135 replies to this topic

[mailin]

Don't worry about KDR, or even win/loss. Look at XP per match as previously mentioned. More XP per match on average means more damage done, more kills, more assists, etc. In other words, a greater likelihood that you're helping your team.

As pointed out previously as well, KDR is too mech dependent. My favorite mech is a Spider 5D with 3 mplas. He's a leg collector. I leg enemies and then usually move on. Sometimes I go for the kill, so my KDR in that mech is only 1.5. But my XP per match in that mech is crazy high because it's a spotter and assist machine. Same with my Raven 3L. He's another legger, but because I have it loaded with streaks and a tag I do get kills in him, so the KDR is 1.87.

If I just went for the kills in those mechs it would probably be higher, but there are two things holding me back on that. 1. While I like taking out the enemies, I really enjoy making them mad by taking a leg and letting the brawlers finish them while I scamper off to find another target. 2. I figure I already got an assist and a component destruction, so I let my friendlies get the final kill XP and C-bills. I'm very cooperative that way.

[Vanguard319]

My overall kdr is .95 atm, though some of my kdrs with individual mechs are higher. (my best is my Flame at 3.00)

[Gremlich Johns]

KDR = e-peen

What's your percentages with a specific weapons? That should mean more to you than the KDR.

Edited by Gremlich Johns, 31 May 2014 - 07:44 PM.

[Rogue Jedi]

anything better than 0 is a good KDR

your KDR depends on all sorts of factors, if you usually pilot a lightly armed scout then you will rarely get a kill but get killed most games, in that situation you could be a huge benefit to the team bit rarely do much damage or get a kill, if you are a sniper using ACs Gauss and PPCs you may rarely be seen and shot at so you could get a high KDR of 5+ but possibly be of less help to the team than the spotter with a 0.1KDR

if you pilot a brawler assault who takes damage for the team to advance then deals massive damage when getting close you would probably have about a 1 KDR

I usually pilot lights in a scout harasser role, my KDR is 0.53 however some of my Mechs have a much higher KDR and 1 mech has a 3 Win loss ratio

[Bartholomew bartholomew]

A positive KDR and W/L feels real nice to look at. But That feeling you get when you KNOW you helped you team win in a close match. That is priceless.

Edited by Bartholomew bartholomew, 01 June 2014 - 05:58 PM.

[RLBell]

The goal is to have fun.

I am just annoyed that you cannot earn the Seriously! achievement and the Philanthropist achievement more than once, because I did both in a single match. I unloaded ten tons of LRM's for 1200 points of damage, twelve assists, and zero kills.

If I had scored a kill in that match, I would have gotten neither.

[Ursh]

In my brand new ember today, I managed to get 3 heavies and one medium to come chase me into the buildings on River City skirmish just as my team was really beginning to push. They chased me for about 90 seconds to two minutes before they finally killed me. I did 50 damage, with five assists.

Many people would tell you that's terrible and I should be ashamed. My team cleaned up most of their heavy strikers during my little dancing interlude, and we won the match with five mechs alive to zero. If those heavies had been focusing on our slower mechs, they might have actually done timely damage.

It's the reason i find it funny when the occasional thread pops up where someone complains about a high K/DR but low match w/l record. I do things that i think will help the team.

[Roughneck45]

As you can see, KDR is a touchy subject on these forums lol.

IMO as a player, you are holding your own at 1.5, an asset to the team at 2.0, and a hard carry at 3+

[Torgun]

Considering the sorry state of the matchmaker, KDR is pretty much meaningless as you can end up in a horrible team and get steamrolled before having time to do anything significant. Or if you're lucky often end up in the teams doing the rolling thus getting better stats than you deserve. It's pretty much a roll of the dice due to the matchmaker. I'm still convinced if we didn't have stats for deaths and kills, there would hardly be anyone running and hiding in a weaponless mech to preserve their precious stats, forcing everyone in that match to have their mechs locked as a result.

Edited by Torgun, 02 June 2014 - 03:03 PM.

[n r g]

This one

[Dunning Kruger Effect]

Quote

Lets try to look at the average K/D ratio of the community:

• Every kill made by a player causes another player to have a death. At this point, the ratio is 1:1.
  
• Can a player kill another player without causing a death? No, so the average ratio across the community cannot be higher than 1:1.
  
• Can a player die without the death being caused by another player, thus causing a kill? Yes, either by running out of bounds, destroyed legs by dropping, overheat, or turrets. Therefore, as long as players have died by alternative methods, the average K/D ratio of the community must be less than 1:1.

So, yeah. Above 1:1 is pretty ok. Heck, even 1:1 is above average. Not that it matters that much, because many of the roles in this game are based around doing something other than getting that final hit.

My response EDITED due to my own lack of reading comprehension. I'll just go with "QFT" here instead.

Random Thought of the Day: Do many people with a high (say 2+) KDRs claim it is meaningless? Or do only people with low-average KDRs claim they are unimportant?

Edited by Dunning Kruger Effect, 03 June 2014 - 03:55 AM.

[Dago Red]

Dunning Kruger Effect, on 03 June 2014 - 03:52 AM, said:

My response EDITED due to my own lack of reading comprehension. I'll just go with "QFT" here instead.

Random Thought of the Day: Do many people with a high (say 2+) KDRs claim it is meaningless? Or do only people with low-average KDRs claim they are unimportant?

I'd say that overall people with higher KD ratio's probably do weight it higher. But then the ones who don't are also less likely to engage in behavior that specifically improves that ratio at the expense of other things.

So there's probably an argument to be made for which is the cause and which is the effect.

Mines not great but then I'm digging myself out of a death hole caused by taking 6 months off and then coming back with a lot of rust and all previous stats being wiped. Is it important? Not really but it's a useful measuring stick for self progress.

[jper4]

i like the point about xp being potentially the most important stat since it includes things like spotting, component destruction (which will lead to kills for someone more often than not), savior/defensive kills which all tie into helping the team/keeping fellow teammates alive and being a good teammate vs i killed X amount more/less than i've died.

as an example, depending on the mech my kdr ranges from 0.0 (only played a match or two in them and didn't kill anything) to 5.0 currently. going 10+ matches per mech (not about to figure out everything including archived stats) then it's 0.36 (curse you locust 1V- actually was 0.64 before the reset though) to 3.0 (wow didn;t notice my battlemaster 1S actually passed my streaktaro- that just seems wrong) so therefore i suck and am good at the same time according to my kdr

[Mott]

Dago Red, on 03 June 2014 - 04:01 AM, said:

I'd say that overall people with higher KD ratio's probably do weight it higher. But then the ones who don't are also less likely to engage in behavior that specifically improves that ratio at the expense of other things.

This is me.

Tanar, on 03 June 2014 - 04:10 AM, said:

so therefore i suck and am good at the same time according to my kdr

This is me also. I have mechs (i only pilot mediums) that possess KDRs of 1.73-2.83, but they don't get me much in the way of XP or CBill bonuses because all they do is capitalize on the weakness of a couple mechs per match.

My mechs with KDRs of <1.5 actually do WAY more damage, get me WAY more XP & CBills per match and end up with better W/L because i'm tearing armament-limbs off several enemy mechs (saving my team from untold amounts of damage), getting 6-8 substantial assists before i even begin working for a kill to pad my stats.

By which time my team is already up by those lopsided 6-3, 8-2, 9-1 scores and it's a race to try to get one of the last killshots - which i only score once in a blue moon.

[Archie4Strings]

I think what really matters is just the Win/Loss-stats... you can play the match of your life, just being smart, good aiming, good maneuvering, make 1000+ damage, your team wins the match without you getting any kills but probably 12 assists and lots of destroyed components!

Getting the kill is often just shooting at the right time on the right target and being the lucky one who gets the final shot!
Sure it also can reflect your skill, for example by focussing on the damaged side-torsos to kill a mech much faster. But in my mind, there are so many situations in which its just luck to get the kill!

It also depends on your weapon set-up! When playing with my dragon (usually sniper) with just pin-point-damage (AC, Gauss, PPC...) i have a K/D-Ratio of 1,94. Far more than with all my other mechs, which i often equip with a few more lasers too (its usually around 1,00 - 1,4)
Uh damn no!! Just noticed that i have 2,0 with my Battlemaster 3M, and its not yet skilled... maybe just cause i play the 3M as a sniper with high pinpoint-damage too (its so slow and such a low torso-radius, that i prefer to stay in a safe distance.. But i made just around 20 Matches with it, so i guess my KDR will go down by time with that Mech...

Anyway: What i want to say is, that the Win/Loss-Stats do count much more, cause its your team that matters! Having 6 (more or less) kills in 1 match is mostly just pure luck! You cannot get 6 Kills by killing 6 Enemies one by one all by your pure skill without the help of your team!! Thats just the way it is! Except you ar playing vs REALLY bad noobs (and maybe not even then!)

[Alaskan Nobody]

Dunning Kruger Effect, on 03 June 2014 - 03:52 AM, said:

Random Thought of the Day: Do many people with a high (say 2+) KDRs claim it is meaningless? Or do only people with low-average KDRs claim they are unimportant?

Both actually - first people I heard talk about how it isn't the best stat to measure success by....

Were people like Koniving and Eglar, who have the stats to brag about.
(or in Koniving's case, had)

[Lootee]

I had some funny KDRs from before the wipe.

These were the most extreme ones, with the junk in the middle cut out:


The ratio doesn't really matter in the end, because often you'll find yourself driving a crappy mech variant with zero upgrades just to try to grind through the trees to unlock Elite/Master. So you end up eating tons of deaths just wishing you could get out of that mech and back into the 'real' one. The stat is mostly meaningless with the matchmaker imbalances and the XP grinding that's required to unlock the higher pilot trees. And sometimes after you've painfully grinded death after death in 1 variant, you feel it'd be a waste to stop now so you go and die some more in 2 more variants of the same mech.

Those damn Spiders and Locusts...


I wouldn't put much stock in KDR numbers. The system is kind of set up to tank your ratio on purpose. Used to be 4.0+ until I started leveling Spiders, Locusts, Trebuchets and all the other less potent mechs.

Edited by PanchoTortilla, 03 June 2014 - 05:25 PM.

[Podex]

Like everyone else has said -- it means nothing, thank goodness. Especially when you run fast mediums and often need to take one for the team to break up a swarm.

Everyone likes to chase the squirrel. I am that squirrel.

[WVAnonymous]

Gigastrike, on 30 May 2014 - 01:27 PM, said:

Lets try to look at the average K/D ratio of the community:

• Every kill made by a player causes another player to have a death. At this point, the ratio is 1:1.
  
• Can a player kill another player without causing a death? No, so the average ratio across the community cannot be higher than 1:1.
  
• Can a player die without the death being caused by another player, thus causing a kill? Yes, either by running out of bounds, destroyed legs by dropping, overheat, or turrets. Therefore, as long as players have died by alternative methods, the average K/D ratio of the community must be less than 1:1.

So, yeah. Above 1:1 is pretty ok. Heck, even 1:1 is above average. Not that it matters that much, because many of the roles in this game are based around doing something other than getting that final hit.

Actually, the average among the "active" community can be much higher than one.

Players sign up, get near zero kills over 30 or 40 games and then quit.

The KDR among players who abandon the game is a stat that PGI should be more interested in than the KDR for the 1000+ drop dual gauss or AC5/PPC ninjas.

[plodder]

Whatever kdr are you are comfortable with is the goal.
Some fellas scream and yell in rage if it drops below 5. Others are mildly perturbed if it drops below .75
Mine has dropped quite a bit since I am dropping less, trying to builds and styles of play.
My opponents seem to be tougher too. Good work enemies o7!
I also don't care, except it used to be 2-3 times as high for over a year. But, if I cared more it would be higher.

When I do my job as an Atlas brawler I might not get a kill, but do over 500 damage, as I entertain the enemy, and my lance brings them down. I like a good scrap and a victory above all. Gl hf