Jump to content

K/d Ratio Not A Good Stat?


73 replies to this topic

#61 Zervziel

    Member

  • PipPipPipPipPipPipPip
  • The Marauder
  • The Marauder
  • 907 posts
  • LocationTharkad

Posted 19 March 2013 - 07:12 AM

My KDR is kinda skewed because for several drops in a row, anytime I got anywhere near dropping an enemy mech, some random ******* would pop up and get a laser blast while my autocannon was still in route. this went on for quite a few games until I got tired of it and tore an arm off of a killstealing puggie who'd pulled that **** about three times in the same match. Naturally he was at the top of the ratings while I was middle. Quite frankly I wouldn't miss stats if they were to disappear as it's basically sheer e-peen.

Ideally it shouldn't matter who drops who, but when you get into a brawling duel with another atlas, both of you having wounded each other so that one more salvo will decide the winner of a long grueling duel only to have a teammate run up and blast him, it kills the fun of that battle drastically. It's not skill that wins, it's the cheap *** splatcat that just trundled around a corner or a light that just blitzed in long enough to steal a kill before bolting.

Edited by Zervziel, 19 March 2013 - 07:17 AM.


#62 Stoicblitzer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,931 posts
  • LocationUSA

Posted 19 March 2013 - 07:15 AM

kdr is also affected by disconnecting now. i'm not against it but my game has been freezing a lot lately. Those are deaths that I'm not responsible for. A higher KDR likely means that you are a solid player but not always especially for people who had KDRs before phase 3 and only did 4 mans. Those used to be roflstomps. Now there is very little class matching in pugs and the games are typically harder, at least in my experience. My KDR is 5.05.

#63 katshing

    Rookie

  • 4 posts

Posted 19 March 2013 - 07:23 AM

View Postkatshing, on 19 March 2013 - 07:12 AM, said:

No, with the current rules of this game what matters in the end is if you win or not nothing else.
thats why you get more c-bills for an assist then a kill (i think)
If i want to farm c-bills i will get a friend and make sure i do all the damage needed for the kill -1 and have my friend take the kill

So basicly the more C-bills/match the more skill you have

#64 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 19 March 2013 - 07:25 AM

View Postkatshing, on 19 March 2013 - 07:12 AM, said:

No, with the current rules of this game what matters in the end is if you win or not nothing else.

You win by killing mechs.
Players with high K/D ratios will generally have higher W/L ratios than folks with negative K/D ratios. Because they contribute more to winning.

View Postkatshing, on 19 March 2013 - 07:12 AM, said:

thats why you get more c-bills for an assist then a kill (i think)

No, you don't get more. All else being equal, they are worth the same, as the kill+the last component destruction which leads to the mech's death will add up to the same as the assist.

#65 Red squirrel

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,626 posts

Posted 19 March 2013 - 07:26 AM

View PostKhobai, on 18 March 2013 - 10:59 PM, said:

KDR is a bad stat because kills mean absolutely nothing. It just means youre good at getting the last shot in and kill stealing.

A better statistic would be damage done divided by damage received


In the end there is no calculatable stat that can tell you how good you are.

Want a good K/D ratio: Take a heavy alpha build and time your shots
Want lots of damage dealt: Use crazy LRM boats
Want good W/L ratio: PUG stomp with your Raven 3L or Atlas D-DC ECM 4 man team

You know that you are a good pilot when:
1) Your team mate salutes you for saving his ***.
2) You get ripped in pieces delaying the enemy cap and secure the victory
3) You shoot the arm off that YLW, the RT out of that Hunchback and then go for the next priority target instead of collecting kills.
4) You announce targets, help your LRM boats to get a lock, come up with a good tactic or support a good plan someone else has.
5) You know that team work and positioning are the most important thing in this game
...etc...

#66 Stoicblitzer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,931 posts
  • LocationUSA

Posted 19 March 2013 - 07:27 AM

View PostRed squirrel, on 19 March 2013 - 07:26 AM, said:


In the end there is no calculatable stat that can tell you how good you are.

Want a good K/D ratio: Take a heavy alpha build and time your shots
Want lots of damage dealt: Use crazy LRM boats
Want good W/L ratio: PUG stomp with your Raven 3L or Atlas D-DC ECM 4 man team

You know that you are a good pilot when:
1) Your team mate salutes you for saving his ***.
2) You get ripped in pieces delaying the enemy cap and secure the victory
3) You shoot the arm off that YLW, the RT out of that Hunchback and then go for the next priority target instead of collecting kills.
4) You announce targets, help your LRM boats to get a lock, come up with a good tactic or support a good plan someone else has.
5) You know that team work and positioning are the most important thing in this game
...etc...

you, i like you.

#67 Teralitha

    Member

  • PipPipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 3,188 posts

Posted 19 March 2013 - 07:39 AM

View PostZervziel, on 19 March 2013 - 07:12 AM, said:

My KDR is kinda skewed because for several drops in a row, anytime I got anywhere near dropping an enemy mech, some random ******* would pop up and get a laser blast while my autocannon was still in route. this went on for quite a few games until I got tired of it and tore an arm off of a killstealing puggie who'd pulled that **** about three times in the same match. Naturally he was at the top of the ratings while I was middle. Quite frankly I wouldn't miss stats if they were to disappear as it's basically sheer e-peen.

Ideally it shouldn't matter who drops who, but when you get into a brawling duel with another atlas, both of you having wounded each other so that one more salvo will decide the winner of a long grueling duel only to have a teammate run up and blast him, it kills the fun of that battle drastically. It's not skill that wins, it's the cheap *** splatcat that just trundled around a corner or a light that just blitzed in long enough to steal a kill before bolting.


Isnt winning all that matters? I play with a coordinated team, and you know what? There is no griping on who gets the kills so long as we win. It soulds to me like your more worried about KDR than the ones who 'steal' your kills. Its a team game.

Edited by Teralitha, 19 March 2013 - 07:41 AM.


#68 xhrit

    Member

  • PipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 976 posts
  • LocationClan Occupation Zone

Posted 19 March 2013 - 07:55 AM

View PostHungus, on 19 March 2013 - 04:30 AM, said:

EDIT: I just watched that video and found it rather humorous after reading your post where you say "First improve your aim. Where you land your shots is more important then actually landing shots." Yet in that video your accuracy is terrible, infact you seem to be just mashing/holding the alpha key quite often and just spamming machine guns (lol) everywhere.


Spams machineguns, outdamages entire 4 man http://house-marik.net/ premade.

Haters gunna hate.

lol?

But yeah, you are right. My accuracy was pretty bad in that video. I was playing on my girlfriends machine, which is quite a bit more powerful then my own computer. My PC is incapable of running MWO on ultrahigh settings, let alone running it on ultrahigh while recording it. In any case, she has a 200$ microsoft natural wireless mouse 6000 and a 150$ XArmor U9 Cherry MX brown wireless mechanical model M clone keyboard, being the haxxor that she is.

It is a really nice keyboard, but has a hard limit of 5 keypresses at once, otherwise there are conflicts. Also, the keys stop repeating if you hold them down for more then 30 seconds. You can see me getting hung up on the controls several times in that video, like when I fall off the bridge at the very start of the match. I hate her mouse too, not because it is a bad mouse, but I just hate mice in general.

Next time I user her machine to record mwo I will bring my 10$ wired Logitech K120 keyboard, and my 20$ wired Kensington Orbit Trackball...

...then maybe I won't get so frustrated trying to snipe that I decide to rush a splat cat with my machineguns, like in this video.


Edited by xhrit, 19 March 2013 - 08:08 AM.


#69 Gaan Cathal

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 2,108 posts

Posted 19 March 2013 - 08:06 AM

View PostZervziel, on 19 March 2013 - 07:12 AM, said:

My KDR is kinda skewed because for several drops in a row, anytime I got anywhere near dropping an enemy mech, some random ******* would pop up and get a laser blast while my autocannon was still in route. this went on for quite a few games until I got tired of it and tore an arm off of a killstealing puggie who'd pulled that **** about three times in the same match. Naturally he was at the top of the ratings while I was middle. Quite frankly I wouldn't miss stats if they were to disappear as it's basically sheer e-peen.


You are aware that this makes you the ****? Height on the leaderboard is almost entirely due to damage dealt, meaning that if he was at the top of the ratings, you got petty and intentionally attack a teammate who was doing a lot of work for your team, because you happened to be shooting one of his targets before he did?

#70 xhrit

    Member

  • PipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 976 posts
  • LocationClan Occupation Zone

Posted 19 March 2013 - 08:17 AM

View PostRed squirrel, on 19 March 2013 - 07:26 AM, said:

Want a good K/D ratio: Take a heavy alpha build and time your shots
Want lots of damage dealt: Use crazy LRM boats
Want good W/L ratio: PUG stomp with your Raven 3L or Atlas D-DC ECM 4 man team

You know that you are a good pilot when:
1) Your team mate salutes you for saving his ***.
2) You get ripped in pieces delaying the enemy cap and secure the victory
3) You shoot the arm off that YLW, the RT out of that Hunchback and then go for the next priority target instead of collecting kills.
4) You announce targets, help your LRM boats to get a lock, come up with a good tactic or support a good plan someone else has.
5) You know that team work and positioning are the most important thing in this game
...etc...


Red squirrel's good tips for good pilots :

1) Always give your dudebros mad props. /brofist

2) Run off by yourself and die, it totally helps win games.

3) Shoot their arms and legs off first, headshots are overrated.

4) Make sure your raven 3L is equipped with TAG, NARC and twin flamers. That way you help your DDC lrm boats more.

5) And remember, a good team can carry a bad player. When in doubt position yourself behind the rest of your team.

Edited by xhrit, 19 March 2013 - 08:19 AM.


#71 Tikkamasala

    Member

  • PipPipPipPipPipPip
  • 210 posts

Posted 19 March 2013 - 08:29 AM

View PostKhobai, on 18 March 2013 - 10:59 PM, said:

A better statistic would be damage done divided by damage received


Why do you think dying after taking an alpha to the head is preferable to making the enemy work for the kill and stripping off every bit of armor before you go down? In the latter case you prevent the enemy from shooting at other perhaps more valuable targets.
Similarly, spreading damage all over the enemy team won't help your team as much as focusing on the right target/part of a target with the intent to kill or disable rather than accumulating a large damage done value.

#72 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 19 March 2013 - 08:57 AM

Ya, ideally you would want to have the highest (damage received)/(deaths) ratio possible.

#73 Vrekgar

    Member

  • PipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 366 posts

Posted 19 March 2013 - 09:05 AM

There is no good metric currently for determineing if someone is a "good player".

K/D Ratio: ELO will trend this to 50/50. Can be heavily skewed by group play, because teamwork and communication is Powerful.

Destroyed/Loss Ratio: Too high of a chance of someone just running and powering down if he cant get kills. Also too many times people will focus a target down only to have the kill taken by another team mate who only got the killshot and never helped.

Match Score: Nobody quite knows exactly what its rating you and how it stacks up.

Damage: High alpha builds along with LRM builds dominate this, despite the questionable usefullness in the grand scheme of a match.

Assists: Only measures that you helped by at some point you shoot the target. The most useful out of the bunch because it means you at least shot X number of targets and the targets were killed.

#74 Ph30nix

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,444 posts

Posted 19 March 2013 - 09:25 AM

personaly i think K/D ratio tracking should be removed, so should seeing # of kills per match,, it just encourages people to try and rambo instead of working as a team.

its not uncommon for me to have zero kills but 5+ assists and have my damage be 400+ so its not like im just tapping every mech and moving on.

im in middle of fighting a mech that i just crippled and i have a teamate still on it and i see a new enemy coming? i switch to the new guy and work on killing or crippling him and let my teammate finish the already crippled enemy (by crippled im talking majority of their weapons are gone and have an almost zero chance of killing anyone before they are finished)





8 user(s) are reading this topic

0 members, 8 guests, 0 anonymous users