106 replies to this topic

[Theodore42]

sneeking, on 13 June 2015 - 08:24 AM, said:

When damage kills locks spots and assists are all in balance then you have had a good game.

So you would judge a player's skill by the c-bills earned? If so, I would agree. That's probably the best metric for affecting the battlefield.

Although I've been known to "slice" through massive swaths of mechs just to rack up the assist points. That's worth a lot of cbills guys!!!

LordNothing, on 13 June 2015 - 04:24 AM, said:

sure you can shotgun damage all over an enemy and get your damage in, usually with lerms, srms, lbs, or poorly placed laser shots (slicers i call em) and get a lot more damage in than an individual who can aim.

Edited by Theodore42, 13 June 2015 - 08:38 AM.

[dezgra]

For myself, if i do 400+ damage in a medium or heavy (rare for me to pilot a light or assault) I feel I did my job. 200,000 c-bills I am smiles all around. Break 100 points...give me space, cause I'm excited...and flailing!

Edited by dezgra, 13 June 2015 - 08:49 AM.

[LordKnightFandragon]

Kotzi, on 12 June 2015 - 11:54 PM, said:

The least amount to kill a mech. Better put the sum of that for a match of course.

But sadly, the faster you kill a mech, the less you get paid....so, it pays to farm in this game. take a mechs arms off, its legs, and wittle it down to 1% before finally killing it.

[Ultimax]

Jman5, on 12 June 2015 - 02:39 PM, said:

Between games I started marking damage numbers of the assaults, heavies, mediums, and lights and split it between the losing team and winning team. This was done in the solo queue and all 3 game modes. I skipped over weekend event numbers to avoid any weirdness. Overall I recorded about 1,400 damage scores. Without further ado, here are the average* damage numbers.

Average Damage for a Winning Player by Weight Class

• Lights: 196

• Mediums: 312

• Heavies: 333

• Assaults: 383

There you have it. While I wasn't really surprised by the anemic light damage, I was surprised at how high the other three were. So next time you want to lash out at your team for being the ultimate noobs, consider whether or not you carried your own weight first.

This is a great conversation to start Jman.

Amount is relative though.

If average 30 ton mechs are doing an average of 6x their own weight in damage, while an average 65 ton heavy class is doing an average of 5x their damage - then the lights aren't doing that bad, relatively speaking.

(I went higher on ave for lights, just based on what I tend to see get played the most, where Heavies as a class have many options in the 65~70T range, the higher the weight of the Heavy average the better the light number would look anyway)


I also think the data is unfortunately, very very polluted. Solo PUG queue Lights as a class seem to be played by either

A: Really tiny group of actually good light pilots.
B: Larger group of people who just go "WEEEEEE!!!!" diving head first into the enemy team while they spin their enormous steering wheel.

Every time I see a Locust on my team at the start, I guesstimate there is an 80% chance they belong to group B.

Jman5, on 12 June 2015 - 02:39 PM, said:

Fun Fact: If you compare average winner and loser damage of each weight class, the medium class had the greatest gulf. Does this mean that your team's medium performance is the most important make or break weight class in the solo queue? Or is it just an anomaly that will eventually even out? I don't know.

That is definitely interesting.

Edited by Ultimatum X, 13 June 2015 - 09:32 AM.

[Felio]

Jman5, on 12 June 2015 - 02:39 PM, said:

Fun Fact: If you compare average winner and loser damage of each weight class, the medium class had the greatest gulf. Does this mean that your team's medium performance is the most important make or break weight class in the solo queue? Or is it just an anomaly that will eventually even out? I don't know.

Well done! And good call on using the median, obviously. That's really what people want when they say average, a lot of the time.

As far as this point about mediums, I would guess it's because they die first, and once a team is 2 kills ahead, the remaining mechs on that team are relatively safe.

Edited by Felio, 13 June 2015 - 09:31 AM.

[Moomtazz]

Jman5, on 12 June 2015 - 02:39 PM, said:

Fun Fact: If you compare average winner and loser damage of each weight class, the medium class had the greatest gulf. Does this mean that your team's medium performance is the most important make or break weight class in the solo queue? Or is it just an anomaly that will eventually even out? I don't know.

It could be that Mediums seem to be recommended most when new players ask about what mech to get. A new player probably scores less, in general, than an experienced player. So if it is true that Medium is the class with the largest percentage of new players, you might see a big swing like that. Low score = new player = loss, high score = experienced player = win

Please note I am not bashing newbies and I use "probably, might, etc" alot because I have no idea if it's true or not.

[Mcgral18]

Joseph Mallan, on 13 June 2015 - 02:27 AM, said:

The face of 1st world problems!

Well, I could have phrased it as video game disappointed, but don't you think that gets a little long winded?

[DaZur]

Interesting statistics...

Mine exhibit a clear correlation with the class I'm most comfortable with piloting:

Light: 100 - 200
Medium: 275 - 350
Heavy: 400 - 600
Assault: 300 - 400

As you can see, I consistently do better better in Heavies. Where one would expect a linear increase in damage from Light through Assault, my metrics show retrograde in the Assaults.

Actually... If it wasn't for my Zeus -6S the disparity between Heavies and Assaults would be even more prevalent (The Zeus pilots like a Heavies IMHO) so I actually do well in it.

[DAYLEET]

Jman5, on 12 June 2015 - 02:39 PM, said:

Over the years there has been a lot of back and forth about what sort of damage numbers people should aim for if they want to do well in Mechwarrior Online. 150? 200? 250? 300? More? With the influx of Clan XL engines and the many hitpoint quirks out there, we're likely at a high water mark of damage required to win a typical game. While most people would agree that a light should not be expected to do as much damage as an assault, but there aren't really any concrete numbers about what sort of damage an average Light or Assault on the winning team does.

Between games I started marking damage numbers of the assaults, heavies, mediums, and lights and split it between the losing team and winning team. This was done in the solo queue and all 3 game modes. I skipped over weekend event numbers to avoid any weirdness. Overall I recorded about 1,400 damage scores. Without further ado, here are the average* damage numbers.

Average Damage for a Winning Player by Weight Class

• Lights: 196

• Mediums: 312

• Heavies: 333

• Assaults: 383

There you have it. While I wasn't really surprised by the anemic light damage, I was surprised at how high the other three were. So next time you want to lash out at your team for being the ultimate noobs, consider whether or not you carried your own weight first.


If you're curious, the average damage for a losing player by weight class is

Light: 131.5
Medium: 170
Heavy: 232
Assault: 273

Fun Fact: If you compare average winner and loser damage of each weight class, the medium class had the greatest gulf. Does this mean that your team's medium performance is the most important make or break weight class in the solo queue? Or is it just an anomaly that will eventually even out? I don't know.

* Ok I said average, but technically I calculated median. It's basically the same thing except median is better if you want to cut through outliers on the high or low end.

Statistics... To put this into context, you have to mention that 400 damage using lasers roughly equals 100 damage using ballistic/ppc. Also disagree completely with the equation of your thinking that the median of a bunch of random people with random mech in a winning match is any indication of anything. Some of my light/meds have a much easier time pushing 500damage than some of my heavy/assault. Dealing damage is easy, dealing effective damage is not. The damage a big guy push and die with less than 100 damage but put it all in the same component of 2 enemy is much more important than the 500 damage another dude sprays over the body of 12 different mech.

Ultimatum X, on 13 June 2015 - 09:23 AM, said:

A: Really tiny group of actually good light pilots.
B: Larger group of people who just go "WEEEEEE!!!!" diving head first into the enemy team while they spin their enormous steering wheel.

Every time I see a Locust on my team at the start, I guesstimate there is an 80% chance they belong to group B.

I definitely belong in group B. I always go head first in the pack with most of my mech, especially the lights because it's so fun and so exciting. It certainly is visceral to pilot a light as opposed to drive a heavy. My stats arent bad however, 1V 240dps 1.12WR / PiratesBane 248dps 1.41WR, what do we learn from that? It's random, too many things can make or break a match at every second that it's going on. People should try to stop to justify themselves when they lose, you win some and then you lose some, it's fine.

Edited by DAYLEET, 13 June 2015 - 10:18 AM.

[thesleepyslam]

my usual metric is just 400+ damage for an average match, with 600+ for good match

[Akhri Mala]

Anything heavier than a light, I usually feel that 100 damage per minute is good.

[Light-Speed]

It should be noted that if a light and stay behind a heavier mech then the light can, say, kill 2 dires with ~200 dmg whereas a heavy or assault would need to do possibly twice that much damage going through the front armor to achieve the same results.
So damage isn't a very good indicator of how much the light contributed to the victory.
Plus the fact that a light doesn't have to do dmg to help the team win, like scouting, splitting, and/or distracting the enemy team.
Just my two cents.

[michaelius]

Ultimatum X, on 13 June 2015 - 09:23 AM, said:

This is a great conversation to start Jman.

Amount is relative though.

If average 30 ton mechs are doing an average of 6x their own weight in damage, while an average 65 ton heavy class is doing an average of 5x their damage - then the lights aren't doing that bad, relatively speaking.

That is definitely interesting.

I'd also say damage is also a function of cooling potential and light mechs will usually have as many heatsinks in engine as other classes while mostly carrying lower range more heat efficient weapons.

[Poggle]

Interesting stats - the lights may score very small but the question I have is how is the impact of force multipliers on overall score taken into account? Does it matter?

E.g like are the high numbers underpinned by support roles?

Edited by Poggle, 13 June 2015 - 11:52 AM.

[Moldur]

I consider over 300 damage pulling my weight, and over 400 damage as "good." Though I understand that's pretty hard to put out if you die too soon due to circumstance. Pinpoint damage builds can get a pass. The 12 Small Laser Nova can get a couple solo kills and still only put out 250ish damage.

Edited by Moldur, 13 June 2015 - 11:37 AM.

[Sarlic]

Quote

This was done in the solo queue and all 3 game modes

What a waste of a stats.

Everybody person has their own variable. Not many other factors have been taken in account.

Interesting that this is apparently a game for inviduals.

Edited by Sarlic, 13 June 2015 - 11:45 AM.

[Kristov Kerensky]

My damage per chassis is less then the averages for some of my Mechs, my PPFLD mainly, but I tend to average less per kill as well, so I'm good with that. Takes 33 to drop a Mech, if you do 400 and aren't using nothing but LRMs, you wasted a lot of time and energy to get that kill. And that's purely a personal skill thing, it may have zero influence on whether your team wins or loses.

I'm not happy with my Atlas DDC, it's at 393 average per kill, and that just ain't good with me, WAY too much wasted time there. My overall average per kill is 200, and that's something I'm not happy with, still too high, I need to cut that down, but I can live with it since I use missiles in one form or another a lot, ups the damages due to lack of precision, but good for the team overall.

And all that tells me is that I need to work on my aim, nothing else, but I already knew that.

We can all see our own averages, and they'll be different for all of us, team oriented players can show high damage to kill OR low damage to kill, they can show low damage a drop OR high damage a drop, it depends on what they were driving and what their function in the team was. I tend to try and open targets up and strip guns, leave the kills to others when I'm in my Assaults, while I go for quick kills in my Lights, and my damage/kill shows that difference in them. I don't see my assists, spots, flanks, formation, UAV, and TAG bonuses, the real helpful measures of team usefulness.

[YueFei]

Kristov Kerensky, on 13 June 2015 - 12:18 PM, said:

My damage per chassis is less then the averages for some of my Mechs, my PPFLD mainly, but I tend to average less per kill as well, so I'm good with that. Takes 33 to drop a Mech, if you do 400 and aren't using nothing but LRMs, you wasted a lot of time and energy to get that kill. And that's purely a personal skill thing, it may have zero influence on whether your team wins or loses.

I'm not happy with my Atlas DDC, it's at 393 average per kill, and that just ain't good with me, WAY too much wasted time there. My overall average per kill is 200, and that's something I'm not happy with, still too high, I need to cut that down, but I can live with it since I use missiles in one form or another a lot, ups the damages due to lack of precision, but good for the team overall.

And all that tells me is that I need to work on my aim, nothing else, but I already knew that.

We can all see our own averages, and they'll be different for all of us, team oriented players can show high damage to kill OR low damage to kill, they can show low damage a drop OR high damage a drop, it depends on what they were driving and what their function in the team was. I tend to try and open targets up and strip guns, leave the kills to others when I'm in my Assaults, while I go for quick kills in my Lights, and my damage/kill shows that difference in them. I don't see my assists, spots, flanks, formation, UAV, and TAG bonuses, the real helpful measures of team usefulness.

You gotta sometimes give credit to enemy pilots for their defensive piloting, too. It's not necessarily a lack of accuracy on your part that causes it to require you more damage per kill.

Look at it this way: pro NBA players sometimes miss free throws, and that's a shot they practice a lot and when they take that shot in a game they have a lot of time to line it up, and nobody in their face to contest the shot.

Now we're talking about you trying to hit an actively moving, thinking opponent, who is doing his hardest not to let you kill him so easily, and who is also shooting back, while his buddies may be swarming forward to give him some help?

Human reflex time is not 0 milliseconds, so it's not like you can instantly react to every change of direction he makes, or every little twist/turn/stutter step. Missing the vital component you wanted to hit might just be because the other guy is moving well, not because you couldn't aim.

[Bishop Steiner]

LordKnightFandragon, on 13 June 2015 - 09:00 AM, said:

But sadly, the faster you kill a mech, the less you get paid....so, it pays to farm in this game. take a mechs arms off, its legs, and wittle it down to 1% before finally killing it.

A lesson I have yet to learn. I fight to win, not farm cbills.

I probably cost myself a lot of epeen by being efficient.

[Water Bear]

YueFei, on 13 June 2015 - 12:54 PM, said:

You gotta sometimes give credit to enemy pilots for their defensive piloting, too. It's not necessarily a lack of accuracy on your part that causes it to require you more damage per kill.

Look at it this way: pro NBA players sometimes miss free throws, and that's a shot they practice a lot and when they take that shot in a game they have a lot of time to line it up, and nobody in their face to contest the shot.

Now we're talking about you trying to hit an actively moving, thinking opponent, who is doing his hardest not to let you kill him so easily, and who is also shooting back, while his buddies may be swarming forward to give him some help?

Human reflex time is not 0 milliseconds, so it's not like you can instantly react to every change of direction he makes, or every little twist/turn/stutter step. Missing the vital component you wanted to hit might just be because the other guy is moving well, not because you couldn't aim.

An often overlooked point. Many one-on-one battles, especially between assaults, boil down to who can twist better and take aim faster. One of the great strengths of Inner Sphere humanoid assaults is their typically large arms which can effectively shield their body. DPS assaults like the King Crab and the Dire Whale almost never take damage to their arms.

I for one am very disappointed when my Atlas dies with even one arm left attached, or my Zeus dies while it's left torso still exists. That means that my opponents have to be scoring at least 2-300 damage on each of those mechs, respectively, to score a kill. Probably much more to bring the Atlas down, since I do my best to Popsicle before I go down.