59 replies to this topic

[WILL WORK FOR AMMO]

Fake edit: I distinguish between player performance and team contribution. A player that contributes greatly to a team's victory does not necessitate good performance from that player.

I find that damage is not a great correlation to performance, with the following case in mind.
If performance were directly correlated to damage, then high pinpoint-damage builds would be performing optimally when they acquire maximum possible damage. However, this is counter to the purpose of those builds, which is to put as much damage as possible into one spot, and so we should be seeing less damage per kill or assist from these builds when the pilot is playing the build skillfully.
As an example, my K2 is set up to do a 50-point pinpoint alpha. Managing 600 damage on 3 kills and 0 assists on a win on skirmish means that I used 200 damage on average over each of those 3 kills. That is pitiful performance, unless those were heavily-armored assaults twisting away my damage.

It is not just a matter of build. A friend and I occasionally group as cicadas with roughly 64 to 72 pinpoint damage between us (depending on whether or not we run with ECM). These builds could be used to support heavies and assaults or just general, up-front damage to opposing forces, and we would expect wasted damage as we cannot hover in one place for too long. However, instead we sneak behind enemy heavies and assaults that are straggling behind and extract them. Again, we perform well when our damage is low and our kill count between us is high.

In sum, kills, assists, and damage should only be considered to measure your performance in light of your build's advantages and the strategies employed for those builds.

Edited by WILL WORK FOR AMMO, 12 October 2014 - 03:03 AM.

[oneproduct]

It also depends whether you mean average damage on a match you've won or average damage on all your matches. On matches where you lose you will do far less damage.

On average, the mechs of mine which I consider serious (i.e. not just playing to get basics) have an average of around 425 damage per game. This is by going to my mech stats page and taking the total damage divided by the number of games played in that mech. That 425 figure is nearly constant for all my mechs, despite their builds. For example:

Firestarter (6x ML): 79,174 / 185 = 428
Jenner (6x ML): 70,047 / 164 = 427
Raven (2x ERLL, ECM): 43,579 / 100 = 436
Cicada (2x ERLL, ECM): 42,718 / 99 = 431
Shadowhawk (2x PPC, 3x SRM4): 24,873 / 58 = 429
Catapult (2x ALRM20, 3x ML): 30,966 / 73 = 424
Stalker (2x ERLL, 4x ML, 4x LRM5): 57,174 / 131 = 436
Awesome (2x LPL, 5x MPL) = 30,094 / 73 = 412
Jagermech (2x gauss): 31,858 / 75 = 425
Jagermech (3x UAC5): 16,651 / 30 = 550
Dragon Slayer (2x PPC, 2x AC5): 81,608 / 151 = 540
Atlas (2x UAC5, 3x ASRM6, 2x ML, ECM): 46,455 / 86 = 540

This is for a 59% win rate and 2.04 profile-wide kill/death ratio playing in a mix of solo or casual group play.

My two jagermech averages are a great example of what people were saying about how damage doesn't always matter. 2x gauss jagers can contribute just as much to a fight while dealing less damage because you are more likely to hit the right spots and kill an enemy mech sooner.

My dragon slayer having a higher score than the others I would genuinely attribute to how the dragon slayer is a better mech than the others. Poptarting, assault class armor and good maneuverability allows him to survive longer than other mechs and push out more damage during that time.

For the atlas I'd say that the damage is inflated due to SRMs causing far more damage to sections of enemy mechs that are unimportant. Also ECM lets him survive long enough to do that damage.

Generally in the matches I play in for my bracket I'm one of the more performant players so based on that anecdotal evidence I would say that these would be quite good averages. The thing is people always remember how much damage they did in their good matches, but in good matches where it's a landslide victory you get a ton more damage than you would normally because you basically have free reign while the enemy routs. In the games you lose most mechs on the losing team tend to get below 250 damage or so.

In general it would be more useful for people to post their total damage / matches played to get a better idea of what their average damage really is. I wouldn't be surprised if most people's average damage was around 300.

Edited by oneproduct, 12 October 2014 - 04:56 AM.

[Mudhutwarrior]

Being a team game I am always happy to have high damage and no kills. Have done +800 and no kills many times. I'm not a frontline pilot like many of these skilled twitch shooters. Never mind I cant see well enough most of the time So I am happy to just contribute to the win.

Just as happy to disarm and opponent and let the other guy close the deal. I think there is greater cruelty to doing that Like leaving a mech smoking and legged and walking off. If I am in an especially good mood I will tell them in chat I will be back to Finnish you off when I am through. Did it to an Atlas In a match where he gave his coordinates to the rest of my team. I just artyed him instead. What a Joker.

Edited by Mudhutwarrior, 12 October 2014 - 03:34 AM.

[BattleBunny]

10 times your tonnage = a good match.

getting no kills while doing a 1000 damage I consider a sloppy game though. it means damage was spread out and not as usefull as it could have been when gunning a little better.

but since 1000 damage is rare for me I am still happy when I see it on the boards.


in the end the team win is what makes a match a good match I suppose.

Edited by BattleBunny, 12 October 2014 - 03:42 AM.

[Chrithu]

Burktross, on 11 October 2014 - 06:42 PM, said:

So, tell me. What are your damage values for good and bad?

The max Damage really depends on the Mech I am in and how the match went down. Thus generally my rule of thumb is: If someone did more than 250 - 300 Damage and scored a kill or two he did good and helped the team.

900 DMG isn't bad in a Mech that has such a low sustained DPS value. It means you survived pretty long while at the same time constantly being shooting at the enemy.

On the other hand my late experience in my scalpel build (Pulse Laser Jester) starts to teach me that such high damage values speak a language of sub-optimal target selection. If you lock on on targets shoot for the weak points and generally always go for a quick killshot and do your job well you usually do not score above 500 dmg. Your targets simply die before you are able to shoot enough at them to gather really high damage numbers.

That's not to talk anyone down, but in my view Damage alone is a pretty bad indicator of skill. Everyone above a certain baseline skill (that I'd set pretty low) can put together a high DPS build and have a lucky match in which he goes largely unnoticed by the other team and can shoot alphas at their assault mechs for 10 minutes.

That aside: Before the clan invasion I had a 1000+ dmg run in my 3 LB10-X Ilya Muromets. I caused a lot of ammo crits in that match.

Edited by Jason Parker, 12 October 2014 - 03:54 AM.

[Eboli]

I tend to go by a multiple of 10xmech tonnage. If I am running a Jenner a score of 350+ would be acceptable (10x35). But it really all depends on the role you are playing. If you are playing a narcing 3L you may not get 350 damage but have given your team players targets to launch LRMs, enemy distraction, targeting info as well as ECM cover. A 3 AMS ECM Kitfox can play an important role to the team without much damage either.

Another example is my specialisation - Oxide. If I have not put out 550 damage with kills I call it an average match with anything below 300 a failure. 700+ with kills - happy, 800+ with kills very happy, 900+ with kills - wow. If for some reason I have not got kills but multiple component destruction I would also tend to be happy knowing that I have made another team members life easier.

But generally 10 x mech tonnage (with some kills and assists) I would be happy with.

It all comes down to if you have a mech specialisation and your expectations. The better player you become the higher the expectations you give yourself.

Cheers from China! Can't play the game atm but at least I can read the forums
Eboli

[Alienized]

having high damage doesnt automatically mean you are skilled. you can aim but what else? a good aim is only part of beeing good. sometimes, doing less damage but help a teammate which leads to a win shows more skill than doing a crapton of damage but lose because you lack supportive teamplay.
and supportive play needs more skill than a good aim alone. especially in ECM mechs. (ecm sniperravens? waste of potential)

[Karl Marlow]

Well. I have only ever had a 1000+ game once. That was with my LRM Stalker back when LRM splash damage was broken.

Since then a really god game for me is in the 850 area and it doesn't happen very often. Usually if I am having a 850 damage kind of game I end up with 3-5 kills and we lose the game. That is an oddity about my damage numbers. If the team does well I usually have 200-300 damage unless I was killed early. maybe 1 or 2 kills with the occasional 500 point game with 3+ kills.. If my team sucks though I either get swarmed and can't do crap or I will have astronomical numbers. Either way it tends to be a Loss.

It probably has something to do with a target rich environment when you are outnumbered.

AS for what I consider good damage. I know some people like to try and average 10xfirepower on the mech. I don't really worry about my damage numbers per se. I suppose I always want to do at least 200. At the end of the day we can't all have 1000 point games. her is no way to do that much damage in a single match. So basing your games off of damage numbers is silly. Instead I consider it pulling your weight if you get at least 1 kill. If everyone could at least do that much we could get a win.

Edited by ThomasMarik, 12 October 2014 - 04:27 AM.

[Pjwned]

Levi Porphyrogenitus, on 11 October 2014 - 07:55 PM, said:

Damage is not an accurate indicator of skill. In fact, inflated damage scores are indicative of poor shot placement. Many Clan mechs, for instance, get far higher damage for the same battlefield impact, due to the nature of Clan weapons and how they tend to spread damage more.

That said, I figure 1 kill and 200+ damage is minimal acceptable performance, as that's basically accounting for your own slot in the team roster.

300-400+ damage, assuming several component destructions and 5+ kills/assists is a decent game, where you've done a bit more than your share.

500-700+ damage, assuming 8+ kills/assists and lots of component destructions, is a good game, where you've made a reasonably dominant impact.

More than that, with 11-12 kills/assists, means that you've had an amazing game.

1k+ damage with only 1 or 2 kills and only a few assists means that you really need to learn to aim better, or you need to stop farming damage on crippled targets and do something meaningful instead.

High damage might mean poor shot placement, yeah, but it also means you have the skill to survive a long time while still dealing damage.

[Lily from animove]

Pjwned, on 12 October 2014 - 02:57 PM, said:

High damage might mean poor shot placement, yeah, but it also means you have the skill to survive a long time while still dealing damage.

this I had 2 TW's one wiht 4xlrm 10 + some lasers and another with 4xlrm 10 + artemis and some lasors.

While the artemis one runs more hot due to lack of heat, and the non artemis one alway shad roughly the better damge, it performed worse, because damage was done more on the non vital parts. very high damage is often a sign that you couldn't hit the right parts needed to be efficient.

[Karl Streiger]

There is a thin line between dealing damage and being effective.

I love PUG examples. Here you got one:
Once i was toying with my Misery -2 Peps, 1 LL, 1 AC 20 - when i went down i had ~ 240-300dmg.... (at this time not a single assist) - so a bad game?

Another Misery - 5- Medium Pulse Laser, LBX and SRM - finally decided that it was time to attack. He did run in the same bunch of persons i have charged before (thinking he is with me)
Well he survived: 5 kills, >800dmg (almost every shot was a kill)

What happened? I did blew huge holes in the enemy armor - 40-80 dmg per location - and i did not concentrate on one target - i shot at all of them

His crit seaking ability - got multiple crits on the enemy (additional damage) - blew ammunition - (additional damage) - destroyed components (additional damage)...
So he would have been quite terrible if he had charged before me - (only armor damage -not very concentrated) - not sure that it would have been possible to kill those boogeys on my own - with the big guns.

---

So - in a TEAM GAME your individual performance is an illusion. If you got high numbers thank somebody that he made it possible for you (even the enemy because they were numb)

So you want a idea about your performance?
LRMs - > dealing much damage is a good performance
LBX, Clan Autocannons, Laser, SRMs -> dealing much damage could be mean:

• you are bad - spreading to much damage
  
• you are lucky - and the RANDOM Rolls decided that you caused some critical hits

PinPoint weapons - dealing much damage > 200 per kill you are bad

Edited by Karl Streiger, 13 October 2014 - 01:23 AM.

[Glythe]

That's a nice score but before your ego gets to big remember that skill is not damage. Damage in an indicator of how well you did but getting a kill with less than 240 damage is way better than dealing 400 damage and also one kill.

I had a game recently in my Atlas where I got 3 kills (not vultures) with about 240 damage. Without getting a headshot you basically need to recognize leg dumpers and XL engine users when you see their loadouts.

[TexAce]

Yesterday, Hunchy 4G, 4 Kills, 5 Assists with 242 damage 2 of those kills were nearly untouched mechs. Damage is not everything. Knowing where to place that AC20 slug matters much more.

[Roadbuster]

Less than 100 dmg with anything except a Locust, a pure TAG/NARC scout or an Assault which was left behind and got taken appart by a wolf pack, is nothing to be proud of.

Since I prefer pinpoint damage to take out mechs/components as fast as possible, in a good match I end up with ~400-700dmg depending if playing a medium, heavy or assault. Average is around 300-500 without any meta builds or clan mechs.

I also like LordKnightFandragon's approach though.

According to that, my Locust would be my most effective mech with 150+dmg compared to 32 armor

[LordKnightFandragon]

mike29tw, on 11 October 2014 - 08:33 PM, said:

You don't need to strip all of that 526 armor to kill it though. You'd be surprise how little damage is required to kill an assault mech with just a few well placed shots.

I don't think damage is an accurate indicator of skill. There are instances where I killed 2 fresh lights with 2 alpha each. That roughly amounts to sub-200 damage, but effectively take out 2 players from the other team. I wouldn't call that a below average performance.

There really is no end all be all, single indicator of player skill. In order to drop an Assault as quickly as you can, placing everything in the CT, thats what? Assuming the person put atleast some armor on the back, of lets say an Atlas, its like 114 for the front and like 62 for internals? So a total of 176 dmg to put him down? A Realistic amount of probably 220 or so, cuz you wont land everything in the CT....

[LordKnightFandragon]

Roadbuster, on 13 October 2014 - 03:41 AM, said:

Less than 100 dmg with anything except a Locust, a pure TAG/NARC scout or an Assault which was left behind and got taken appart by a wolf pack, is nothing to be proud of.

Since I prefer pinpoint damage to take out mechs/components as fast as possible, in a good match I end up with ~400-700dmg depending if playing a medium, heavy or assault. Average is around 300-500 without any meta builds or clan mechs.

I also like LordKnightFandragon's approach though.

According to that, my Locust would be my most effective mech with 150+dmg compared to 32 armor

Lol, I honestly dont know how to grade a properly played Light mech. I understand they are not damage dealers, I use my metric as its from World of Tanks and that is pretty much a game of subtraction vs your own team's health pool. First one to 0 loses, and I play heavy hitters whose main job is to deal the damage.

I suppose, if the Light was tagging his tail off and narcing for those missile boats, getting assists left and right, he would know he did his job as a light, if he helped his fellow Dire Whale NOT get circle jerked by the enemy lights, helped other mediums and lights gang bang the enemy Dire Whale.....its all the things NOT displayed in in game stats that make a good Light/medium player.

[Wicked12B]

Damange = skill end of story

[Vandul]

I can load up my cheeze 6xLBX5 Direwolf and generate all the damage in the world. Or I can throw lurms at every target I can get a lock on and land 800+ damage. Or run around in my Firestarter/Arrow spraying wildly and taking down damaged mechs landing 4+ kills/match.

But at the end of the show, If my name isnt in the top 12, I failed. Doesnt matter how many kills or how much damage or how many assists.

I.
Didnt.
Win.

/game

[Jman5]

I'm seeing a lot of low-damage apologists in this thread.

The only consistent exception where I feel it's safe to disregard damage numbers are for one-sided fights. When you're on a really good team that's ripping your opponents to shreds, the damage is generally much more equally spread and so it's lower overall. You see this on the good premade teams. Alternatively, when you're on the receiving end there simply isn't enough time for you to do respectable damage; you're quickly outnumbered and overrun. Tough break.

Other than that, you have to use the law of averages. That your average teammate in your average game has average accuracy. I disagree with all the arguments that these low-damage, multi-kill players are hotshot sharpshooters who only hit the vital component that they need to hit. No. These low-damage players are often missing and hitting irrelevant components just as much as anyone else (or even more so).

Mech selection and build does make it easier or harder to do good damage so I'm not expecting a spider to out-damage a direwolf. However, you can at least consistently be in the top 6 damage on your team in any chassis. With the one big exception I listed above it almost always comes down to...
1. poor survivability: dying before you can do decent damage
2. playing too passively: sitting behind cover too much and not finding opportunities to hurt the enemy

The first step to improving is to stop pretending that you're some sort of hyper-accurate crackshot and work on increasing your damage without dying.

[Y E O N N E]

I also operate on a per-tonnage mechanic.

4x: You pulled your own weight, but it's nothing commendable.

10x: This was a good, solid match.

15x: Excellent game. You punched above your weight.

20x: Phenomenal. Your team probably would have lost without your input.

25x: God-like.

All that said, I also distinguish between dishing out pain and actually being effective in combat. The amount of damage you do is meaningless if you haven't contributed to earning kills. I had a game last night on Canyon where I was in my LCT-1M and we lost. Somebody complained about all of us doing terribly. He had 461 damage for 0 kills and 0 assists. I had 372 damage for 1 kill and 0 assists on a Stalker I worked over. Whom do you think was the more valuable player?

It's actually better to get more kills for less damage because it means you are a good shot and that you can destroy the enemy more rapidly. This is good. This is very good. Draining the enemy of his resources so he can't fight, and doing it faster and better than him, is what combat is all about.

Edited by Yeonne Greene, 13 October 2014 - 11:08 AM.