135 replies to this topic

[Ursh]

As someone alluded to earlier, a lot of it can depend on your mechs load out and playstyle.

One of my battlemasters has a 4.0 kdr because I use it to ambush people with a bunch of pulse lasers, and it only takes 2-3 45+ damage alphas to kill a fresh mech, and considerably less to kill people later in the match.

I've got 57 mechs, and my overall KDR since statwipe is about 1.09, and I have a slightly positive win/loss record as well. I've lonewolfed about 75-80% of matches since I came back from a six month break.

I'm also incredibly impatient and aggressive, which often leads to deaths.

[Mister Dubis]

I wouldn't mind if they changed it from KDR to something that deals with assists and dmg done. I feel like that would be a much better indicator of how good you are.

[Dunning Kruger Effect]

Quote

I wouldn't mind if they changed it from KDR to something that deals with assists and dmg done. I feel like that would be a much better indicator of how good you are.

Disagree with this statement.

(#1) Assists might mean something if they weren't the easiest stat to pad in the game. Brush a ERLL over every enemy mech and "voila" you have 12 assists. I might have done only a dozen damage points and got 12 assists. Deliberately getting assists takes far less skill than "kill-stealing."

(#2) Damage done is a bit more debatable, it's true. However consistently getting high damage and very few kills might also indicate an inability to target weak components.

i.e.

Averaging 4 kills a game with a 4:1 (4:00) KDR and 400 damage
--vs--
1 kill every 4 games with a 1:4 (0.25) KDR and 800 damage

...is the one with the higher damage truly a better player? Maybe he is just spraying damage wildly all over the enemy mechs, whereas the "low damage" player finished off all his targets swiftly, targeting weaker components or XL side torsos...

Edited by Dunning Kruger Effect, 04 June 2014 - 04:13 AM.

[Kinski Orlawisch]

Wayne?

Stop checking it. Play and have fun.

[Ursh]

They should have some kind of combined ratio, sort of like how the NBA does PER- Player Evaluation Ranking

For MWO it would combine kills, assists, damage per match, cap assist, spotting assist, savior kills, component destruction, TAG bonus, NARC bonus. Stats would have to be weighted so that having a massive advantage in a single stat couldn't elevate your PER beyond a certain point.

I think you'd see a more comprehensive picture of someone's contribution that way. Everyone has had games where they felt like they did almost nothing, but walked away with 4-5 kills because they happened to get the last shots in on mechs other people were destroying. Other times you've walked away from a game where you totally carried your team but you had one kill and eight assists to go along with your 750 damage.

[Exilyth]

The best KDR is 0.0 because in a perfect world you'd disarm your enemy and win by objective every match

[Bulletsponge0]

Dunning Kruger Effect, on 04 June 2014 - 04:07 AM, said:

Disagree with this statement.

(#1) Assists might mean something if they weren't the easiest stat to pad in the game. Brush a ERLL over every enemy mech and "voila" you have 12 assists. I might have done only a dozen damage points and got 12 assists. Deliberately getting assists takes far less skill than "kill-stealing."

(#2) Damage done is a bit more debatable, it's true. However consistently getting high damage and very few kills might also indicate an inability to target weak components.

i.e.

Averaging 4 kills a game with a 4:1 (4:00) KDR and 400 damage
--vs--
1 kill every 4 games with a 1:4 (0.25) KDR and 800 damage

...is the one with the higher damage truly a better player? Maybe he is just spraying damage wildly all over the enemy mechs, whereas the "low damage" player finished off all his targets swiftly, targeting weaker components or XL side torsos...

So some kind of stat like damage per kill....although that's not perfect by any stretch of the imagination either

[Alaskan Nobody]

Bulletsponge0, on 04 June 2014 - 10:41 AM, said:

although that's not perfect by any stretch of the imagination either

That is the problem though - in a game with as many variables as MWO has - the only stat you can really epeen with is the W/L ratio (hence all the upset over Elo's implementation)

[Bulletsponge0]

Shar Wolf, on 04 June 2014 - 10:44 AM, said:

That is the problem though - in a game with as many variables as MWO has - the only stat you can really epeen with is the W/L ratio (hence all the upset over Elo's implementation)

true

but that must mean I have a tiny e-peen

[Alaskan Nobody]

Bulletsponge0, on 04 June 2014 - 10:47 AM, said:

true

but that must mean I have a tiny e-peen

Humility can be something worth being proud about.

[xMintaka]

Shar Wolf, on 04 June 2014 - 10:44 AM, said:

That is the problem though - in a game with as many variables as MWO has - the only stat you can really epeen with is the W/L ratio (hence all the upset over Elo's implementation)

Which is skewed by playing in a group or not, trying new tactics in game, etc.

It begs the question really, why the need to epeen? We are (mostly) adults with lives outside of gaming. We game for fun. Getting hung up on stats within those games seems ridiculous and counter productive to the fun factor.

I love stats, I check mine after every play session, but it's not the end of the world if my KDR drops a point or three. The only stats I'm actually serious about maintaining are on the ~5 mechs that I enjoy playing the most and feel I should be getting kills and actively contributing to wins. But again, dropping a few points is merely a time to consider where I went wrong and how to improve.

[Alaskan Nobody]

Lunatech, on 04 June 2014 - 10:57 AM, said:

It begs the question really, why the need to epeen?

A great deal many reasons - ranging from not having anything else in life to brag about , to finally being better than someone at something, to trying to use it to support your arguments (IE my best weapon is the LRM - that proves it is OP!) to several other reasons.


For some people - the epeening is the fun factor.

I would sit about where you seem to on the matter though.

Although I don't seem to care as much about my stats
(you are giving them the 15 minutes or so to update, right?)

[xMintaka]

Shar Wolf, on 04 June 2014 - 11:08 AM, said:

I would sit about where you seem to on the matter though.

Although I don't seem to care as much about my stats
(you are giving them the 15 minutes or so to update, right?)

The only reason I care about the stats is that they help me improve. Checking them gives me a frame of reference as to how well I was playing in those mechs on a given day and any changes I might need to make in order to continue improving.

Even in this case, the stats I check are vastly different for some mechs. I keep track of my W/L and KDR in my Cicada's, Jenner's, Flame and Firestarter's because all those mechs are geared for damage and kills (and I feel comfortable piloting them, unlike my laughably bad KDR in poptarts).

But my newly equipped NARC Commando I only keep track of W/L because I'm not likely to get any kills with a Flamer and SRM2.

My Locust(s) and DRG-5N are geared to amputate components and flee so they also dont get many kills but contribute to the team by crippling mechs. For these a "Components Destroyed" stat would be fantastic.


(Yeah, I usually check my stats later in the day when there's ****** all to do )



You make a valid point though. We all enjoy this game for different reasons, it's not for me to comment on those reasons. The only thing I'd like to say about epeens is along the lines of that "religion is like a penis" saying.

It's fine to have one. It's fine to be proud of yours. But please, please don't wave it around in my face.



Well, I didn't think a Canadian forum would censor that. Guess they've been reading up on British "swears" lol.

Edited by Lunatech, 04 June 2014 - 11:38 AM.

[Celem]

Im with the various folks who have suggested ignoring KDR entirely. Note I only pilot light mechs and have a sub 1.00 kdr on all of them but the Ember. I would hotly argue with anyone that suggests this makes me a poor player.

MWO is not CoD or BF. Kills really don't matter at a personal level. As long as the enemies die and you pull your weight, nobody should care who strikes that final blow.

Why?:

1) Mechs are highly durable compared to most FPS characters, there are few situations where you and you alone single-handedly slaughter an opposing mech. Even if he had no damage, he's distracted by your team, trying to avoid being ganged-up-on, watching for your lurms etc etc. Also due to this mechanic its possible, and indeed common, to do most of the work and have someone else get the kill credit.
There is NO kill stealing in MWO. NONE. I hate when people say this, its a tactical team game, the concept is not possible. He died, we didnt, everyone gets paid = Win.

2) This is recognised in-game, you will earn just as much exp/c-bills through assists as you will through kills, the difference (if any) is negligible.

3) All mechs are not equal. I kinda expect my assault pilots to out-kill my light mech. If they dont then wtf did I bring them for? Likewise in a semi contradiction to point #1 an assault can 1-shot a healthy light mech, the light needs some true skills to solo the assault (its doable, but a lot more effort and depends on some privacy, your team ties up his mates so you can do your job)

Ultimately its all depending on your role. Many light mech pilots will see a low KDR compared to other weight classes at least in the early days of their MWO career, thats not because you suck. You should however be seeing braces of assists and spots which get you paid just the same. (There are always role-based exceptions. If you are a Ember/Hugin packing MGs or fit for striking like an fs9-a ml boat then your job description has changed and you ought to be seeing some kills.)

Pay attention to the Win/Loss ratios, the KDR is nice to see if specific mechs in a given class are performing like they should be, I do watch my Ember to make sure im getting the job done compared to a Tag scout or anti-light skirmisher.

Edit: and unfortunately Luna, the one you picked there has other connotations in other English cultures. I know how Brits use it and how you tried to, but other meanings pop first to mind for much of the rest of the world.

Edited by Celem, 04 June 2014 - 11:51 AM.

[InspectorG]

Bulletsponge0, on 04 June 2014 - 10:47 AM, said:

true

but that must mean I have a tiny e-peen

Not the size of the mech but how you pilot it!!!

[Bulletsponge0]

InspectorG, on 04 June 2014 - 12:00 PM, said:

Not the size of the mech but how you pilot it!!!

so fast and furious, constantly crashing into things, before finally putting a shot on target

[Dawnstealer]

Mine's pretty bad, but that's because I have a blast running Locusts and a spotter-style Commando that rocks an ECM and single MPL (along with a NARC and two tons of ammo). Especially that second one doesn't kill too many people, so it's always way in the negative of the death scale.

Locusts get one-shot more than anything else, so they don't survive long.

In my legit brawling, mess-things-up mechs? 1.5 or so. I'd say 1+ is a good goal, but really, I think if you're the doorbuster (like I usually am), your KDR is going to be a little less impressive, but SOMEONE has to be first through the door.

For me? It's just a game, so I don't mind "dying" if my team's moving pixels kills the other teams moving pixels. How's that for immersion?

[Mazzyplz]

i will explain this finally for all newbs to easily grasp, in plain english;


kdr will depend on the mech and role (loadout) you play - generally less than 1 is not good! but some mechs kill more than others, keep that in mind and look at your per mech KDR not the base one (that one is useless as even a yardstick)
more than 1 is desirable for improvement measurement upon your own tactics as has been stated before.


losing kdr (going into the negative) is undesirable, unless! and this part is key: unless you can trade KDR for WINLOSS ratio.
if you trade some KillDeathRatio to gain WINLOSS ratio then you will step up into a higher elo with tougher opponents. because WINLOSS is what actually determines everything. so your suicide push that won the match by distracting the enemy? yea that actually paid off, but only paid off by punishing you with tougher competition; enjoy!!

anyhow, the good player will keep a good kdr without padding it when he could trade KillDeathRatio for winloss-ratio in an all or nothing push.
padding your kdr only to survive is only excusable when the fight is completely lost and capping loss is imminent

that's all

[focuspark]

Nikkoru, on 30 May 2014 - 07:46 AM, said:

What's a good kill/death ratio?

One you can live with.

[Dawnstealer]

Mazzyplz, on 04 June 2014 - 01:10 PM, said:

i will explain this finally for all newbs to easily grasp, in plain english;


kdr will depend on the mech and role (loadout) you play - generally less than 1 is not good! but some mechs kill more than others, keep that in mind and look at your per mech KDR not the base one (that one is useless as even a yardstick)
more than 1 is desirable for improvement measurement upon your own tactics as has been stated before.


losing kdr (going into the negative) is undesirable, unless! and this part is key: unless you can trade KDR for WINLOSS ratio.
if you trade some KillDeathRatio to gain WINLOSS ratio then you will step up into a higher elo with tougher opponents. because WINLOSS is what actually determines everything. so your suicide push that won the match by distracting the enemy? yea that actually paid off, but only paid off by punishing you with tougher competition; enjoy!!

anyhow, the good player will keep a good kdr without padding it when he could trade KillDeathRatio for winloss-ratio in an all or nothing push.
padding your kdr only to survive is only excusable when the fight is completely lost and capping loss is imminent

that's all

Ugh - you're making me like Marik. Stop.