins games, he was just posting that it is a metric which points out trends. High KDR teams are likely to be winning. Low KDR teams are likely to be losing.
<!--quoteo(post=1914875:date=Mar 18 2012, 11:55 PM:name=Wheeee)--><div class='quotetop'>QUOTE (Wheeee @ Mar 18 2012, 11:55 PM) <a href="index.php?act=findpost&pid=1914875"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->NS and NS2 aren't really games where you can project KDR into wins and losses. While it can be a decent general measure of player effectiveness and skill, especially at the extremes (a player with a 10:1 kdr likely knows how to play the game very well and is executing a lot of high level strategy as well as clearing the map of hostiles, whereas a player with 1:10 kdr is likely clogging up the spawn queue so much and is so ineffective in the field that it equates to the team being down a player), there are plenty of other factors to consider: a skulk who goes around chomping down extractors the whole game might have a worse KDR than his skulk buddy who solely ambushes rambo marines, but which one is the more effective in assisting the team win? It's difficult to measure what the true impact of KDR is other than in generalities. A skulk rush might be successful even if 3 skulks die to 2 marines, if they manage to kill some structures as well. A solo skulk might be effective in delaying a push by marines even if he dies without getting any kills, simply by projecting his presence near the marines. There's way too many factors involved in the game to analyze KDR in a vacuum.<!--QuoteEnd--></div><!--QuoteEEnd-->
Yep. I think this decently sums up the value and limits of analyzing KDR for wins/losses.
<!--quoteo(post=1914875:date=Mar 18 2012, 11:55 PM:name=Wheeee)--><div class='quotetop'>QUOTE (Wheeee @ Mar 18 2012, 11:55 PM) <a href="index.php?act=findpost&pid=1914875"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->There's way too many factors involved in the game to analyze KDR in a vacuum.<!--QuoteEnd--></div><!--QuoteEEnd-->
Right, that's why I didn't analyze KDR in a vacuum. I tied it to<i> everything that players do</i>. I also discussed how 'aggregate KDR' applies to teams. If you can show me a game where a team with an incredibly bad KDR wins, I'll show you an opposing team that <b>didn't make use of</b> their huge KDR. Like I said in the OP, it's not a deathmatch game where getting to a certain kill count wins the game. However, every kill is an opportunity to take map control, spend enemy resources and so on.
For instance, I can walk in a pub where the aliens just got to 3 hives. I can morph onos, see 2 of our hives die and then go kill the chair with smash and win the game. Chances are the marines' KDR over the last few minutes of the game will look like they're winning, but they didn't counterbalance their big increase in map control with defensive presence. I do this a lot in pubs, because people are bad. That doesn't mean KDR isn't useful.
It's the same thing every time someone bites down a chair in a pub server full of confused new players, and thinks they're a hero. If the opponents just killed you every time you ran headlong at the chair, you would be dead weight the entire game. The 'KDR changes' that come with killing you would equate to raw gains for the enemy team.
This game is still young, so it's easy for some people to think the different player tasks are not related. You can win a game without being conscious of your KDR efficiency, but that won't hold up at higher levels of play, because that isn't the real character of the game.
<!--quoteo(post=1914871:date=Mar 18 2012, 11:47 PM:name=IeptBarakat)--><div class='quotetop'>QUOTE (IeptBarakat @ Mar 18 2012, 11:47 PM) <a href="index.php?act=findpost&pid=1914871"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I tend to favor lemming tactics such as being a distraction or die in the process killing or at least mostly killing a hive. There are also people who suicide just to get to the other side of the map faster. Which is one of the reasons why games like tribes ascend removed the death count, as it didn't really matter when people started killing themselves tactically.<!--QuoteEnd--></div><!--QuoteEEnd-->
It's great that you are able to be effective like this, but should everyone to play in these weird niche roles? Of course not. There's a place for every playstyle, but the 'lemming' you've described cannot be applied to all players. If a new player comes to this forum and wants to learn "the gist" of how to play NS2, I want them to walk away thinking about how they can shoot some aliens and bite some marines. The reason you are able to play this way is because you understand that it's not the norm, and that it's a piece of a larger machine (the team with other players fighting on the main front, shooting the things you are distracting).
TalesinOur own little well of hateJoin Date: 2002-11-08Member: 7710NS1 Playtester, Forum Moderators
<!--quoteo(post=1914873:date=Mar 18 2012, 09:48 PM:name=GORGEous)--><div class='quotetop'>QUOTE (GORGEous @ Mar 18 2012, 09:48 PM) <a href="index.php?act=findpost&pid=1914873"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I agree that KDR doesn't win games by its self, but that doesn't make it useless. I'm not sure how you are holding these two contradictory thoughts at the same time. You're basically saying a) dying is a bad and b) killing is meaningless. But a kill is about equal and opposite to a death. When you die (negative), I get a kill (thus a positive).
The only way to rationalize that KDR is meaningless is to claim that both kills and deaths are meaningless. That's nonsense.
If you're just trying to say that KDR isn't the whole story or the biggest factor in winning, then you would be correct in my opinion. But to say KDR, kills, and deaths are meaningless is just ridiculous.<!--QuoteEnd--></div><!--QuoteEEnd--> Honestly, they are meaningless, as a tally-count. You can have stupid deaths, and you can have useful deaths. Jumping off into a bottomless pit, or diving through the defenders/static-D to get the last shot on a Hive both count as one death. One is pointless, the other is useful. K/D doesn't take either of those into account. As said earlier, you could be part of a 10-Marine group and be the lucky one to get the last shot on a Skulk everyone is firing at, or you could heroically kill the Onos or Fade that's wrecking your base singlehandedly. Both count as one kill.
The system is, again, entirely <i>worthless</i> as a real performance metric. The real/only metric to pay attention to (until an individual player scoring method is introduced which has a real bearing on player performance) is which team wins the match.
So should players not strive for a higher KDR when they play, because of the caveats you mentioned? This isn't a theory of everything - I'm explaining one of many ways to look at the game, and its benefits. You are explaining its shortcomings (which is fine and necessary too), but that's all we're doing. There's still plenty of use for players thinking about KDR when they play.
You could write up a thread explaining your way of analyzing the game, and it would probably be good. The main thing here is to discourage people wandering around getting mad about halo or whatever.
<!--quoteo(post=1914883:date=Mar 18 2012, 10:09 PM:name=internetexplorer)--><div class='quotetop'>QUOTE (internetexplorer @ Mar 18 2012, 10:09 PM) <a href="index.php?act=findpost&pid=1914883"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->So should players not strive for a higher KDR when they play, because of the caveats you mentioned? This isn't a theory of everything - I'm explaining one of many ways to look at the game, and its benefits. You are explaining its shortcomings (which is fine and necessary too), but that's all we're doing. There's still plenty of use for players thinking about KDR when they play.<!--QuoteEnd--></div><!--QuoteEEnd--> Would you like to see PRES for kills then? What do you propose?
A proper scoring system will better indicate player performance, but it's development would take time away from more pressing matters. Lets get the game working first before we start such a project.
If it really does bother you, start thinking about all the ways scoring can be done, and develop a framework for the devs to work from eventually.
For now, that kick ass marine that saved his base from a pushing force 10x in a row now, but the aliens always backed off before dying, will just be considered a nub by some.
IeptBarakatThe most difficult name to speak ingame.Join Date: 2009-07-10Member: 68107Members, Constellation, NS2 Playtester, Squad Five Blue, NS2 Map Tester, Reinforced - Diamond, Reinforced - Shadow
Also comes down to who you regard as the important player. The 2 players killing all the aliens in alien start, or the player killing the hive with the axe. Obviously the two players who're killing all the aliens are going to have a sweet k:d ratio. But they're only stopping the aliens. The guy with the axe killing the hive is going to win the game for them, but he only gets a score increase for killing a structure. They're all equal in contributing to the objective, but the one player who is actually completing the objective is the player with the lowest k:d ratio.
<!--quoteo(post=1914878:date=Mar 19 2012, 12:00 AM:name=internetexplorer)--><div class='quotetop'>QUOTE (internetexplorer @ Mar 19 2012, 12:00 AM) <a href="index.php?act=findpost&pid=1914878"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Right, that's why I didn't analyze KDR in a vacuum. I tied it to<i> everything that players do</i>. I also discussed how 'aggregate KDR' applies to teams. If you can show me a game where a team with an incredibly bad KDR wins, I'll show you an opposing team that <b>didn't make use of</b> their huge KDR. Like I said in the OP, it's not a deathmatch game where getting to a certain kill count wins the game. However, every kill is an opportunity to take map control, spend enemy resources and so on.<!--QuoteEnd--></div><!--QuoteEEnd--> theoretically, yes. However, because of the way the asymmetry of the game works e.g. marines' power scaling up exponentially when traveling in groups, aliens being able to cover a lot of ground quickly, people not having psychic abilities to know exactly where everything is at the moment, many of these opportunities are meaningless.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->For instance, I can walk in a pub where the aliens just got to 3 hives. I can morph onos, see 2 of our hives die and then go kill the chair with smash and win the game. Chances are the marines' KDR over the last few minutes of the game will look like they're winning, but they didn't counterbalance their big increase in map control with defensive presence. I do this a lot in pubs, because people are bad. That doesn't mean KDR isn't useful.
It's the same thing every time someone bites down a chair in a pub server full of confused new players, and thinks they're a hero. If the opponents just killed you every time you ran headlong at the chair, you would be dead weight the entire game. The 'KDR changes' that come with killing you would equate to raw gains for the enemy team.<!--QuoteEnd--></div><!--QuoteEEnd-->
not necessarily true; a solo skulk dealing substantial damage to the chair can cause the marines to play more defensively. A lone marine running in the hive and knifing it down to 50% might force a gorge to withdraw from the front, or a skulk to go gorge, reducing the aliens' opportunity to have an attack class in the field. Even if you die 5 times, you may force a change of strategy, a slowdown in tech while marines research welders/mines (in case they didn't already), or force the khamm to place a whip or crag... your respawns are free, but those things aren't. And delaying tech means that fades come out before armor 2, or jetpacks come out before the aliens are ready to deal with them...etc.
Of course it's a no-brainer to say "killing is better than dying" but it really is very limited in what you can say about it. I can have, for example, 5-10 for the first 5 minutes of the game, and then solo a fade. I have a ###### KDR but that single kill i just contributed to the team did far more than anything else. Could the marines be in a better position if I hadn't died? maybe. But let's say, I have a 10-5 ratio instead, but the comm had to medspam me 10 medpacks and 5 ammo packs to achieve this. The cost to the team is significant, yet my KDR shows that i'm supposedly a good player. I could have a decent ratio going LMG the entire game, which could be improved if I bought shotties all the time, but then my teammates wouldn't be getting overflow res, and perhaps their skill with a shotty is better than mine. So I take the hit in KDR to improve their abilities. I've seen fades who try to get their KDR as high as possible, killing a marine and then retreating for a minute to heal up. They are 25-0 as a fade but serve no real purpose in projecting map control because they play like they're made out of glass, and are just generally ineffective.
Just wanted to say we have been watching our scores since the days of Pong. Calling K/D Ratios a Cod thing or a cs thing is just dumb. Besides, a dead army is a useless army. This can be applied to NS2; no ip or eggs means no soldier to fight back the enemy.
Talking about Pong, has anyone watched the 'long play' videos where elite guys walkthrough non stop old amiga games? Theres like 200+ of them, all about 2 hours long. The one for 'GODS' is truly awsome, same for 'The Adams Family'.
In team play/competitive gaming environment K:D is irrelevant simply because there are so many factors determining who got the kill and and who died. Go watch some old cs/ns matches and I'll guarantee even the best players sometimes have low K:D. If you're working as a team, some people are going to get the kill and some are going to die. Just because someone has a low ratio does not mean they're playing bad.
Worrying about your ratio is a solo player type mentality only useful in public team deathmatch games were killing the other players is the ultimate goal. Stop worrying about your ratio and focus on playing as a team. Its especially useful in NS.
Quick Scenario: Your score is 5-1 and you're chomping on a resource node that's almost dead. Marines are on the way and you'll most likely die if you don't run now. Do you stay and kill it knowing you'll die or run ? The player worrying about his ratio will run. The one playing for the team kills the node and dies. - Protip: Be the player that kills the node.
I think it's possible to be conscious of concepts like KDR without being a bullheaded idiot who can only follow one line of thinking. I wish the same were true of posting on forums!
<!--quoteo(post=1914903:date=Mar 18 2012, 10:49 PM:name=internetexplorer)--><div class='quotetop'>QUOTE (internetexplorer @ Mar 18 2012, 10:49 PM) <a href="index.php?act=findpost&pid=1914903"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I think it's possible to be conscious of concepts like KDR without being a bullheaded idiot who can only follow one line of thinking. I wish the same were true of posting on forums!<!--QuoteEnd--></div><!--QuoteEEnd-->
Then why are you still posting? You whole post is focused on linear thinking. Obviously I made a valid point in objection to your statement that K:D matters and since you can't respond with anything sensible, you resort to attacking. Do us all a favor and just stop playing any team oriented games. Thanks.
I find this whole thread irrelevant. KDR does not determine a games outcome. All it takes is a chair/hive to be defenseless and a distracted comm to instantly end a match due to the ninjaing of a chair/hive. I've seen games where aliens were losing horribly and they still won cause a few skulks gathered up at the right moment to hit the CC and take it down. KDR is irrelevant.
Also, in all my years of NS playing (been playing since 10/31/02) I've never once heard of someone complaining about anything related to KDR.
<!--quoteo(post=1914909:date=Mar 19 2012, 02:08 AM:name=MetalGarret)--><div class='quotetop'>QUOTE (MetalGarret @ Mar 19 2012, 02:08 AM) <a href="index.php?act=findpost&pid=1914909"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I find this whole thread irrelevant. KDR does not determine a games outcome. All it takes is a chair/hive to be defenseless and a distracted comm to instantly end a match due to the ninjaing of a chair/hive. I've seen games where aliens were losing horribly and they still won cause a few skulks gathered up at the right moment to hit the CC and take it down. KDR is irrelevant.
Also, in all my years of NS playing (been playing since 10/31/02) I've never once heard of someone complaining about anything related to KDR.<!--QuoteEnd--></div><!--QuoteEEnd-->
While that is true that it can happen, KDR still tends to show who is winning.
If one team is going 10K per 1 D, they're probably winning the game, right?
Obviously I'm not saying they will win the game, but chances are that they will. That's the point of KDR. If you use it as a theory with absolute truth value, it will fail you. But if you consider it as part of the puzzle which can give you insight into who is winning or losing, then you're doing it right.
I gotta say, this was a well put together article. I agree that the kill/death ratio is an important factor in natural selection 2. Especially since it's quite hard to maintain a high KDR in ns2. One example of this is it can help lesser skilled players identify the skilled players and support that skilled player to attack enemy positions more effectively.
You state that the most important reason to kill enemies is to get the opposing team to spend resources. This is 100% true. However, there are better ways to cost the enemy team resources, and so it follows that there are more important things than KDR. NS2 is a game of priorities, and since there are better ways to do something, <b>killing</b> is almost always a very low priority.
What you have essentially stated is that keeping all other variables constant, a KDR of 0.9 is better than a KDR of 0.8, which is fine. However, it says absolutely nothing about a game such as natural-selection.
<!--quoteo(post=1914915:date=Mar 19 2012, 08:44 AM:name=GORGEous)--><div class='quotetop'>QUOTE (GORGEous @ Mar 19 2012, 08:44 AM) <a href="index.php?act=findpost&pid=1914915"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->While that is true that it can happen, KDR still tends to show who is winning.
If one team is going 10K per 1 D, they're probably winning the game, right?
Obviously I'm not saying they will win the game, but chances are that they will. That's the point of KDR. If you use it as a theory with absolute truth value, it will fail you. But if you consider it as part of the puzzle which can give you insight into who is winning or losing, then you're doing it right.<!--QuoteEnd--></div><!--QuoteEEnd-->
This is another interesting point. KDR can be a good indicator of how a game is going in symetric games and games which are strictly shooters. Take COD for example, I have a gun, you have a gun, pew pew I die. Guess you are better with a gun. Natural selection on the other hand; you have a gun, I have nomnom teeth, pew pew, I die. What does that mean? Are guns currently better than teeth? Are you defending an area which better suits range over melee? Do you have more upgrades than me? The kill tells you absolutely nothing about anything. Furthermore, strategic shooters and the related KDRs are largely affected by what has come before. That is to say, even in a game that is kind of "strategic" like battlefield 3, you could control the entire map, but when you and me come face to face with our weapons, the playing field is essentially balanced. Natural selection is in no way like this. When one team has a massive KD advantage over another, its usually because the losing team has nothing, and has been in this situation for some time.
What this means is that the other more important strategical aspects affect how well a team is doing, and this affects KDR, <b>not the other way around</b>.
What if you have 1-0 k/d. Your k/d is approaching infinity so you must obviously be the best player in the world! :p
I enjoy getting a good k/d. Its one useful measure for how much map pressure your putting on the other team (due to egg and marine mobility limits) and it makes me feel happy that i am being useful in that way, and i like feeling happy. Any experienced player however knows theres alot more than kdr though..
The biggest travesty (as the OP has done) is when you start creating systems which relate player 1's k/d to player 2's k/d and draw bad conclusions such as player 1 is better than player 2. K/d is personal vanity and is only really useful as a measure of usefulness if your playing a pressure role.
ArgathorJoin Date: 2011-07-18Member: 110942Members, Squad Five Blue
This thread is one huge mess of valid points.
It is obvious (I would think) to everyone that some information and value can be gained from viewing a players/teams KDR. Also, as is obvious, you do not get the full story.
As was clear from the original post, you cannot simply ignore KDR and pretend that only egotistical prats think KDR is important. You can use it to make useful judgements about the game/player, yet it will not tell you everything. There really is no major controversy here.
<!--quoteo(post=1914954:date=Mar 19 2012, 01:37 PM:name=Argathor)--><div class='quotetop'>QUOTE (Argathor @ Mar 19 2012, 01:37 PM) <a href="index.php?act=findpost&pid=1914954"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->This thread is one huge mess of valid points.
It is obvious (I would think) to everyone that some information and value can be gained from viewing a players/teams KDR. Also, as is obvious, you do not get the full story.
As was clear from the original post, you cannot simply ignore KDR and pretend that only egotistical prats think KDR is important. You can use it to make useful judgements about the game/player, yet it will not tell you everything. There really is no major controversy here.<!--QuoteEnd--></div><!--QuoteEEnd-->
I think it is only egotistical prats that think it is important.
Wanting to kill things is fine. The enemy needs to be killed sometimes, and I'm sure you are just the person to do it. However, the problem starts when thinking that killing the enemy is <u>important</u>, and even more dangerously, when you start thinking that the ratio of kills to <b>deaths</b> is important. No player should ever be sat thinking "I really need to go in there, but then I will die and my KD will go down". Similarly, no player should ever be sat thinking "I could attack that res tower with my teammate, or I could sit over here on this wall and wait for marines to come". Those approaches are conducive to KD but bad for the team.
ArgathorJoin Date: 2011-07-18Member: 110942Members, Squad Five Blue
edited March 2012
<!--quoteo(post=1914960:date=Mar 19 2012, 11:56 AM:name=Imbalanxd)--><div class='quotetop'>QUOTE (Imbalanxd @ Mar 19 2012, 11:56 AM) <a href="index.php?act=findpost&pid=1914960"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Wanting to kill things is fine. The enemy needs to be killed sometimes, and I'm sure you are just the person to do it. However, the problem starts when thinking that killing the enemy is <u>important</u>...<!--QuoteEnd--></div><!--QuoteEEnd--> Are you seriously attempting to argue that killing the other team, in an FPS designed around killing the other team, is <b>not important</b>?
This is such a fruitless arguement. Killing will always be important in NS2 and so will other tasks (like defending certain hallways, repairing, building, scouting, etc).
I play clan matches with a bad KDR on purpose. I play Skulk (and preferably stay skulk) aggressively and often carry a large amount of deaths. I do this because I understand that <b>other things</b> are more important. I focus on key enemy players, stopping momentum, forcing defensive play, killing important structures, etc.
However, I still realise that KDR can give you information about how the game is going. NS2 is based around map control, momentum and raiding. You need (in all but the odd strategy, all lerks, etc) to kill, or not be killed by, the other team to do these things. Kills and deaths will always be an important aspect in NS2, as will other things brought into NS2 from the FPS and RTS elements.
<i>edit - It is not a question about whether you are worried about going into a room as you might die, its about the fact that when you do go into the room it is important that you win the encounter (whether or not you kill the opposition or die is important).</i>
If a player can get a good k/d it probably means that whatever he can do it without dieing. This is very efficent since it saves res, gives more map control and does not slow down the spawn cue. This can be number of things, such as: -Capping rts/biting down rts -getting a forward base -scouting while picking out solo players -defending a key potition -putting pressure
However you can get good k/d by doing worthless things like staying in base all game do get 3-4 kills without dieing. But even so he is doing it very efficently since he can defend the base without allowing the skulk to take anything down.
You should try to suicide as rarly as possible, there are timings in the game where it can be effective but most of the time you should focus on getting your task done without dieing.
--- The strategy/tactic used is important in game but you need players capeable of doing they part for it to be effective. If you want a skulk to keep pressuring rts its important he is able to stay alive, if he can pick out the marines who try the save the rt he is already being much more efficent than just suiciting 1 rt down. This is the same with marines, if you want a marine to cap the other side of the map he must be capeable of defending himself if a skulk arrives. Even though he might only get 0-2 kills (since he is not in the main fight) he is doing something very efficent while he stays alive.
Its bit similar to starcraft, it doesnt matter what strategy/tactic you use, if you are not good enough to do it efficiently you will lose vs a better player. It does not mean that tactics are useless. -- You can get these statistics from K/D, If players are doing whats important it can tell you how well they are peforming that task. But other factors are also important.
It makes me laugh when people go on and on about KD ratios, seriously who cares except those who like to camp and waste time trying to be pro, even in something a bit more competitive like cs or cod i don't care, all that matters is overall points and to some extent kills, as long as im at the top when the round ends that's all that matters! :)
There is still a hidden strategy portion outside of KD ratios..
I've seen games won by the team with lower KD overall (by quite a bit)
The fact that PRES is not given per kill anymore also reduces the effectiveness of getting kills.
OP has addressed this slightly from this post but if all you do is camp for kills while leaving the other teams res nodes alive then they can still easily come back... etc, etc
in the end its still all about the res nodes (and hive 2 :p)
hey, internetexplorer it may have been you I was thinking of :) nice thread, btw.
i'm kind of with elodea on this one. it is a useful stat for your front line guys - those wins in the corridor match-ups are huge.
but there are less aggressive (& yes less talented) players that we need to encourage to continue to play, so the scoring system should give them a sense of accomplishment as well. we don't want anyone looking at their KDR & say 'I s*ck' (or worse be told 'you s*ck') & quit playing the game.
i like the assist idea - maybe there can be some scoring enhancements around groups
NS2 is probably <u><b>THE WORST</b></u> game you could possibly associate KDR with, so much so that it is f*cking stupid to even post a topic about it (no offense).
I have been in at least 4 different games where our team has come back and won from circumstances that we should not have (such as being a marine against 3 hives, or being an alien against marines who control the majority of the map.)
One of the main reasons we won this is because everyone on the team was willing to give up their life for the cause, they could go -30 and it would not matter as long as we won which we did. When you look at players who thrive on their KDR they become cowards afraid to die ingame, which leads to them doing nothing to turtling and base which leads to the opposing team having free unopposed reign on the map.
tl;dr Version
If i see some one on the team who has a +30 KDR but isnt helping the team to do ###### other than kill and i see some one with a -30 KDR who is doing everything in their power to help the team win guess who is the noob? That's right, the self important KDR loving ######.
<!--quoteo(post=1915012:date=Mar 19 2012, 09:05 AM:name=Xostean)--><div class='quotetop'>QUOTE (Xostean @ Mar 19 2012, 09:05 AM) <a href="index.php?act=findpost&pid=1915012"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->NS2 is probably <u><b>THE WORST</b></u> game you could possibly associate KDR with, so much so that it is f*cking stupid to even post a topic about it (no offense).<!--QuoteEnd--></div><!--QuoteEEnd-->
Indeed but i would actually go further than that and say KDR does even matter in any game where overall score or kills alone places a player at the top winning position, for example in MW3 not only does the biggest score get top position but the more points you get per round the quicker you rank up, my KDR can be pretty bad by some peoples standards yet more often then not ill be the player with the most points, to me deaths are meaningless when you can respawn a few seconds later, there's little downtime and zero cost so over the course of a game i can be out there running around scoring more kills and points while someone playing more carefully doesn't.
Comments
If points are applied for most/all useful team-positive acts, it should actually provide a useful measurement, rather than who got luckiest at putting the last bullet into X enemy players, regardless of their upgrades or the quality of the fight. Except of course for the CoD kiddies harping on about K/D, who can't figure out why their score is abysmal when their entire focus is on K/D, instead of helping their team to win the game in ways other than finding things to shoot.
Player 1 kills 10 skulks and dies once. All 10 were last-shots while travelling in a large group.
Player 2 kills 3 Skulks, 5 Fades, 2 Lerks and an Onos, and scares off a raiding party on a minibase, welds up a few RTs and follows the Comm's orders to scout out an area. All singlehanded. She dies 14 times over the course of the game.
K/D says that player 2 sucks, and player 1 waggles his K/D around like a typical pre-adolescent. When anyone with half a brain can see that player 1 was terribad, and player 2 was an incredibly useful asset to her team. K/D is WORTHLESS.<!--QuoteEnd--></div><!--QuoteEEnd-->
You basically just described the problem with any scoring system. Obviously score needs to be taken with a grain of salt. Ultimately the end result (W or L) is what matters from a 'skill' perspective. Any time you try and quantify a player's contribution toward winning, you're going to run into problems like what you describe in your post. You're not wrong, you're just missing the point. The problems you laid out aren't about KDRs but about scoring in general.
There is no completely accurate scoring method. Anyone trying to prove that something is the be-all-end-all has quite the burden of proof. I'm not sure that IE was trying to prove that KDR alone wins games, he was just posting that it is a metric which points out trends. High KDR teams are likely to be winning. Low KDR teams are likely to be losing.
Yep. I think this decently sums up the value and limits of analyzing KDR for wins/losses.
Right, that's why I didn't analyze KDR in a vacuum. I tied it to<i> everything that players do</i>. I also discussed how 'aggregate KDR' applies to teams. If you can show me a game where a team with an incredibly bad KDR wins, I'll show you an opposing team that <b>didn't make use of</b> their huge KDR. Like I said in the OP, it's not a deathmatch game where getting to a certain kill count wins the game. However, every kill is an opportunity to take map control, spend enemy resources and so on.
For instance, I can walk in a pub where the aliens just got to 3 hives. I can morph onos, see 2 of our hives die and then go kill the chair with smash and win the game. Chances are the marines' KDR over the last few minutes of the game will look like they're winning, but they didn't counterbalance their big increase in map control with defensive presence. I do this a lot in pubs, because people are bad. That doesn't mean KDR isn't useful.
It's the same thing every time someone bites down a chair in a pub server full of confused new players, and thinks they're a hero. If the opponents just killed you every time you ran headlong at the chair, you would be dead weight the entire game. The 'KDR changes' that come with killing you would equate to raw gains for the enemy team.
This game is still young, so it's easy for some people to think the different player tasks are not related. You can win a game without being conscious of your KDR efficiency, but that won't hold up at higher levels of play, because that isn't the real character of the game.
<!--quoteo(post=1914871:date=Mar 18 2012, 11:47 PM:name=IeptBarakat)--><div class='quotetop'>QUOTE (IeptBarakat @ Mar 18 2012, 11:47 PM) <a href="index.php?act=findpost&pid=1914871"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I tend to favor lemming tactics such as being a distraction or die in the process killing or at least mostly killing a hive. There are also people who suicide just to get to the other side of the map faster. Which is one of the reasons why games like tribes ascend removed the death count, as it didn't really matter when people started killing themselves tactically.<!--QuoteEnd--></div><!--QuoteEEnd-->
It's great that you are able to be effective like this, but should everyone to play in these weird niche roles? Of course not. There's a place for every playstyle, but the 'lemming' you've described cannot be applied to all players. If a new player comes to this forum and wants to learn "the gist" of how to play NS2, I want them to walk away thinking about how they can shoot some aliens and bite some marines. The reason you are able to play this way is because you understand that it's not the norm, and that it's a piece of a larger machine (the team with other players fighting on the main front, shooting the things you are distracting).
The only way to rationalize that KDR is meaningless is to claim that both kills and deaths are meaningless. That's nonsense.
If you're just trying to say that KDR isn't the whole story or the biggest factor in winning, then you would be correct in my opinion. But to say KDR, kills, and deaths are meaningless is just ridiculous.<!--QuoteEnd--></div><!--QuoteEEnd-->
Honestly, they are meaningless, as a tally-count. You can have stupid deaths, and you can have useful deaths. Jumping off into a bottomless pit, or diving through the defenders/static-D to get the last shot on a Hive both count as one death. One is pointless, the other is useful. K/D doesn't take either of those into account.
As said earlier, you could be part of a 10-Marine group and be the lucky one to get the last shot on a Skulk everyone is firing at, or you could heroically kill the Onos or Fade that's wrecking your base singlehandedly. Both count as one kill.
The system is, again, entirely <i>worthless</i> as a real performance metric.
The real/only metric to pay attention to (until an individual player scoring method is introduced which has a real bearing on player performance) is which team wins the match.
You could write up a thread explaining your way of analyzing the game, and it would probably be good. The main thing here is to discourage people wandering around getting mad about halo or whatever.
Would you like to see PRES for kills then? What do you propose?
If it really does bother you, start thinking about all the ways scoring can be done, and develop a framework for the devs to work from eventually.
For now, that kick ass marine that saved his base from a pushing force 10x in a row now, but the aliens always backed off before dying, will just be considered a nub by some.
theoretically, yes. However, because of the way the asymmetry of the game works e.g. marines' power scaling up exponentially when traveling in groups, aliens being able to cover a lot of ground quickly, people not having psychic abilities to know exactly where everything is at the moment, many of these opportunities are meaningless.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->For instance, I can walk in a pub where the aliens just got to 3 hives. I can morph onos, see 2 of our hives die and then go kill the chair with smash and win the game. Chances are the marines' KDR over the last few minutes of the game will look like they're winning, but they didn't counterbalance their big increase in map control with defensive presence. I do this a lot in pubs, because people are bad. That doesn't mean KDR isn't useful.
It's the same thing every time someone bites down a chair in a pub server full of confused new players, and thinks they're a hero. If the opponents just killed you every time you ran headlong at the chair, you would be dead weight the entire game. The 'KDR changes' that come with killing you would equate to raw gains for the enemy team.<!--QuoteEnd--></div><!--QuoteEEnd-->
not necessarily true; a solo skulk dealing substantial damage to the chair can cause the marines to play more defensively. A lone marine running in the hive and knifing it down to 50% might force a gorge to withdraw from the front, or a skulk to go gorge, reducing the aliens' opportunity to have an attack class in the field. Even if you die 5 times, you may force a change of strategy, a slowdown in tech while marines research welders/mines (in case they didn't already), or force the khamm to place a whip or crag... your respawns are free, but those things aren't. And delaying tech means that fades come out before armor 2, or jetpacks come out before the aliens are ready to deal with them...etc.
Of course it's a no-brainer to say "killing is better than dying" but it really is very limited in what you can say about it. I can have, for example, 5-10 for the first 5 minutes of the game, and then solo a fade. I have a ###### KDR but that single kill i just contributed to the team did far more than anything else. Could the marines be in a better position if I hadn't died? maybe. But let's say, I have a 10-5 ratio instead, but the comm had to medspam me 10 medpacks and 5 ammo packs to achieve this. The cost to the team is significant, yet my KDR shows that i'm supposedly a good player. I could have a decent ratio going LMG the entire game, which could be improved if I bought shotties all the time, but then my teammates wouldn't be getting overflow res, and perhaps their skill with a shotty is better than mine. So I take the hit in KDR to improve their abilities. I've seen fades who try to get their KDR as high as possible, killing a marine and then retreating for a minute to heal up. They are 25-0 as a fade but serve no real purpose in projecting map control because they play like they're made out of glass, and are just generally ineffective.
Worrying about your ratio is a solo player type mentality only useful in public team deathmatch games were killing the other players is the ultimate goal. Stop worrying about your ratio and focus on playing as a team. Its especially useful in NS.
Quick Scenario: Your score is 5-1 and you're chomping on a resource node that's almost dead. Marines are on the way and you'll most likely die if you don't run now. Do you stay and kill it knowing you'll die or run ? The player worrying about his ratio will run. The one playing for the team kills the node and dies. - Protip: Be the player that kills the node.
Then why are you still posting? You whole post is focused on linear thinking. Obviously I made a valid point in objection to your statement that K:D matters and since you can't respond with anything sensible, you resort to attacking. Do us all a favor and just stop playing any team oriented games. Thanks.
Also, in all my years of NS playing (been playing since 10/31/02) I've never once heard of someone complaining about anything related to KDR.
Also, in all my years of NS playing (been playing since 10/31/02) I've never once heard of someone complaining about anything related to KDR.<!--QuoteEnd--></div><!--QuoteEEnd-->
While that is true that it can happen, KDR still tends to show who is winning.
If one team is going 10K per 1 D, they're probably winning the game, right?
Obviously I'm not saying they will win the game, but chances are that they will. That's the point of KDR. If you use it as a theory with absolute truth value, it will fail you. But if you consider it as part of the puzzle which can give you insight into who is winning or losing, then you're doing it right.
What you have essentially stated is that keeping all other variables constant, a KDR of 0.9 is better than a KDR of 0.8, which is fine. However, it says absolutely nothing about a game such as natural-selection.
<!--quoteo(post=1914915:date=Mar 19 2012, 08:44 AM:name=GORGEous)--><div class='quotetop'>QUOTE (GORGEous @ Mar 19 2012, 08:44 AM) <a href="index.php?act=findpost&pid=1914915"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->While that is true that it can happen, KDR still tends to show who is winning.
If one team is going 10K per 1 D, they're probably winning the game, right?
Obviously I'm not saying they will win the game, but chances are that they will. That's the point of KDR. If you use it as a theory with absolute truth value, it will fail you. But if you consider it as part of the puzzle which can give you insight into who is winning or losing, then you're doing it right.<!--QuoteEnd--></div><!--QuoteEEnd-->
This is another interesting point. KDR can be a good indicator of how a game is going in symetric games and games which are strictly shooters. Take COD for example, I have a gun, you have a gun, pew pew I die. Guess you are better with a gun. Natural selection on the other hand; you have a gun, I have nomnom teeth, pew pew, I die. What does that mean? Are guns currently better than teeth? Are you defending an area which better suits range over melee? Do you have more upgrades than me? The kill tells you absolutely nothing about anything. Furthermore, strategic shooters and the related KDRs are largely affected by what has come before. That is to say, even in a game that is kind of "strategic" like battlefield 3, you could control the entire map, but when you and me come face to face with our weapons, the playing field is essentially balanced. Natural selection is in no way like this. When one team has a massive KD advantage over another, its usually because the losing team has nothing, and has been in this situation for some time.
What this means is that the other more important strategical aspects affect how well a team is doing, and this affects KDR, <b>not the other way around</b>.
I enjoy getting a good k/d. Its one useful measure for how much map pressure your putting on the other team (due to egg and marine mobility limits) and it makes me feel happy that i am being useful in that way, and i like feeling happy. Any experienced player however knows theres alot more than kdr though..
The biggest travesty (as the OP has done) is when you start creating systems which relate player 1's k/d to player 2's k/d and draw bad conclusions such as player 1 is better than player 2. K/d is personal vanity and is only really useful as a measure of usefulness if your playing a pressure role.
It is obvious (I would think) to everyone that some information and value can be gained from viewing a players/teams KDR. Also, as is obvious, you do not get the full story.
As was clear from the original post, you cannot simply ignore KDR and pretend that only egotistical prats think KDR is important. You can use it to make useful judgements about the game/player, yet it will not tell you everything. There really is no major controversy here.
It is obvious (I would think) to everyone that some information and value can be gained from viewing a players/teams KDR. Also, as is obvious, you do not get the full story.
As was clear from the original post, you cannot simply ignore KDR and pretend that only egotistical prats think KDR is important. You can use it to make useful judgements about the game/player, yet it will not tell you everything. There really is no major controversy here.<!--QuoteEnd--></div><!--QuoteEEnd-->
I think it is only egotistical prats that think it is important.
Wanting to kill things is fine. The enemy needs to be killed sometimes, and I'm sure you are just the person to do it. However, the problem starts when thinking that killing the enemy is <u>important</u>, and even more dangerously, when you start thinking that the ratio of kills to <b>deaths</b> is important. No player should ever be sat thinking "I really need to go in there, but then I will die and my KD will go down". Similarly, no player should ever be sat thinking "I could attack that res tower with my teammate, or I could sit over here on this wall and wait for marines to come". Those approaches are conducive to KD but bad for the team.
Are you seriously attempting to argue that killing the other team, in an FPS designed around killing the other team, is <b>not important</b>?
This is such a fruitless arguement. Killing will always be important in NS2 and so will other tasks (like defending certain hallways, repairing, building, scouting, etc).
I play clan matches with a bad KDR on purpose. I play Skulk (and preferably stay skulk) aggressively and often carry a large amount of deaths. I do this because I understand that <b>other things</b> are more important. I focus on key enemy players, stopping momentum, forcing defensive play, killing important structures, etc.
However, I still realise that KDR can give you information about how the game is going. NS2 is based around map control, momentum and raiding. You need (in all but the odd strategy, all lerks, etc) to kill, or not be killed by, the other team to do these things. Kills and deaths will always be an important aspect in NS2, as will other things brought into NS2 from the FPS and RTS elements.
<i>edit - It is not a question about whether you are worried about going into a room as you might die, its about the fact that when you do go into the room it is important that you win the encounter (whether or not you kill the opposition or die is important).</i>
This can be number of things, such as:
-Capping rts/biting down rts
-getting a forward base
-scouting while picking out solo players
-defending a key potition
-putting pressure
However you can get good k/d by doing worthless things like staying in base all game do get 3-4 kills without dieing. But even so he is doing it very efficently since he can defend the base without allowing the skulk to take anything down.
You should try to suicide as rarly as possible, there are timings in the game where it can be effective but most of the time you should focus on getting your task done without dieing.
---
The strategy/tactic used is important in game but you need players capeable of doing they part for it to be effective. If you want a skulk to keep pressuring rts its important he is able to stay alive, if he can pick out the marines who try the save the rt he is already being much more efficent than just suiciting 1 rt down. This is the same with marines, if you want a marine to cap the other side of the map he must be capeable of defending himself if a skulk arrives. Even though he might only get 0-2 kills (since he is not in the main fight) he is doing something very efficent while he stays alive.
Its bit similar to starcraft, it doesnt matter what strategy/tactic you use, if you are not good enough to do it efficiently you will lose vs a better player. It does not mean that tactics are useless.
--
You can get these statistics from K/D, If players are doing whats important it can tell you how well they are peforming that task. But other factors are also important.
I've seen games won by the team with lower KD overall (by quite a bit)
The fact that PRES is not given per kill anymore also reduces the effectiveness of getting kills.
OP has addressed this slightly from this post but if all you do is camp for kills while leaving the other teams res nodes alive then they can still easily come back... etc, etc
in the end its still all about the res nodes (and hive 2 :p)
i'm kind of with elodea on this one. it is a useful stat for your front line guys - those wins in the corridor match-ups are huge.
but there are less aggressive (& yes less talented) players that we need to encourage to continue to play, so the scoring system should give them a sense of accomplishment as well. we don't want anyone looking at their KDR & say 'I s*ck' (or worse be told 'you s*ck') & quit playing the game.
i like the assist idea - maybe there can be some scoring enhancements around groups
I have been in at least 4 different games where our team has come back and won from circumstances that we should not have (such as being a marine against 3 hives, or being an alien against marines who control the majority of the map.)
One of the main reasons we won this is because everyone on the team was willing to give up their life for the cause, they could go -30 and it would not matter as long as we won which we did. When you look at players who thrive on their KDR they become cowards afraid to die ingame, which leads to them doing nothing to turtling and base which leads to the opposing team having free unopposed reign on the map.
tl;dr Version
If i see some one on the team who has a +30 KDR but isnt helping the team to do ###### other than kill and i see some one with a -30 KDR who is doing everything in their power to help the team win guess who is the noob? That's right, the self important KDR loving ######.
Indeed but i would actually go further than that and say KDR does even matter in any game where overall score or kills alone places a player at the top winning position, for example in MW3 not only does the biggest score get top position but the more points you get per round the quicker you rank up, my KDR can be pretty bad by some peoples standards yet more often then not ill be the player with the most points, to me deaths are meaningless when you can respawn a few seconds later, there's little downtime and zero cost so over the course of a game i can be out there running around scoring more kills and points while someone playing more carefully doesn't.