KDR

Very interesting way to look at it and thank you for correcting me on that statement. I have never done the math to calculate in a component destruction + kill cbills = kill assist cbills. That being said for me at least it still is probably much more feasible to be able to get an easy kill assist on every enemy if I'm not to focused on killing the every mech I see. (Purely looking from a cash grinding perspective).

Almost off topic but I can't stand it when people continue to engage a mech that has no weapons for the easy kill (padding that KDR?) especially when they score is close or my team is down.

One question that I've never actually looked into, but I'm sure someone here knows. How does crash/disconnecting from the match after your mech has entered the game effect your KDR?

 

The horse has been beaten many times before. There is no magical statistic that will tell you whether a player is good or not.

For myself, I am very big on statistics because it is the only way to you can real about your improvement as a player. However, I can't go and compare myself to other players because their situation is different. There is no one out there that plays the same game as you because every game is different. We all play different mechs, at different ratios, in different brackets, at different times, against different people.

If you were trying to gauge yourself as a player, all you can do is try to improve on your own numbers and play an increasingly high-level and consistent game. When I get into a good groove, I can do 30 game stretches where my KDR is above 4 and average damage above 600. I know that is about the pinnacle of my personal skill with that mech, so I try to duplicate that success continually. I scribble little notes above my games about what went well, what went poorly, and I try to not make the same mistakes again. When I go on bad stretches, I take a step back and try to identify what I've been doing differently.

As for the original question, I think MntRunner's scale is pretty fair for pugging. Michael's is probably a bit off because I'm well into the 'you don't pug' sections and I don't consider myself a great player; good but not great.

 

Another issue with counting on Assists is that you are relying on someone else to actually kill the mech in order to get the cbill and xp bonus. If you get the kill yourself then you are guaranteed the bonus.

[quote author=Sassafras link=topic=5797.msg37820#msg37820 date=1393520645]

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.

[/quote]

I think they give us enough information that we could normalise this or calculate a kill steal coefficient to help determine if you're efficiently killing a mech or stealing a kill from someone else who has done most of the work. I'll have to think for a little while how to do this but I'm pretty sure it's possible.


[quote author=e92 link=topic=5797.msg37823#msg37823 date=1393521729]

For myself, I am very big on statistics because it is the only way to you can real about your improvement as a player. However, I can't go and compare myself to other players because their situation is different. There is no one out there that plays the same game as you because every game is different. We all play different mechs, at different ratios, in different brackets, at different times, against different people.

[/quote]

This is probably one of the reason why PGI only let's players view their own statistics.

 

In putting together some competitive 12 man matches over the last month I went so far as getting all my pilots to send me their stats from the MWO site and put them into the spreadsheet from the QQ Mercs toolbox. In pouring over all the stats what I leaned towards was the damage per match stat. However, I wouldn't even consider a mechs stats unless the pilot had at least 50 games in that mech (I preferred 75-100 minimum, but had to open my search parameters). KDR is fine but is not the only stat to look at and definitely not one that a pilot should worry about (I'm pointing at you random pug guy who runs and hides in the match).

 

Personally, I think a weighted consideration of Damage/Match and KDR can be combined to give some estimate of a pilot's skill, but, the only true way to measure is dropping with them.

This doesn't even take into consideration ROLE driven mechs. Spotters, Tanks, Cappers and Harassers may never get 1000 damage and 5 kills, however, these pilots can be just as pivotal to a win as the brawlers who rain death! Four Atlas' can destroy most of an enemy team with their single lance if they have the support to distract and harass the enemy and control where the engagement begins. Does this make those harassers and support mechs any less vital to victory? Also, I have also had plenty of matches (including 12 mans) where a light team has won by capping while they were the only mechs left. Again, low damage an only one kill between three lights, however, they won the match...

Moreover, I feel pilot skill is much more than numbers. I have dropped with plenty of pilots who are phenomenal commanders and leaders who are only middle of the road at best for damage and Kills/Deaths.

On the same token, I have dropped with a LOT of players who rack up damage and kills like no other who have terrible communication, terrible attitudes and are all around unpleasant.

I am very lucky to have a group of friends that I can drop with who appreciate community as well as skill. I am certainly not the most feared pilot, however, I can hold my own. We all work on communication skills together and try to have fun.

End of the day, if I'm not having a good time playing, I don't care how great any player on my team can score hits and drop enemy mechs.

My 2¢

Chatt.

 

Neat to read all y'alls thoughts on this.

Right now my overall KDR hovers in what Michael would call pretty good and you don't pug drop often do you?. Buying new mechs and xp'ing them up typically causes a drop with driving mechs I'm more comfortable with/familiar with driving it back up. I've some that are abysmally low, all Commando's but the Death Knell, and my Kintaro's that are all above 3 KDR.

I'm not sure what stat I would say would be the one to measure success by. I tend towards C-bill earnings per match or damage done per match. C-bill earnings because military activity revolves around economics. Damage done because in military activities killing the enemy isn't always the objective but imposing your will upon them is.

The problem with using C-bills is that after a certain point they become meaningless. This level might vary between different folks but for me that point came at the 100 million mark. With that much I can buy any mech I want, trick it out, and make enough C-bills xp'ing it up that I'll have earned back what I spent by the time I'm finished with that chasis.

The problem with damage done is that the way this game plays is that kills is more important than strategy or tactics. Without something outside of the matches we play effecting the game damage alone is meaningless. CW I hope brings something to the table on this otherwise this game is just shallow.

In the end Cpt Chatty's measure is probably the best, is the game fun or not?. As shallow as MWO is right now it can be loads of fun. Since I only ever pug after a week or two of steady playing I end up wanting to kill my 'own' team more than I want to kill the 'enemy' team and that just sucks all the fun out of it for me. I expect to die each game and don't even care if I win or lose each game but my God the Mountain Dew swilling, Cheetos eaters I end up grouped with piss me off to no end.

 

This is based only on my own experience, but on my stats page, I have a total of 1 match played on the CN9-A, which crashed before getting into the game. I have 0 wins, 0 losses, 0 Kills, and 1 Death, so it looks like disconnects do affect your KDR, but do not affect your W/L.

 

KDR means jack shit.

I had just below 2.0 when I stopped playing.

I remember after beta went open with all the crazy UI/framerate/hardware bugs, I used to simply AFK grind variants for literally a month straight when it was super bad. It wasn't even worth actively playing there for a minute. That absolutely skewed the fuck out of my KDR.

And yeah, if you pug at all, it isn't going to be super high.

KDR means literally nothing. There are too many exceptions to the 'range' theory, which is bs. It doesn't reflect ELO; I was in top ELO and yet my KDR was 2~. Think about it.

And if any of you doubt me, then I'll bother to reinstall and challenge any doubter to a duel for account.

 

Agree with the 0.9-1.1 being average
less being needs a team.

We have an award in my Unit for 2.0 KDR in a Mech we classify that mech Pilot as being an Ace.
and if he attains a 4.0 or better a double ace.
We give medals and everything (lol)
There is a Minimum number of matches needed to qualify. because a pilot who plays one good match and kills 4 mechs cant earn Ace.

P.S. when i started playing MWO i had about a .8 KDR as I was experimenting with different builds and the game mechanics and lost more than i won. then I got an understanding of the game and Joined a Group and Now my KDR has been built up to 1.4+ over all. I feel that was Hard as I was handy capped by my poor start.

I Whole Heartedly WELCOME the new Stat Wash. this will allow me to test my Mettle and see where my true KDR is.

 

I agree with those who say that around the 1.0 KDR is average. I (almost) always pug, and if I drop in a "group" it's a "group" of two. At least in a pug game, I feel that if you pull in a 1.0 KDR, I am happy. You killed someone before you died, so we're generally even. Of course, I always try to take into account what mech you are piloting and what you killed. In terms of the actual match, if you are piloting an atlas, and get a lucky headshot on a locust, but do no damage and die early, I feel as if that is a "low" 1.0 KDR, and you didn't contribute much. Conversely, I've seen some very nice medium pilots do massive damage/have a well place air/arty strike where they do significant damage to the enemy team and come out with maybe one kill before they die. I give that a "high" 1.0 KDR. There are way too many factors in this game that it's difficult to just look at stats to determine "He's got a 4.0 KDR, he's great" (maybe a bad example, because he probably is great if he has a 4.0)

Another think I would like to mention, that someone also touched on, is how else they attribute to the game. Often in pug matches, I find myself trying to desperately run around in my Spider trying to cap points in conquest so we don't lose, while the giant pug ball rolls around blindly chasing Firestarters. I feel that the lights/mechs working for the team are "better for the team" than the guy in the ac40 jager trying to get his extra kills.

Of course, I am no expert at all in statistics, or this game, I just follow baseball a lot, and we (baseball players/fans) seem to love our stats. Also, my point of view is more pug oriented than team oriented, as that is where a majority of my play time comes from.

I also welcome the stat cleansing, as I think my KDR will be significantly improved. (I had a bad, bad phase where I thought it was a great idea to run crawl around in my 4 AC5 Single Heat Sink no backup weapons, a bajillion tons of amo CTF-4X.) I just worry that I'll have a bad string of games and be back in the under 1 KDR/Win Loss record.

 

I freely admit I copy off all my stats on the same date each month and throw them into a spreadsheet which calculates a lot of metrics like damage/kill, damage/match, kills/match etc. It even produces nice little pretty tables tracking my progress without me having to do any work.

What it doesn't do is in any way reflect on you using that chassis as odds are I'm not using the same build, fighting with the same people etc. In fact because I have a lot of mechs I find I deliberately make every variant in a chassis different.... optimal they're not maybe but I try to get them to fit my playstyle or push me to develop it. For example I'm truly awful with LRM's so every now and again a poor new mech gets designated as a guinea pig and I force myself to grind it all the way to mastery to work on my effectiveness. Doesn't help my stats much but at least I know why and can see a steady improvement ... from awful to average.....

Its mainly a mental prod to myself so I can see my progress, push myself and take the appropriate mech/setup to fit a role in a lance or 12 man if one isn't dictated.

I know a few people with insane KDR but then they essentially probably play 5? mechs each. Not even 5 chassis just 5 specific variants. Personally I've never really thought much of that as you're just working by rote. Pull a good performance in a locust though and I'm impressed.

Pull in the stats I do in my poptart meta Slayer and I'm really not impressed despite them being quite good.

 

Buyine and leveling causes a drop, having buggy connection causes a drop, getting disconnected seem to cause a drop, playing that piece of crap variant of a chassis just so you can elite others causes a MAD drop.

But in the end, who cares! It is having fun and in game performance that is important to me.

 

Not convinced it does say that as it could be a reflection of artie usage, running in packs, precision meta volleying... only you know.

Class - Damage/match - Kills/match

A 332 1.24
H 322 1.15
L 377 0.81
M 312 1.13

Not sure what that tells you but its certainly a hint I don't use artillery and prefer precision weaponry.

What it doesn't show you is the variance in a category:
Dragon Slayer: 228 - 2.35
AWS-8V: 464 - 0.72

Hint one is an LRM boat ....

For amusement a mech most people do well in but I can't get to work...

JR7-F: 624 - 0.56

It was my first light mech of course but then the bare stats won't tell you that (its also not that statistically useful as I only ever payed 16 matches in it).

 

All this no longer matters because TOMORROW IS RESET DAY!!!!

Now you will be able to track your stats much more accurately as you will have had experience with the game, wisdom with your builds and practice on the field. Also, since a lot of you won't be grinding mechs anymore you will be playing with fully elited or mastered equipment and that will make a difference.

 

Mmmh I am still steadily climbing to 125 (closing in on 100) .... aka one of every variant (excluding champion mechs) so the "grind with a goal" continues. Call it a mental achievement.

 

KDR says nothing about Mech Type, assigned task, etc.

I reckon its hard to have a decent KDR in a Locust, or Commando. But if your Team knows where the enemy is, nicely tagged, narced, spotted, 9+ assists, led chasers in the hands of a waiting lance, confusion in the other camp in preparation of a push, capping far off Bases... and some lucky guy blasts your legs off, with a happy hit, you have zero kills. Am i a bad player? No i helped my team.

A good KDR in a great, damage dealing Mech is far easier. Are you a better pilot for choosing the stronger, Metamech?

Also: Luck. Did i have the last shot on that guy everyone pounded? Am i a better player for hanging out with the ganking squad?

So.. a good KDR had what significance? No teamplayer? MWO is more than just straight killing.

 

1: I don't see how you can talk about Locusts and Commandos while also talking about helping your team. The Locust is clearly a bad choice whose role can be done better and more effectively by other mechs. I don't have any experience with Commandos, but I suspect that Spiders, Ravens, and Jenners fill that role better as well.

2: 1,500 damage Raven builds are a thing. I've never sniffed that with any other mech I pilot. Now, are you a better pilot for choosing this mech? Perhaps not. But you most definitely are a better teammate for choosing this over a Locust or Commando. (Incidentally, that 1,500 damage comes with a number of kills, generally.)

3: Addressed the teamplayer aspect. You've confused making sub-optimal mech and build choices with being a good teammate. A pilot can make an optimized build choice for each mech class and have a great KDR, all the while helping the team. These are not mutually exclusive goals.

Finally, over a period of time, the luck factor for killing blows becomes a non-factor. If you find yourself getting the "lucky shot" often, it's not really luck.

This is not to say that KDR is the be-all, end-all stat in the game. But it's not meaningless.