Off Topic: The Flood
This topic has moved here: Subject: Let's have a talk about K/D
  • Subject: Let's have a talk about K/D
  • Pages:
  • 1
  • 2
  • of 2
Subject: Let's have a talk about K/D

Expressing my strong liberal views without shame. Favorite quotes below:

"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe."
"One starts to live when he can live outside himself."

- Albert Einstein

Let me preface my post by saying that my K/D in Halo 4 is slightly less than 1 as of this post's original submission. I am not here to justify my skill--or lack thereof--in this discussion. The conclusions I reach and the arguments I present are all based on my observations of Halo multiplayer for the past five years. I am also not here to arbitrarily assign all ranges of kill to death ratios to certain skill levels.

So, now that that's out of the way, let's get to it.

When measuring skill in competitive arenas in video games--and usually shooters--it has become custom to use one datum: the kill to death ratio. And for good reason. In these games, the primary objective is generally to obtain more kills than the opposing team(s), so it would be logical to presume that a player who killed more than was killed positively contributed to the team.

What disturbing "consensus" I have seen, however, is the placement of the average K/D ratio. Please note that I am not talking about players' K/Ds who are average; I am literally talking about the statistical average when analyzing data. While there is a correlation between these two sets, they are not identical.

Many users of the B.net forums--and of the Waypoint forums as those have emerged as the Halo HUB--seem to think that the average K/D ratio is above 1.0, and players who have a ratio of 1.0 are below average.

This is a conclusion I completely disagree with, both in the statistical analysis and skill-gap aspects. If the statistical arithmetic average K/D for the entire game's competitive population is significantly less than or greater than 1, then there is something wrong with the game's balancing or matchmaking. Now, let's move on to the more controversial aspect: skill.

I believe Recon (Number 54) said some time recently that a 1:1 K/D is an indicator that the matchmaking system and the player are working together properly (apologies if this is out of context or incorrect). I completely agree. In fact, I would argue that any significant (say, 0.5) deviation from this is a result of poor matchmaking results or, less likely, balancing. This makes sense: clearly a player with 0.5 ratio who tried is not playing against the right people and the same goes for a player who has far surpassed 1.5. Now, this needs to be analyzed in trends, not single matches, as outliers do occur.

My ultimate point is this. While K/D is a measure of player proficiency to some degree, we as a community need to reach the consensus that an average K/D is in fact around 1. I feel if this is agreed upon, a lot of the squabbling and biting seen in some of the more heated debates about matchmaking systems will settle down in the future.

  • 12.01.2012 9:09 PM PDT

Studies show that men think about sex every 7 seconds. I do my best to eat hotdogs in under 6, just so things don't get weird.

Please allow me to introduce Myself
I'm a man of wealth and taste
I've been around for a long, long year
Stole many a man's soul and faith

I still don't know how to link my PSN stats on here to show you my mad skillz at Halo 4.

  • 12.01.2012 9:10 PM PDT

"Halo! Its divine wind will rush through the stars, propelling all who are worthy along the path to salvation."

So what you are saying is that if you have a K/D above the 1.1 mark then you're okay?

  • 12.01.2012 9:10 PM PDT
  •  | 
  • Legendary Member


Posted by: MyNameIsCharlie
I still don't know how to link my PSN stats on here to show you my mad skillz at Halo 4.


dud

letz trade friend codes on the wii so i can show u my mad skills at halo 4

OT: K/D doesn't really matter.

[Edited on 12.01.2012 9:11 PM PST]

  • 12.01.2012 9:10 PM PDT
  • gamertag: [none]
  • user homepage:

Kites and Dinosaurs? Sure!

I like 'em both green.

  • 12.01.2012 9:11 PM PDT

Gather around the campfire, once it goes out, it's out for good.

tl;dr

k/d communism

  • 12.01.2012 9:12 PM PDT
  • gamertag: [none]
  • user homepage:

And I don't give a damn if you don't like me Cause I don't like you Cause you're not like me And I don't give a damn if you don't like me Cause I don't like you Cause you're not like me And I don't give a damn if you don't like me Cause I don't like you Cause you're not like me And I don't give a damn if you don't like me Cause I don't like you Cause you're not like me


Can not be touched.

Didn't even read half of that.

I couldn't care less about K/D. I'm usually drunk when I play, anyway.

  • 12.01.2012 9:12 PM PDT

Ignore my gamertag. It's actually Dragonzzilla.

Men should define their K/D, not the other way around.

  • 12.01.2012 9:14 PM PDT

Expressing my strong liberal views without shame. Favorite quotes below:

"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe."
"One starts to live when he can live outside himself."

- Albert Einstein


Posted by: Eat More Babies
Didn't even read half of that.

I couldn't care less about K/D. I'm usually drunk when I play, anyway.

If you didn't read half of it, why'd you feel compelled to post?

I don't care that much about my K/D's value either; I do care, however, about how K/D is viewed in general.

  • 12.01.2012 9:14 PM PDT

Unreal, isn't it?

I didn't read the entire OP, but yeah, ideally, over an infinite period of time, everybody's k/d would even out as everyone moves to their correct place in the MM system. However, that isn't the reality. Good players beat the worse players and move upwards, and have a k/d greater than 1, implying that they are better than the majority of people they've come up against.

k/d is a good indicator of relative skill until you reach your correct MM rank. At that point your rank is the indicator of relative skill.

Edit
Ah, I see. You're saying that people view the average k/d as greater than 1 and look down on people who have a k/d of 1. Well, to that I'd say that in general our (American at least) society generally looks down on people who do average (i.e. a C in a class is usually considered bad), if you aren't above average you're stupid. Not saying it's right, but yeah you're right about the general perception.

[Edited on 12.01.2012 9:23 PM PST]

  • 12.01.2012 9:16 PM PDT
  • gamertag: [none]
  • user homepage:

We all wear masks and the time comes when we cannot remove them without removing our own skin.

Yup, every single K/D thread is made by one of these two people.

*Sucks at the game and is annoyed that people have actual proof they suck / tries to give an excuse on why a low K/D is still good.
*Someone who is really good at the game and likes to brag about it since they have actual proof of it / calling anyone with a lower K/D bad or a scrub.

[Edited on 12.01.2012 9:17 PM PST]

  • 12.01.2012 9:16 PM PDT

Sometimes I am tempted to quit gaming. I can no longer play without seeing ponyphags, horrendously bad teammates and mexicans.

Gaming community, I want you to know that you suck.

I love KD. Its so cheesy, and it ain't easy bein cheesy.

  • 12.01.2012 9:17 PM PDT


Posted by: MyNameIsCharlie
I still don't know how to link my PSN stats on here to show you my mad skillz at Halo 4.

You're so hip with us kids Charlie!

  • 12.01.2012 9:17 PM PDT
  • gamertag: [none]
  • user homepage:

Contrary to popular belief, friendship is not magic. Unless sex is involved.

tl;dr

it's a video game, no one cares.

  • 12.01.2012 9:18 PM PDT

Expressing my strong liberal views without shame. Favorite quotes below:

"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe."
"One starts to live when he can live outside himself."

- Albert Einstein


Posted by: RealUnrealRob
I didn't read the entire OP, but yeah, ideally, over an infinite period of time, everybody's k/d would even out as everyone moves to their correct place in the MM system. However, that isn't the reality. Good players beat the worse players and move upwards, and have a k/d greater than 1, implying that they are better than the majority of people they've come up against.

k/d is a good indicator of relative skill until you reach your correct MM rank. At that point your rank is the indicator of relative skill.

A K/D higher than 1 would imply that you are better than a majority if means 50%, but that framing is misleading when you consider a good 45-49% could potentially be better. I agree that K/D is a good indicator of skill until you reach your equilibrium rank, however I disagree with where that average has been placed.

  • 12.01.2012 9:18 PM PDT

Join Halo Haven for all things related to Halo 4


Tell her that If you jingle my bells, Ill promise you a white Christmas - Call Me Venom
The world can't end next month. My yogurt expires in 2013 - Princess Cadence
If Apple invented a car, would it have windows? - Xxembers

If the majority of the population's K/D is below 1.0 (<.99) and the remaining percentage is above 1.0 (>1.01) then I think I'd agree with this 1:1 theory.

However, since we really have no practical way to quantify such data, I think it would be okay to assume that the average is above given the fact that a good portion of players have a k/d at least above 1.0 and that the extreme outliers in the 2.0+ range could be enough to jump the average passed that.

  • 12.01.2012 9:19 PM PDT


Posted by: Buddha Milk
Yup, every single K/D thread is made by one of these two people.

*Sucks at the game and is annoyed that people have actual proof they suck / tries to give an excuse on why a low K/D is still good.
*Someone who is really good at the game and likes to brag about it since they have actual proof of it / calling anyone with a lower K/D bad or a scrub.

I hate the way K/D is viewed and I have what I consider an ok K/D in Halo 4. (1.83 or something last time I checked).

  • 12.01.2012 9:19 PM PDT

Expressing my strong liberal views without shame. Favorite quotes below:

"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe."
"One starts to live when he can live outside himself."

- Albert Einstein


Posted by: Buddha Milk
Yup, every single K/D thread is made by one of these two people.

*Sucks at the game and is annoyed that people have actual proof they suck / tries to give an excuse on why a low K/D is still good.
*Someone who is really good at the game and likes to brag about it since they have actual proof of it / calling anyone with a lower K/D bad or a scrub.

I don't see where I say a K/D below 1 or even at 1 is good.

  • 12.01.2012 9:19 PM PDT

I find win to loss ratio to be a better indicator of skill. Just saying.

[Edited on 12.01.2012 9:22 PM PST]

  • 12.01.2012 9:22 PM PDT

Unreal, isn't it?

See my edit. I agree that (in general, not just in this game) we've put the average or expected value at not average.

  • 12.01.2012 9:22 PM PDT
  • gamertag: [none]
  • user homepage:


Posted by: Plasma Prestige
Let me preface my post by saying that my K/D in Halo 4 is slightly less than 1 as of this post's original submission. I am not here to justify my skill--or lack thereof--in this discussion. The conclusions I reach and the arguments I present are all based on my observations of Halo multiplayer for the past five years. I am also not here to arbitrarily assign all ranges of kill to death ratios to certain skill levels.

So, now that that's out of the way, let's get to it.

When measuring skill in competitive arenas in video games--and usually shooters--it has become custom to use one datum: the kill to death ratio. And for good reason. In these games, the primary objective is generally to obtain more kills than the opposing team(s), so it would be logical to presume that a player who killed more than was killed positively contributed to the team.

What disturbing "consensus" I have seen, however, is the placement of the average K/D ratio. Please note that I am not talking about players' K/Ds who are average; I am literally talking about the statistical average when analyzing data. While there is a correlation between these two sets, they are not identical.

Many users of the B.net forums--and of the Waypoint forums as those have emerged as the Halo HUB--seem to think that the average K/D ratio is above 1.0, and players who have a ratio of 1.0 are below average.

This is a conclusion I completely disagree with, both in the statistical analysis and skill-gap aspects. If the statistical arithmetic average K/D for the entire game's competitive population is significantly less than or greater than 1, then there is something wrong with the game's balancing or matchmaking. Now, let's move on to the more controversial aspect: skill.

I believe Recon (Number 54) said some time recently that a 1:1 K/D is an indicator that the matchmaking system and the player are working together properly (apologies if this is out of context or incorrect). I completely agree. In fact, I would argue that any significant (say, 0.5) deviation from this is a result of poor matchmaking results or, less likely, balancing. This makes sense: clearly a player with 0.5 ratio who tried is not playing against the right people and the same goes for a player who has far surpassed 1.5. Now, this needs to be analyzed in trends, not single matches, as outliers do occur.

My ultimate point is this. While K/D is a measure of player proficiency to some degree, we as a community need to reach the consensus that an average K/D is in fact around 1. I feel if this is agreed upon, a lot of the squabbling and biting seen in some of the more heated debates about matchmaking systems will settle down in the future.


'Nuff said.

  • 12.01.2012 9:22 PM PDT

"There's a very fine line between not listening, and not caring. I'd like to think that I walk that line every day."

I suppose that makes sense in a way. Does this imply that nearly every multi-player-game's matchmaking system is broken?

I think it's important to note that some people, like myself, will leave during games where they're not doing well. Stuff like that affects K/D.

But overall it's really not significant whatsoever. So many variables go in to it that it can't express.

  • 12.01.2012 9:23 PM PDT
  • gamertag: [none]
  • user homepage:

We all wear masks and the time comes when we cannot remove them without removing our own skin.


Posted by: Crazy LlamaX

Posted by: Buddha Milk
Yup, every single K/D thread is made by one of these two people.

*Sucks at the game and is annoyed that people have actual proof they suck / tries to give an excuse on why a low K/D is still good.
*Someone who is really good at the game and likes to brag about it since they have actual proof of it / calling anyone with a lower K/D bad or a scrub.

I hate the way K/D is viewed and I have what I consider an ok K/D in Halo 4. (1.83 or something last time I checked).


A K/D of 1.5 to 1.9 is good.
The people with really high K/D's are the ones who only go for K/D. As long as you PTFO and don't go horribly negative, you help the team more than a stat padder.

  • 12.01.2012 9:23 PM PDT

Expressing my strong liberal views without shame. Favorite quotes below:

"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe."
"One starts to live when he can live outside himself."

- Albert Einstein


Posted by: A 3 Legged Goat
If the majority of the population's K/D is below 1.0 (<.99) and the remaining percentage is above 1.0 (>1.01) then I think I'd agree with this 1:1 theory.

However, since we really have no practical way to quantify such data, I think it would be okay to assume that the average is above given the fact that a good portion of players have a k/d at least above 1.0 and that the extreme outliers in the 2.0+ range could be enough to jump the average passed that.


Well, we as players don't have any way to calculate this practically. However, calculating an arithmetic mean for a developer who knows how to structure a matchmaking system properly should be cake. If the average K/D of the Halo 4 population is significantly deviant from 1, I would argue that there is something with the game's matchmaking or balancing. My theory would rely on the fact that extrema are accounted for on both ends (as that is what means do).

  • 12.01.2012 9:23 PM PDT

I am Brosidon, King of the Brocean

This is all that's coming out of your mouth, OP

(Press dialectize)

[Edited on 12.01.2012 9:24 PM PST]

  • 12.01.2012 9:24 PM PDT

  • Pages:
  • 1
  • 2
  • of 2