109 replies to this topic

[Wintersdark]

Mcgral18, on 31 March 2015 - 03:50 PM, said:

At first glance I can't help but feel those numbers are very small....but then saw, PUG LIFE.

Perhaps you should use yourself as a benchmark?


Fair assessment though, Big robots get big damage, which is certainly nicer than smaller damage numbers.

Not so small, really.

There's lots of decent numbers going on, you've got to remember that you're looking at a cross section where there's roflstomps going on and whole teams score practically nothing drawing down the average.

When you ask someone what their average damage done is, you'll always get a number that (assuming they're trying to answer honestly) will be a rough average of what they feel is normal games, not affected by those horrible matches where they got killed right at the start and did nothing of note... and everyone has those matches.

[Quxudica]

Jman5, on 31 March 2015 - 02:38 PM, said:

This probably wont come as a surprise to many, but I thought it would be nice to illustrate why the light queue tends to be so low, and why the heavy queue is always so high. Damage number are abnormally high for Heavy mechs, and dismally low for light mechs. It's not exactly an Earth shattering revelation, but people like doing damage and the heavy is the damage king by a mile.

Here, I broke down a bunch of games for you guys. I'll admit this isn't a huge sample size, but it's about 300 points of data in the solo queue.



167 average damage is terrible. Is it really any surprise no one wants to play lights? Also look at all those people who were in the 1-99 damage section. It's the biggest chunk!




At least you're usually breaking 200 damage. Still not that impressive. Again no surprise that mediums are the second least played. Also of note, the biggest chunk is in the 100-199 section. Not good.




TO THE MOON!!! So we went from 231 to 375. That's a huge spike. And the biggest piece of the pie is 500+ damage!




And here we come crashing back to Earth. Assault damage numbers are actually what you would expect to see from heavies, which is a little weird. However it does seem pretty nicely split between each damage section.




As you can see average damage numbers go way out of wack when we get to heavies.


Weirdly though when you flip Assault and Heavy around you get a pretty nice line.


So what the heck is going on with Heavies? Are they the perfect storm of mobility, survivability, and firepower? Or is this simply a matter of good players gravitating toward heavies? But then shouldn't Elo correct for skill?

A couple simple explanations and they feed into eachother:

First, Pinpoint Alpha Strike damage is so high it severely compromises the survivability of Lights and Mediums and just as severely punishes any kind of mistake made piloting a slow assault. (this is also why some mechs in those classes are so superior, part of why the Storm Crow is so good is simply because it's hitboxes let it deal with massive damage better than other mechs its size).

Second, because of this fact the Heavy class is the easiest class to pilot. Most heavies have the right balance of firepower, armor and speed/maneuverability to deal with the massive alpha damage flying everywhere.

Lastly, because of how quickly mechs can dish out crippling damage it creates a self perpetuating cycle: Firepower becomes the single most important aspect of a mechs build, as it needs to keep up with everyone elses massive damage numbers to have any real chance at being competitive. Even chassis that were intended to be support vehicles are instead being loaded for bear because that's the name of the game This is also exacerbated by the fact most support equipment can be taken by almost any mech in the game with very little trade off and support equipment simply plays such a minor role in this game.

If PGI would finally commit to rebuilding the heat system (reduce the cap considerably, increase dissipation, the result being fewer weapons fired at the same time but smaller groups of weapons being fired more often) and breaking 100% convergence, you would quickly see a leveling out of class representation. If PGI would actually put some effort into fleshing out the concepts of support mechs, reduce the games overall emphasis on dealing damage as fast as possible, you'd see even more improvement.

Edited by Quxudica, 31 March 2015 - 04:00 PM.

[XX Sulla XX]

Combination of speed, ability, hit boxes, firepower. At the same time lack of good combinations of those in mechs like the Victorm and Highlander etc.

1. Fix 80 ton mechs.

2. Fix other assaults like Highlander.

3. Give medium more mobility and a cbill and xp bonus.

4. Give lights a cbill and xp bonus.

Edited by XX Sulla XX, 31 March 2015 - 04:01 PM.

[Jman5]

Wintersdark, on 31 March 2015 - 03:47 PM, said:

Damage done is only a small part of earnings. What pushes earnings up is assists, kills, component destruction, and very significantly things like Savior Kills, Solo Kills, etc.

People (incorrectly) correlate high damage done with high earnings, as high damage done often goes hand in hand with the above bonuses. But the important part is if you understand how the earnings work, you can make a lot more money with less damage done.

When I say damage is directly and indirectly tied to earnings, I don't just mean damage = money. At the moment you get 21 cbills per point of damage, which adds up, but is only one part of it.

However people with high damage will tend to have many more of the other stats that are related to doing damage. Assists and kills bonuses go up the more people you hit. People with low damage usually have lower assists/kills, while people with high damage usually have more. Kill-most-damage and solo kill bonuses are again skewed toward high damage players for obvious reasons. Savior kill bonuses rely on you actively shooting at mechs before they die. again indirectly tied to damage. Component destructions, same story.

So while, yes you can make lots of money with modest damage, I guarantee you'll typically make more in your higher damage games because you're creating more opportunity to earn those big-ticket indirect rewards.

The same story for experience.

Mcgral18, on 31 March 2015 - 03:50 PM, said:

At first glance I can't help but feel those numbers are very small....but then saw, PUG LIFE.

Perhaps you should use yourself as a benchmark?


Fair assessment though, Big robots get big damage, which is certainly nicer than smaller damage numbers.

Unfortunately I would make a pretty bad test subject because I have very few heavy games.

Edited by Jman5, 31 March 2015 - 04:11 PM.

[Rushmoar]

InspectorG, on 31 March 2015 - 03:40 PM, said:

Slow the turning rate on mediums(slightly), Heavies( a decent degree), assaults(greatest degree)

Basically this, it's risk vs reward. Most people say it is not worth taking a light and risk getting legged in 1 alpha strike. Fire power is a big reason to take a heavy or an assault but survivability is the biggest reason.

Why was 3/3/3/3 introduced? It's because you could see a match with 10 assaults and 2 heavies on the same team. There are no reasons for mediums and lights to be in the game. Lights have evey thing stacked against them except for speed and size. But assaults only real disadvantage is ground speed. They have enough agility right now to shot most mechs so where is the draw back for the assaults? Besides, if a light gets on one of your assaults, it's the teams responsibility to get rid of the light.

So if it is 3 light vs. 3 assaults, i'll bet on the assaults every time.

[Averen]

There are a lot of good reasons in this thread, the versatility of heavies is certainly a factor. Mediums (nvm stormcrow) are still limited and risky mechs. Most IS mechs use XLs, cent's/hunchies are challenging, and novas easily disarmed. Lights take a lot of situational awareness (or the magical firestarter), and assaults can be punishing despite their strength. Often your team is also just plain stupid and leaves them to die.

It would be very interesting to see which mechs did so much damage, tho. I'd bet my money on timberwolfs and to a lesser degree hellbringer. Feels like TBRs are absolutely everywhere these days, and why wouldn't they.

I think I'll actually note down the 500+ dmg heavies i see. Might as well put my assumptions to the test and add some actual data.

Edited by Averen, 31 March 2015 - 05:10 PM.

[WarZ]

Solid thread.

Also I agree with damage being a large indicator of performance. As Jman stated above more damage is more opportunities to earn cbills and xp, as well as more opportunities to influence the outcome of a match.

Sure there are other factors that can be counted. But damage is the most consistent and telling indicator across most players.

[Darlith]

Lights also have a huge spread of very low damage because I have noticed often in the Solo queue you have a couple lights per match that always seem to charge off to death immeditately as if they are the god of death incarnate and then scream they were unsupported leaving them at under 100 damage. Then you get that one guy who sniped the whole round getting 400+ damage.

Likewise you get some low damage assault skewing the numbers because they charge in calling the team pansies and die to massed fire. Or they expected to be supported when someone called pushed and no one else moved.

Heavies and Mediums I think get a bit less of that because they tend to stick together more. But the mediums get skewed lower because they are just that much easier as a whole to alpha to death before putting up high performance numbers.

[OznerpaG]

all i know is i'm excited for the urbie so the the heavy queue can clear out a bit - it's been over a month since i'v been able to pilot a heavy. i'd like to play with the Summoner to see how it runs with the new quirks, but it's a heavy and if i have to wait several minutes for a game i'm taking a REAL heavy out

Edited by JagdFlanker, 31 March 2015 - 05:08 PM.

[Y E O N N E]

Lord0fHats, on 31 March 2015 - 03:38 PM, said:

Something that would really help several mechs is for Engine Tonnage to be tied to Mech weight rather than mech class. This would be a god send; Locusts. Commandos. Cicadas. Ice Ferrets. Dragons. Quick Draws. Victors. Gargoyles. Just off the top of my head.

What do you mean by this? You mean engine size? Because a given engine size is the same weight regardless of class. Engine size, however, is determined by what the 'Mech comes equipped with. A 'Mech with a small engine to begin with will have a lower cap. Hence, slow-ass Stalkers and Blackjacks not getting 250-size engines.

To be fair, though, a fast Stalker is a bit nightmarish.

Or were you saying that engine weight should be scalar, i.e. a STD 160 on a Locust weighs less than a STD 160 in an Atlas?Because that would be weird.

[Jman5]

Darlith, on 31 March 2015 - 04:48 PM, said:

Lights also have a huge spread of very low damage because I have noticed often in the Solo queue you have a couple lights per match that always seem to charge off to death immeditately as if they are the god of death incarnate and then scream they were unsupported leaving them at under 100 damage. Then you get that one guy who sniped the whole round getting 400+ damage.

Likewise you get some low damage assault skewing the numbers because they charge in calling the team pansies and die to massed fire. Or they expected to be supported when someone called pushed and no one else moved.

Heavies and Mediums I think get a bit less of that because they tend to stick together more. But the mediums get skewed lower because they are just that much easier as a whole to alpha to death before putting up high performance numbers.

This was one of the questions I set out to answer and it was why I chose to use the Pie Chart. I wanted to know if there were simply lots of extremes on both the high and low end of the damage spectrum. As you can see on the Light's pie chart this was not the case. There were plenty of lights that did terribly as you said, but there were almost no lights on the other end of the spectrum. Only 7% managed to hit even the average damage that Heavies were getting.

Funny enough, it seems that the Assault Class is the one with the most even damage distribution. I have no idea why.

Edited by Jman5, 31 March 2015 - 05:32 PM.

[JaxRiot]

Dont know if I have room to talk here, but just speaking from my limited experience..

I believe Lights are supposed to be more support roles. But playing support in this game is all but useless in pugs. So people try to get as much dps out of them as they can and at least zip around tagging multiple targets. They dont do a lot of damage but can make c-bills from the kill assists. The IS Light mechs are best for that because their engines can be swapped out for blinding speeds.

Medium Mechs are the red headed step children of this game imo. They can have decent fire power but mostly dont have the speed to get out of a tight situation or the durability to last very long. Which means they are generally picked off after the first big skirmish of the map unless they are very careful, and cant get the kill assists or any higher numbers. Medium Mechs are not good for c-bill earnings imo. Decent damage, but low survivability.

Heavies are where its at imo. They have the armor to last longer than one encounter and can pack on the fire power to get some decent damage scores.

Assaults only get the scores that they do because they are durable enough to last, and can dish out some damage before they go down. Other than that I dont see anything special about Assaults damage capabilities.

Just my 2 cents

Cheers

Jax

Edited by JaxRiot, 31 March 2015 - 05:46 PM.

[Jman5]

I just thought of another hypothesis. Perhaps the community has simply figured out Heavies to a large extent but there are flaws in how many players tends to build and play the other three classes.

Heavies tend to get the most scrutiny by the competitive scene and that knowledge does eventually permeate into the non-competitive playerbase. It's possible this simply indicates a flaw in conventional wisdom when it comes to building and playing the other weight classes in the solo queue.

Food for thought.

[Ghogiel]

Myaverage FS9 dmg 366dmg

And I am very bad at lights

[Jman5]

Ghogiel, on 31 March 2015 - 06:19 PM, said:

Myaverage FS9 dmg 366dmg

And I am very bad at lights

I think you're underestimating yourself. I just buzzed through the last 10 lights I played with and against and their damage.

469
187
55
30
231
106
142
166
78
18

[Ghogiel]

Jman5, on 31 March 2015 - 06:25 PM, said:

I think you're underestimating yourself. I just buzzed through the last 10 lights I played with and against and their damage.

469
187
55
30
231
106
142
166
78
18

No I am bad. Just because most people are worse doesn't make me good.

[Jman5]

Ghogiel, on 31 March 2015 - 06:32 PM, said:

No I am bad. Just because most people are worse doesn't make me good.

Actually it does. Everything is relative!

Edited by Jman5, 31 March 2015 - 06:33 PM.

[Darlith]

Jman5, on 31 March 2015 - 05:30 PM, said:

This was one of the questions I set out to answer and it was why I chose to use the Pie Chart. I wanted to know if there were simply lots of extremes on both the high and low end of the damage spectrum. As you can see on the Light's pie chart this was not the case. There were plenty of lights that did terribly as you said, but there were almost no lights on the other end of the spectrum. Only 7% managed to hit even the average damage that Heavies were getting.

Funny enough, it seems that the Assault Class is the one with the most even damage distribution. I have no idea why.

True, but this is also colored by the sorts of people you play beside regularly. I see a more broad spread of damage normally across all groups I'd think. I must start keeping track though, curious what numbers I'd come up with. Not sure I want to keep track of 300 matches of data though.

[Y E O N N E]

Jman5, on 31 March 2015 - 06:01 PM, said:

I just thought of another hypothesis. Perhaps the community has simply figured out Heavies to a large extent but there are flaws in how many players tends to build and play the other three classes.

Heavies tend to get the most scrutiny by the competitive scene and that knowledge does eventually permeate into the non-competitive playerbase. It's possible this simply indicates a flaw in conventional wisdom when it comes to building and playing the other weight classes in the solo queue.

Food for thought.

I can tell you that I don't master weight classes, I master 'Mechs. If you plunk me down in a Mist Lynx, I will derp my way to an early grave because I haven't figured it out yet. When I want to get serious, I will drop in a Locust because I consistently do nasty, nasty things with it, regardless of variant.

So, some 'Mechs have steeper learning curves because of some combination of bad geometry, bad engine caps, and bad hard-points. These effects are more pronounced at the extreme ends of the weight spectrum. Thus, players tend to stick to the middle two classes, with a preference for heavies since heavies tend to be barely any slower than Mediums but sport better equipment potential.

I think it was Adiuvo who once mentioned it, but Heavies are too agile. They really are.

[Mcgral18]

Jman5, on 31 March 2015 - 06:25 PM, said:

I think you're underestimating yourself. I just buzzed through the last 10 lights I played with and against and their damage.

469
187
55
30
231
106
142
166
78
18

Mind taking a look at your own stats?

Damage divided by matches played....with some borked stats every now and then.

Cute Fox Prime is at 362 damage with 176 drops

Myth Lynx at 368 damage after 87 drops

Ember at 347 damage after 77 drops

FS9-S at 439 damage after 72 drops


Haven't played many lights since the stat reset, those were closer to the 270 mark, with ~500 drops in the Jenners (I've gotten better! Thank you Seismic Wallhack)