225 replies to this topic

[Roadbuster]

It depends if you play in a group or in pugs most of the time.

If your K/D ratio is 3+ you either play in a group more often or run a cheese build.
If you solo pug +3 is pretty much impossible to keep up if you don't run a specialized mech or steal kills.

After 3200+ games I'm sitting at K/D 2.05 and a W/L 1.07, about 90% of them in pugs the rest in a team of 2-3.
I'm not going for kills, I try to do my best to win the match.
W/L is more important than K/D.

EDIT: Mechs used - all except Commando and Spider

Edited by Roadbuster, 08 April 2013 - 11:06 PM.

[Chavette]

Roadbuster, on 08 April 2013 - 10:44 PM, said:

It depends if you play in a group or in pugs most of the time.

If your K/D ratio is 3+ you either play in a group more often or run a cheese build.
If you solo pug +3 is pretty much impossible to keep up if you don't run a specialized mech or steal kills.

After 3200+ games I'm sitting at K/D 2.05 and a W/L 1.07, about 90% of them in pugs the rest in a team of 2-3.
I'm not going for kills, I try to do my best to win the match.
W/L is more important than K/D.

Guess me and atleast 6 more from this thread are Martians then.

All because you didnt hit it.

Not like it means anything, but thats not the point here, the point is your denial.

[Ghogiel]

Erasus Magnus, on 08 April 2013 - 10:38 PM, said:

then i think you arent in the elo bracket you belong to (you need to be way up). or you are a good shot and doing lots of kills but dying often and losing the match.
thats the difficulty of isolated stats like this.
you cannot create a clear picture of your doing ingame.

in my spider 5 d i have a k/d ratio of over 5 and a win loss ratio of over 3. this is due to the fact that when i see that we cannot winby kills, i run and try to capwin. as an assault this option goes right out of the window. you either fight or die. in conquest mode i often try to cap 3 additional basesto make the enemy mad, then turn around to help my team with the brawling. so i often arrive and find enemies in bad shape.

My W/L prior to Elo wasn't very good. I think I won maybe about 60% of my matches.

Now that Elo has come in, I run with a group probably a bit more often then I run PUG, because it matters more, and I win more now.

I started a little experiment with my STKs.




5M is pretty much all PUGged, which I did after the 3F, which is pretty much all premade.
W/L suffers a lot sure, but KDR is easy enough even pugging, after 140+ matches into being seeded. not even including the matches I played in my DDC lol.

[Erasus Magnus]

Chavette, on 08 April 2013 - 10:48 PM, said:

Guess me and atleast 6 more from this thread are Martians then.

All because you didnt hit it.

Not like it means anything, but thats not the point here, the point is your denial.

denial of what? K/d IS meaningless as an isolated stat. W/l is, too, because you can win by contributing exactly nothing to the teams effort.

[Arete]

*Sigh*

I love when this comes up in Elo matchmaking games. If you are below "your" skill level rating-wise, your KDR will be huge. If you're roughly at where you should be, it will rarely exceed 3, if you're at 1 you're fine.

This is where I usually say "Your rating is the best indication of your skill", but since we can't see it my point is moot.

[Roadbuster]

Chavette, on 08 April 2013 - 10:48 PM, said:

Guess me and atleast 6 more from this thread are Martians then.

All because you didnt hit it.

Not like it means anything, but thats not the point here, the point is your denial.

If you can take down 3 undamaged mechs with any mech you drive, all ranging from light-assault with a mixed weapon loadout, then more power to you.

The above is my opinion.

[Rocdocta]

KDR is a bit of a nice stat that is really meaningless. I regularly do alot of damage each game in my jenner but rarely kill anything. i am more likely to be killed than do the killing as a couple of lucky AC20 hits = dead. 150 kph only does so much in a pug.

[OneEyed Jack]

I'd say you're fine if you're at or around 1:1 or better, and/or at any level and improving, as long as you aren't stat-padding to the detriment of your team.

The KDRs that are significantly higher than 1:1 are pretty much exclusively either pre-made players that boosted their stats artificially with PUG-stomping or stat-padders who over-inflate their stats by doing things like hiding until they can rush out and snag a kill-shot (or spend half the game chasing the mech with no guns), often losing the game because they were too afraid to expose themselves to fire. There's also the bandwagon-jumpers who only play whatever is currently considered OP. They're kind of like the stat-padders, but more likely to get into the fight, since they tend to favor more kills over less deaths as a way to increase stats. Except right now when they both seem to be playing pop-tarts.

[Chavette]

Erasus Magnus, on 08 April 2013 - 10:59 PM, said:

denial of what? K/d IS meaningless as an isolated stat. W/l is, too, because you can win by contributing exactly nothing to the teams effort.

My last line said just that. Its meaningless. But why do people get upset about and call killstealer, groupstomper, cheesebuiler when someone says hes got a high one?

[StandingCow]

Zylo, on 08 April 2013 - 09:57 PM, said:

Players can get a really high W/L ratio just by capping...

That isn't a bad thing though, it means you out-thought/out-maneuvered your enemy, so you deserve the win.

[Ghogiel]

1:1 kdr isn't good.

It's medium or average.

[KinLuu]

Chavette, on 08 April 2013 - 11:15 PM, said:

My last line said just that. Its meaningless. But why do people get upset about and call killstealer, groupstomper, cheesebuiler when someone says hes got a high one?

Because you feel the need to wave yours in peoples face like a baboon feels the need to show off his red butcheeks.

[Kiiyor]

Mine was around 4 for ages, all PUG. For some reason I was a god of destruction in the Awesome 8Q, of all mechs. For the first 20 matches or so after the utter bloodbath that was open beta, it was close to 7. Noobs were popping all over the place.

Then came the cataphract. I persevered with mastering that albatross until my KDR was around .9. I just couldn't gel with it. I've clawed it back to 1.99 now, which is probably where I'm meant to be.

[Chavette]

KinLuu, on 08 April 2013 - 11:25 PM, said:

Because you feel the need to wave yours in peoples face like a baboon feels the need to show off his red butcheeks.

I didn't even write mine at first I just said 3+ is good and alot of people wrote their 3+ scores.

[Kiiyor]

KinLuu, on 08 April 2013 - 11:25 PM, said:

Because you feel the need to wave yours in peoples face like a baboon feels the need to show off his red butcheeks.

If I was a baboon, I would work it. Nothing but assless chaps. I'd shake it like a Polaroid picture.

[Hayashi]

>1 means you're above average. 0-1 means you're below average. Remember, for every kill there is a death except cases where you blow yourself up. Therefore the number of deaths on the whole is greater than the number of kills on the whole. If you kill more than you die, you're generally doing well. But this is only very broadly speaking.

KDR depends greatly on what your role in a team is. Assault and Hunchback players tend to have lower KDRs as part of their job is taking damage for everyone else - meaning they also die more often. Light mechs and snipers/LRM boats have the job of dealing more damage are not really supposed to take any, so they tend to have higher KDR.

Chassis also changes a lot. My own KDR varies from 0.94 in the worst mechs in the game (COM-3A, SDR-5K) to 7+ in the cheesiest mech (RVN-3L). People who cheese in a RVN-3L often, and even drop in 4mans with nothing but RVN-3Ls in the 4-man, can reasonably be expected to exceed 5, often 10 in KDR.

Unless they are really terribad.

Also consider lag factors. If two people shoot at an enemy mech with enough damage to destroy it at the same time, the one with less ping will get the kill every time.

So what is 'good' is hard to determine since it depends on what you're piloting and who you drop with. You could have a KDR of 4 in a RVN-3L with 4man drops and 50 ping, and you could still be a bad player anyway. And if you PuG only in a SDR-5K with a KDR of 3 and ping of 500, you'd be probably considered as godly by any player who knows what that means.

WLR better reflects your ability to contribute to a team in the long term. If you only killsteal your WLR will be terrible even if your KDR is good since you're not actually doing much to help. But WLR is still biased by how good your chassis is, how good your teammates are and even the time of day (some times have more good players than others, more cheeserunners than others etc etc) and isn't very good in isolation either.

ELO would probably be a better stat value to base your estimates of skill around, but unfortunately we can't see our actual ELO scores. And even so, there's a limit to ELO - once your skill level is high enough it may become impossible for the matchmaker to find an actual match for you within the matching timeframe, which means that even if you drop alone, your team may end up overpowered far more often than not because it contains one of you. If you drop with 3 friends that all exceed the ELO's normal limit... the 4 of you can very well have a greater level of skill than the entire 8 mech team on the other side combined.

Edited by Hayashi, 08 April 2013 - 11:48 PM.

[JohnnyC]

My overall is 1.74, but it depends on the mech.

Jenner-F is at 5.20
Stalker-5S is at 6.25
Stalker-5M is at 1.29
Stalker-3F is at 1.97

I don't think the overall matches quite right with the individual mech k/d ratio right now because they didn't start tracking individual mechs until well after they tracked overall.
That's almost entirely solo-launching. I have jumped on a TS server and dropped with a couple of 4 mans but only a few times and not for very many matches. I just generally want to drop in, play a few rounds, and then do something else.

[Faithsfall]

With kdr being alot by chance you could be lucky and get 4/5/6 kills a match without doing that much dmg, you were just the lucky one to hit at the time the mech is almost dead, surely a better stat is your avg dmg per match?

[GetinmyBellah]

I think what you should be looking for is average experience. I see a lot of high KDR's with only mid 500 avg xp. While I have a modest KDR I think my stat that truly matters, to me at least is an avg xp of 703, per match.

[Faithsfall]

Vegentius, on 08 April 2013 - 11:54 PM, said:

I think what you should be looking for is average experience. I see a lot of high KDR's with only mid 500 avg xp. While I have a modest KDR I think my stat that truly matters, to me at least is an avg xp of 703, per match.

Though with avg exp can't this be affected by premium time?