40 replies to this topic

[Lostdragon]

Bud Crue, on 26 September 2016 - 06:06 AM, said:

Yes, but Yellonet's OP is trying to get to a comparison based on actual data. Any quantitative comparison of performance should account for those things that you point to as potentially making a mech OP...at least as far as establishing an average performance baseline value(s). It is our inability to know the full scope of the data that makes up those baselines (because PGI doesn't share) which makes us have to fall back to our subjective views of what is OP and what is not. Alas, unless we know everything, we can only make assumptions based on incomplete data and our own subjective experiences.

Which brings us back to the features you list in your post.

To our subjective point of view a mech having all those features is going to be OP over ANY other mech lacking even one. Only real data provided by PGI can show us otherwise, and we just don't have it, not even with the leader boards.

I don't think you need quantitative data to make a good case a mech is OP or at least superior to other offerings. I predicted well before the KDK came out that it would obsolete the DWF because it has all the strengths of the DWF with none of the weaknesses and I was 100% right (and I definitely wasn't the only one saying this).

But we do also have some qualitative data we can look at, not as much as PGI, but the results of tournaments as well as personal stats can give you a good idea what is going on. In the Weight of the World tournament of the top 20 assault players 16 had the most contribution from the KDK with 3 DWFs and 1 KGC rounding out the leaderboard. Then there are personal stats, I ran the 4x CUAC10 on the DWF for a long time before I put it on the KDK3 but my stats in the DWF were never nearly as good as the KDK3, which had nearly twice the w/l and KDR.

I would love to be able to slice and dice PGI's data but I don't think it is necessary to do that to know which mechs are dominant and which are not doing well. I would also point to the rescale that screwed most lights as a case study in how using quantitative data without any regard for qualitative results can backfire.

[FupDup]

I think the "balanced range" of your diagram is a tad too broad. Mechs in the upper part of "balanced" would curbstomp the daylights out of mechs in the lower end of that same range.

[C E Dwyer]

Yellonet, on 26 September 2016 - 07:15 AM, said:

It's sad how many actually say things like that and mean it...

This is one of the reasons I'm looking forward to Battletech so much, zero need for balancing.

[Bobzilla]

Yellonet, on 26 September 2016 - 04:07 AM, said:

Clearly some people here have misunderstood the term "overpowered".
I'm going to clear things up for you.

Something being overpowered, or underpowered for that matter, has really nothing to do with the abilities and performance of the object itself, but rather the abilities and performance of the things it relates to and is used in conjunction with be it cooperatively or competitively.



If we take MWO for instance, when someone says that mech X is overpowered it means that the mech in question is in some regard performing better than most or all other mechs, on average.
The average bit is the important one, it's important to stress this as this seems to be what a bunch of people don't understand.
So when someone says that mech x is overpowered there is very seldom someone saying that the mech in question is bad or average, pretty much everyone agrees that the mech is quite good.
However there is usually much disagreement on whether or not it is overpowered, this IMO comes partly from some people misunderstanding what overpowered means, some will say:
- "No, it's not OP because I die in it as much as any other mech".
- "It's not OP because I see people doing less than 100 damage in them here and there."
- "It's not OP because I can do just as well in mech Y."
None of those arguments invalidates the claim that mech X is OP because overpowered doesn't mean automatic win or invincibility or that terribads suddenly become T1 superstars, no, it just means that on average mech X is performing better in some regard than most or all other mechs, enough so that it gives the player in such a mech an advantage over other players.
It might not be visible in all matches and in all situations, but (again) on average over several matches mech X comes out with better K/D or higher damage output or less deaths, or something else so that a pattern emerges that this mech is clearly better than other mechs.

Balanced mechs are the ones within the range where they can still be a little better or a little worse than other balanced mechs but in which player skill or tactical situation is always a larger difference than the abilities and performance of the mech.

Overpowered mechs are those that stick out as performing better than most mechs and as such gives an advantage to the player in many situations which in turn can in some degree make up for player skill.

Some big problems, one mech is always going to be better, and then in peoples minds OP.

Power creep. If the new mech is better than the old 'OP' mech, the old mech is automatically no longer OP due to the average. If something like this was set in stone a while ago, it wouldn't be a problem.

What stats do you base performance off of? That's a whole mess right there that you couldn't get even a small % of the community to agree on.

[Bud Crue]

Lostdragon, on 26 September 2016 - 07:29 AM, said:

I don't think you need quantitative data to make a good case a mech is OP or at least superior to other offerings. I predicted well before the KDK came out that it would obsolete the DWF because it has all the strengths of the DWF with none of the weaknesses and I was 100% right (and I definitely wasn't the only one saying this).

But we do also have some qualitative data we can look at, not as much as PGI, but the results of tournaments as well as personal stats can give you a good idea what is going on. In the Weight of the World tournament of the top 20 assault players 16 had the most contribution from the KDK with 3 DWFs and 1 KGC rounding out the leaderboard. Then there are personal stats, I ran the 4x CUAC10 on the DWF for a long time before I put it on the KDK3 but my stats in the DWF were never nearly as good as the KDK3, which had nearly twice the w/l and KDR.

I would love to be able to slice and dice PGI's data but I don't think it is necessary to do that to know which mechs are dominant and which are not doing well. I would also point to the rescale that screwed most lights as a case study in how using quantitative data without any regard for qualitative results can backfire.

Totally agree. My posts above are intended to be dripping with sarcasm but at the same time based on what I see as the only possible rationale for PGI's selective nerfing of mechs: The presumed actual use of collected data. Its either that or random chance (dart board of balance).

If they aren't making decisions based on actual comparison of data then what are they basing these decisions on?
To wit, go look at the last two nerf passes. Now, identify for me those mechs that were nerfed and which were truly OP based on leader board data or subjective comparison. The Oxide perhaps? What else? Were the Grasshoppers H, N and J truly so OP to warrant nerfs, but the P wasn't and still isn't? Did all of the Quickdraws save the IV-4 warrant TWO nerf passes? Was the Bounty Hunter really dominating play?

Yet here we are, with even fans of the Kodiak 3 stating that by any reasonable comparison it is the best...perhaps most OP...mech in the game, and PGI is silent and has remaind so since they "corrected" the UAC10 performance. Why? I can only assume it is because that the data shows that it is NOT OP, or they just haven't had a game of darts recently. Its one or the other.

[Almond Brown]

Cathy, on 26 September 2016 - 07:45 AM, said:

This is one of the reasons I'm looking forward to Battletech so much, zero need for balancing.

So you feel that ALL the BattleMechs^tm of BattleTech^tm are all "balanced"? Really? Wow! I hope you are not to disappointed when the games arrives and much "BattleMech/Weapons balancing" has been already done...

[PhoenixFire55]

Lostdragon, on 26 September 2016 - 05:48 AM, said:

When you compare a TBR to a SMN it is hard not to conclude the TBR is just better. The SMN has anemic hardpoints, lacks the best upgrades (ES), all its E hardpoints are on low slung arms, and comes with 5 tons of locked JJ. The TBR has a lot more hardpoints, good options for high mounted weapons, and more tonnage to work with due not just to being 5 tons heavier but to having ES and not having JJ locked on every variant.

Ok, lets talk about this particular example.

I build a 5xSRM6 brawler out of my Summoner. Hardpoints suit my build perfectly and I can't care less about low hardpoints in arms because I'm a brawler and I don't take g*y pop shots over ridges. I have plenty tonnage for all my weapons, ammo and lots of space for DHS. Apart from that I have a much better side profile and lots of JJs to tank and spread damage like a boss. I.e. from my point of view Summoner is just better than Timber.

See my point yet?

[Davers]

Bobzilla, on 26 September 2016 - 07:47 AM, said:

Some big problems, one mech is always going to be better, and then in peoples minds OP.

Power creep. If the new mech is better than the old 'OP' mech, the old mech is automatically no longer OP due to the average. If something like this was set in stone a while ago, it wouldn't be a problem.

What stats do you base performance off of? That's a whole mess right there that you couldn't get even a small % of the community to agree on.

Well, it would be based on the only stats MWO cares about- damage, kills, and w/l. What other stats matter?

[PhoenixFire55]

Jetfire, on 26 September 2016 - 06:29 AM, said:

I think it boils down to meta builds possible on some mechs being OP, not the mechs themselves. There have definitely been Meta's which left most mechs builds in the dust and only a few chassis' could compete.

"Meta" is dictated by the way players choose to play the game, i.e. hide and take occasional alphas. The fact that people are choosing to play only this kind of game and to whine about how certain mechs are better suited for it instead of adapting and playing the game differently when they drive other mechs have nothing to do with mechs themselves.

I'll give you an example with an all-powerful OP KDK-3. Put it on Polar against a Raven or ACH with 2xERLLs and see what happens. Or take an Atlas-S with 4xSRM6s + AC20 and make them engage in a basement on HPG. Suddenly KDK-3 sux. But instead of playing the strenghts of their mechs people put LRMs and LLs on their Atlases and then complain how KDK-3 is OP.

[PhoenixFire55]

Mister Blastman, on 26 September 2016 - 06:51 AM, said:

They can be brought down just like any other 'mech. That's not the problem, though--it is how they literally outclass any other robit I own with hardly any seat time at all. Would I say they spread damage well? No, not really. But they sure as hell can pump it out like nothing I've ever used in this game before (beyond the Dragon I hacked in closed beta to have an AC/5 that fired three times a second... 15 dps. Muhahahaha. I reported a detailed explanation of how to PGI and now we can't do it yay! ).

Sorry, but quad-UAC10 Dire was a thing for years before KDK showed up and it pumped the same damage and still does. Problem obviously is somewhere else, but its not mechs in question, its people who play the game with stompy robots as if it was a stealth simulator. But when there is nowhere to hide (like Polar etc.) suddenly even LRMs become viable.

[Ultimax]

Yellonet, on 26 September 2016 - 04:07 AM, said:

Clearly some people here have misunderstood the term "overpowered".
I'm going to clear things up for you.

Something being overpowered, or underpowered for that matter, has really nothing to do with the abilities and performance of the object itself, but rather the abilities and performance of the things it relates to and is used in conjunction with be it cooperatively or competitively.



If we take MWO for instance, when someone says that mech X is overpowered it means that the mech in question is in some regard performing better than most or all other mechs, on average.
The average bit is the important one, it's important to stress this as this seems to be what a bunch of people don't understand.
So when someone says that mech x is overpowered there is very seldom someone saying that the mech in question is bad or average, pretty much everyone agrees that the mech is quite good.
However there is usually much disagreement on whether or not it is overpowered, this IMO comes partly from some people misunderstanding what overpowered means, some will say:
- "No, it's not OP because I die in it as much as any other mech".
- "It's not OP because I see people doing less than 100 damage in them here and there."
- "It's not OP because I can do just as well in mech Y."
None of those arguments invalidates the claim that mech X is OP because overpowered doesn't mean automatic win or invincibility or that terribads suddenly become T1 superstars, no, it just means that on average mech X is performing better in some regard than most or all other mechs, enough so that it gives the player in such a mech an advantage over other players.
It might not be visible in all matches and in all situations, but (again) on average over several matches mech X comes out with better K/D or higher damage output or less deaths, or something else so that a pattern emerges that this mech is clearly better than other mechs.

Balanced mechs are the ones within the range where they can still be a little better or a little worse than other balanced mechs but in which player skill or tactical situation is always a larger difference than the abilities and performance of the mech.

Overpowered mechs are those that stick out as performing better than most mechs and as such gives an advantage to the player in many situations which in turn can in some degree make up for player skill.

You're not wrong, the problem is that we do not have a definition of what is the ideal power level for any given mech in any given weight class to compare all of the other mechs to, nor do we have those mechs to cross-reference against other weight classes.


So when many of us say a mech is fine or that it's OP or not OP, it's more likely based on our ideal place of balance and not compared to a heap of mechs that suffer from clear design issues.

[SmithMPBT]

Cathy, on 26 September 2016 - 06:23 AM, said:

Well quite clearly the Kodiak isn't Op because I've not had a 1k game in mine, quite a few 900's, or got Ace of spades.

While the Giant GrassHopper clearly is OP, because i've done both in one.

This is all the data you need

An awesome pilot can pull 900 in a bunch o mechs, the true measure is the casual player doing way way more damage than they would in a comparable mech of that class. For instance, dude does about 300 in most clan assaults, then averages 800-900 in a Kodiak-3.

[Lostdragon]

PhoenixFire55, on 26 September 2016 - 08:25 AM, said:

Ok, lets talk about this particular example.

I build a 5xSRM6 brawler out of my Summoner. Hardpoints suit my build perfectly and I can't care less about low hardpoints in arms because I'm a brawler and I don't take g*y pop shots over ridges. I have plenty tonnage for all my weapons, ammo and lots of space for DHS. Apart from that I have a much better side profile and lots of JJs to tank and spread damage like a boss. I.e. from my point of view Summoner is just better than Timber.

See my point yet?

Being able to do one thing well doesn't mean the mech is fine or that it is as good as a TBR. I have a SMN build I love that I can do great things with too. Overall, though, the TBR has more flexibility and many more viable builds. The TBR can do SRMs with JJ and backup weapons and it has high mounts so while you may prefer your SMN build I don't think you can successfully argue it is objectively better than the TBR for SRMs anyway.

[Novakaine]

So my six large laser or lrm 100 Stalkers are not OP?
Get real.
Can't speak to stinky clanker mechs.
I have my diginity ya know.

[PhoenixFire55]

Lostdragon, on 26 September 2016 - 09:05 AM, said:

Being able to do one thing well doesn't mean the mech is fine or that it is as good as a TBR. I have a SMN build I love that I can do great things with too. Overall, though, the TBR has more flexibility and many more viable builds.

This is nothing more than your personal opinion. Yeah, it can do laservomit poking better than Summoner. If that is all you care about when you talk about flexibility ... Plus, "flexibility", i.e. the ability to use more loadouts means nothing in terms of being OP or not. KDK-3 can do a gazillion of builds with different ballistics but somehow only select few work for most people.

[AlphaToaster]

PhoenixFire55, on 26 September 2016 - 08:40 AM, said:

Sorry, but quad-UAC10 Dire was a thing for years before KDK showed up and it pumped the same damage and still does. Problem obviously is somewhere else, but its not mechs in question, its people who play the game with stompy robots as if it was a stealth simulator. But when there is nowhere to hide (like Polar etc.) suddenly even LRMs become viable.

I think in regards to comparing the Dire to KDK, The KDK has all the strengths of the Dire with none of the weaknesses.

Dire is slow/lumbering with a big nose so it's difficult to hide the CT. The twist is slower to make that nose even harder to hide.

KDK, it's a more humanoid shape with better ability to shield CT when twisting. It's taller so it gets the top peek advantage, though the dire can side peek, it's speed still hurts it when poking.

Dire is what? 56kph, KDK is 70kph? This is quite substantial when engine speed translates to torso twist speed.

I am not a supporter of making all mechs equal. That would be boring and defeat the purpose of variety. I think we would see stronger balance if we ranked mechs similar to how War Thunder has their rankings. Some mechs would be rank 1 some would be rank 5.

But honestly, I'm OK with the KDK being the best mech in the game right now. Why? Because it's 100 tons. I would expect the best mech in the game to also be one of the heaviest. So I feel we're on the right track here. I would be more perplexed if the top mech remained a 75 tonner after all these assaults were released.

I wouldn't mind if ammo was more dangerous to have on a mech but I am against weapon nerfs to try and bring a specific mech down a few notches.

Edit - I also wanted to add in general, I think the problem is perceived to be worse when there are floodgates that get opened so players don't have to wait a long time in queue. There is only supposed to be 3 Assaults per team in an ideally balanced match. If players keep refusing to queue in anything other than Heavy or Assault because they know there's a gate that makes 5 assault per team games, then what's the point? I say had limits!! 3 Lights/Med/Heavy/Assaults per game hard cap. If players don't like a 20minute wait every game get in a medium or light.

Matchmaker is designed to balance assault pilots out of the queue by making the wait time longer. Players ignore that they are bringing this on themselves by flooding the queue with heavies and assaults.

Edited by AlphaToaster, 26 September 2016 - 10:37 AM.

[Lostdragon]

PhoenixFire55, on 26 September 2016 - 10:07 AM, said:

This is nothing more than your personal opinion. Yeah, it can do laservomit poking better than Summoner. If that is all you care about when you talk about flexibility ... Plus, "flexibility", i.e. the ability to use more loadouts means nothing in terms of being OP or not. KDK-3 can do a gazillion of builds with different ballistics but somehow only select few work for most people.

That's not an opinion, the TBR can run more builds that work because it has more hardpoints and tonnage. It can do builds the SMN just can't do like 2x ERPPC + GR, which is one of the most powerful heavy builds out there.

[Moonlight Grimoire]

What is kind of funny is that we have, finally, the last mech weight class leader board event going on which then will give us hard data for each weight class and with that allow ranking of mechs by any form of average we wish to apply. Through this we can see what mechs out outliers of a weight class by filtering results by weight classes, we can see trend lines and what not, filter by tech bases, omni or non omni, all that lovely stuff.

As much bad game play comes out of events in general we still now will have hard data in two days for everything. This I think is part of PGI wanting to get data, though, obviously they already have this internally, or should. However, this gives us said data as well which maybe that is the intent, probably not, but, it is interesting to go over the data. Seems the Hellbringer is the best clan heavy mech, the Archer predictably (and sadly) is the worst heavy mech in the game (it needs a redo on the geometry, that low cockpit is not doing it any favors). Night Gyr is a strong clan mech which is pretty nice not to have it as another OP mech in the game, and they do exist, as the op said there are mechs that fall outside the bounds of what is considered average strength plus/minus a bit of standard deviation, we all know there are plenty of under powered mechs as well, so, yeah. Maybe this will get PGI to go and help the hurting mechs and tweak the very strong mechs that are just outside the envelope of what is considered balanced.

[Mcgral18]

Mister Blastman, on 26 September 2016 - 06:47 AM, said:

KDK-3 epitomizes the term. Click to win, yo.

Never have I had so many 1k games in a robit before, without breaking a sweat.

And that's post nerf
The SupaHunchDakkaBear was nothing short of God tier
Similar agility to a mech half it's size, with twice the guns, twice the ammo, and half again the heatsinks (in PoorDubs)

It had, without exaggerating, 1000 average damage after over 100 matches.



It is no longer the SupaHunch, but the DakkaBear lives in a much less potent version, albeit cooler.

Personally, I've moved into the PokeBear
All the damage of a WubShee, at 4x+ the effective range. I'll be coming up to my 300th headshot soon

FupDup, on 26 September 2016 - 07:41 AM, said:

I think the "balanced range" of your diagram is a tad too broad. Mechs in the upper part of "balanced" would curbstomp the daylights out of mechs in the lower end of that same range.

Ideally, that would be a slowly decreasing gap every patch cycle, with some iterative changes here and there on the outliers


#PGIPLZ

PhoenixFire55, on 26 September 2016 - 08:40 AM, said:

Sorry, but quad-UAC10 Dire was a thing for years before KDK showed up and it pumped the same damage and still does. Problem obviously is somewhere else, but its not mechs in question, its people who play the game with stompy robots as if it was a stealth simulator. But when there is nowhere to hide (like Polar etc.) suddenly even LRMs become viable.

You know, only with worse hitboxes, half the agility, less ammo, significantly worse weapon mounts.
DakkaBear outdoes the Whale in every way, when it comes to a quartet of Dakka (or Goose)

[Yellonet]

FupDup, on 26 September 2016 - 07:41 AM, said:

I think the "balanced range" of your diagram is a tad too broad. Mechs in the upper part of "balanced" would curbstomp the daylights out of mechs in the lower end of that same range.

It doesn't have a scale, besides it was just a quick and dirty example to illustrate my point.

Cathy, on 26 September 2016 - 07:45 AM, said:

This is one of the reasons I'm looking forward to Battletech so much, zero need for balancing.

Until the multiplayer expansion?