KDR

What is a bad, average, good, and great KDR? Without gaming the system of course. Thanks, just wondering.

 

IMHO

1.0 KDR? Very very bad
1.0-2.0 KDR? Pretty good
2.0-3.0 KDR? You don't PUG drop often do you?
3.0+ KDR? You don't PUG drop hardly EVER!

 

Given that list, then I am downright shitty.

 

First I will concede that I have not been playing very long. Second I have a negative KDR (had to look it up after reading this post, did not even know they had those stats). So might be biased.

Going out on a limb here. IMHO KDR is an absolutely worthless meaningless statistic. To name just one example, when you are running a light mech, harassing the other team, providing laser or Narc targets, you rarely get a kill, I may get 6-10 assists but I rarely get a kill. So the KDR might be horrible but is it really reflective of you contribution? I submit the answer is no.

 

Off the cuff, I'd say:

Too much under 1 is bad. Assuming you're breaking even on wins and losses (which is the Elo based MM's goal), it'd make sense for you to get a kill on most wins, and die on most losses. Deaths on a win, and kills on a loss may well cancel each other out. Playing smarter than, and aiming better than the average bear will pick your KDR up.

Above 1.5 is bordering on average competency. You've been around the block, and know enough tricks to outsmart some opponents or cheat death slightly more often than not.

Above 2 is above average, by definition. You just might have figured out how to spot an open XL side torso, where and how to shoot the most important components, and strategic positioning for the different maps.

Double that and you're got a 4.0 KDR, which is pretty good if you're going solo.

Double that again, and 8 KDR is pretty great. Probably only attainable in a powerful, kill-oriented mech, dropping in a lance if your weight class elo is where it should be.

 

Any mechspeccers account is borked already, KDR is pointless given the amount of different builds we try. I've lost track of the number of builds I've tried on any given chassi except some few where I fell in love with a particular setup. On those I may concede measuring skill by KDR could indicate a skill level.

Compared to a pal of mine that nearly exclusively runs AC40 Jaegers, our scores are way different. Again, another pal of mine runs MWO rarely but is a big fan of the 1X Blackjack with 8 MLAS and he gets obscene amounts of kills and damage.

Stats only make sense with a large enough sample over a range.


That's just my humble opinion mind you.

 

KDR is not a sufficient Indicator for the player's skill. It is rather an Indicator how deadly your Mech(s) is(are). It doesn't say if you are doing your Role correct (most times) with your Mech(s).

As said already you can run Cheese.Builds and say you are skilled with a high KDR. But imo you still can have a low WLR. What about Kill-Assistances?

 

KDR is kind of important, as it shows how much you effect the game. Granted their are some other rules in a game that people can play that effect their KDR negatively, but I don't buy that most people do that every time.

For example I know several lights who can easily keep a KDR over 1.0. I'm not even a great light pilot and my Spider is over 1.5...

The one that will get you the most is line breaker, or damage sponge. You know that guy who leads the charge to break a formation. He's pretty much guaranteed dead, and will most likely not kill anyone, but if he does it right and the team supports the charge like a team its great for over running a position.

That all being said I'm not supportive of running to hide to protect your KDR. It's a silly stat. It means something, as long as you're not obsessed over it.

I do agree WL is very important, however it's easily mitigated by your team. I've pulled the WL up on some really bad players when they've dropped with me simply because I can carry their weight when the ELO machine drops me down into their class of player instead of the other way around.

Once you roll all of that together, every bit of it needs to be taken with a grain of salt because you never know how much team play is involved in generating the stats. And we all know Team Work is OP.

 

In my opinion, depending on the type of drops you do, KDR is tits on a bull.

Why?

If you are dropping 12 mans with guys like FM all the time and you NEVER PUG drop? Then your KDR matters. Those guys, and teams like them, all know what they are doing so you can do extremely well and your KDR is more representative of your skill.

If you are PUG dropping all the time? KDR doesn't mean shit. It just means "I've managed to get lucky and kill/kill-steal this many mechs before we won/lost the match". Also, in PUG drops you can be the best most awesome ace player on that team and have 11 other people fuck it up for you causing you a death AND a loss. PUG matches plunge your KDR by an order of magnitude.

My overall KDR is something like 2.4 right now but I've got mechs that I grinded through with KDRs in the .8 etc (I think one of my Spiders was like less than 1.0 as I was speed+PUG grinding it).

I also do a lot of PUG dropping when team mates aren't around or when I just want to quickly do what I want to do without worrying about team work and all that jazz. That affects my KDR as I've managed to go matches where I have killed 2-3 enemy mechs and still get my ass killed because the PUGs drop the ball at epic levels.

 

I think this is such an arbitrary stat. I remember when I first started playing MWO I would charge all lone wolf style into the other group and overheat using my trial mech Jager S with no armor. I would do 64 damage while getting shredded by 6 other mechs because I didn't understand that this wasn't Halo3, and I would be hyped to get in the next battle so I could drop 80 on those losers from the other team next match. (all while thinking SWEEEEET! I just made 450k in cbills for that). I was sooo bad and my KDR was abysmal.

I took a few months for the light to actually go off and me understand that teamwork really matters as well as enough time to gather enough cbills (hadn't bought any premium time yet or hero mechs) to buy the things that made playing easier. (Mechs, double heatsinks, different weapons.)

Looking at my stats I see a trial drg-5n that has .06 kdr and aws-8q with .17 kdr and a slew of trial mechs from back in the day that you all remember were just unplayable mechs that I am still trying to recover from KDR-wise. So to answer the question my KDR stands at 1.55 now and I pug 95% of the time and I know to get it to 1.5 means that I've been doing higher than twice that amount while inching my way back up the KDR ladder. I know that number spikes when I run with members from my house.

I'm sure its that way for many pilots who finally GET IT once they GET IT. The Kdr number means something...but what exactly...who knows. I guess it means what you want it to mean.

 

I agree and disagree with you. For some roles you just aren't going to get a lot of kills, and you listed a good example of one. However, once someone has played a statistically significant number of games in several different types of mechs KDR does become an indication of how effectively you contribute to a battle when looking at your overall KDR. It may not apply very well for specific mechs, but for your overall account it does tell you something. Anyway, you point is well taken that a single stat like KDR doesn't necessarily tell you how a good a player is. I think you really need to look at several statistics, damage, kill efficiency (damage/kills), kills per battle, etc, when evaluating a players performance.

Just looking at KDR though I feel that I'm not contributing to a match unless I'm taking out at least one enemy mech on the other team. If I take out 2 or more than I'm exceeding the average, and if I don't take out any then I'm not pulling my weight. So here is how I would rank KDR for someone who mostly pugs.

KDR < 0.5 - Bad (Getting killed more than twice as much as killing)
KDR 0.5-0.9 Below Average
KDR 0.9-1.1 Average (Roughly taking out 1 mech per match)
KDR 1.1-1.5 Above Average
KDR 1.5-2.0 Good players but can't always carry a match when the team fails
KDR > 2.0 Great Players who are capable of picking up the slack for bad players, maybe even carrying matches

Personally though I'm trying to improve my game by getting a better kill efficiency. Killing a mech while doing the least amount of damage to them and not just kill stealing. I figure if I can take a mech out of play faster then my team ends up having more guns on the field, ends up winning more often, and as a result my KDR goes up too.

 

Am I the only one when cash grinding that doesn't worry about kills? Personally the perfect round for me in "cash grind mode" would be a win with 12 assists (pays more) and 800-1k DMG. I never drop alone and usually there is at least someone in my group that actually needs the kills for XP so I try to leave the last hit for them. That being said one thing that really has been bothering me with the MWO community is all the definitive statements made. For example "this mech will always beat this mech" or "KDR is a completely worthless metric".

Yes KDR has a place on evaluating a players skill but as already stated you can have a .00001 KDR but if your winning 99 out of 100 games your probably an amazing player. You really need to look at win/loss, kill/death, damage done, and somehow put a metric on any support role you might be doing. Does it matter if your in an Atlas and you die every round with 500+ DMG if because of your sacrifice your team always wins? I would lean towards no.... That being said I also agree that less then 1 KDR even as a light probably means you need to work on your skill. Most lights should be able to last hit and get a kill in most rounds before taking that rolling mech nap.

In the end of the day until/if they release global stats it really doesn't matter, no one knows except you and the people you tell (lie?) to. =P

 

I think the Damage/Match Ratio is even telling more of a player's skill than the KDR. Honestly you can compare DMR inbetween Weight Classes.

 

Yes and No, you can get a lot of kills with low damage by killing a mech with a few well placed alphas to the rear CT or you can get 500 damage just striping apart and Atlas. Which one is truly the better player? High damage is generally good, but I would think killing or at least stripping a mech of its weapons in the most effective way (least damage needed) would be better.

 

Most times you need to deal plenty of damage to make the Kills than having chances to place enough accurate Hits on weak components. At least it is against skilled enemy pilots. My Mechs with high KDR are usually mid range fighters with Dakkas.

 

I do agree, but I still don't feel that damage is the only metric required to get a good handle on a player's skill.

 

Yeah sure, i agree - still not sufficient. But imo better skillpointer than KDR. Just from my feeling and observations.

 

A few months ago someone posted a bunch of polls about what everyone's best average damage per round was in each weight class, and this discussion was debated there. One thing I thought about at the time was seeing what your DMG/Kill ratio was for each mech. This would basically be a kill efficiency (or kill steal) stat.

My numbers were pretty interesting, as my worst mech by far was a HBK-4P which did a ton of damage, but never got any kills (DKR=856). I hadn't played it since like January of 2013, though, which is likely why the numbers were so skewed. My best mech was surprisingly my COM-2D with a DKR of 199, though these were definitely kill steals since it's damage mainly comes from SSRMs which you can't aim. The next best was my TDR-5SS with a DKR of 242, which actually made sense, since I had only run a 7x MPLas build on this mech when I ran the numbers and that thing was designed to get in close and core CTs.

As a summary, my average DKR for each weight class were:

• Lights: 293
• Mediums: 471
• Heavies: 417
• Assaults: 509
• Overall: 428

So what this tells me is that I basically kill steal in my lights, and am most efficient in my heavies.