Jump to content

The Number Is In, And It's 90%


692 replies to this topic

#601 Mcgral18

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • CS 2019 Top 8 Qualifier
  • CS 2019 Top 8 Qualifier
  • 17,987 posts
  • LocationSnow

Posted 09 August 2014 - 11:06 AM

View PostPhaeric Cyrh, on 09 August 2014 - 11:01 AM, said:


Really? I have heard clan players say they are so light and take so few slots, that they throw them on almost as an afterthought, and that at close qrtrs the screenshake if nothing else is worth it.

Regardless, they are better than the IS equivalent. The stream style of launch is only affected by those with AMS (most players do not use AMS) and at their weight and slot reduction that is easily overcome by simply equipping more tubes. That is aside from the fact that the stream launching style cause WAY more cockpit/screen shake than the IS version.. When you are trying to brawl, and a clanner is constantly pelting you with LRM 5's it can be damn near impossible to get your shots off.


The stream against AMS is a bigger factor than most people give credit to.

My cLRM25 or 30 (can't quite remember) couldn't get any missiles past a dual AMS Summoner, and he was the only one left, so it was only his pair of AMS. From 500M to 150M, nothing got past.


Though, I don't have numbers for 20 isLRMs VS 40 cLRMs.

Edited by Mcgral18, 09 August 2014 - 11:08 AM.


#602 IraqiWalker

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 9,682 posts
  • LocationCalifornia

Posted 09 August 2014 - 11:06 AM

View PostPhaeric Cyrh, on 09 August 2014 - 11:01 AM, said:


Really? I have heard clan players say they are so light and take so few slots, that they throw them on almost as an afterthought, and that at close qrtrs the screenshake if nothing else is worth it.

Regardless, they are better than the IS equivalent. The stream style of launch is only affected by those with AMS (most players do not use AMS) and at their weight and slot reduction that is easily overcome by simply equipping more tubes. That is aside from the fact that the stream launching style cause WAY more cockpit/screen shake than the IS version.. When you are trying to brawl, and a clanner is constantly pelting you with LRM 5's it can be damn near impossible to get your shots off.

View PostKin3ticX, on 09 August 2014 - 11:01 AM, said:


nevertheless, you can chime in with indirect fire support without hurting your direct fire capability compared to what you can do with IS-tech.

I can can understand these two points, however:

Saying that clan LRMs are only good against opponents that don't mount AMS is like saying Flamers are good against opponents that don't put armor on.

The weapon shouldn't be considered OP because the playerbase refuses to counter it. That is flawed reasoning.

In close ranges (under 180 meters), they are better.

#603 Sephlock

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,819 posts

Posted 09 August 2014 - 11:10 AM

View PostMcgral18, on 09 August 2014 - 11:06 AM, said:


The stream against AMS is a bigger factor than most people give credit to.

My LRM25 or 30 (can't quite remember) couldn't get any missiles past a dual AMS Summoner, and he was the only one left, so it was only his pair of AMS. From 500M to 150M, nothing got past.


That settles it, there needs to be a tiered set of missile interception challenges , with the final one being entitled "You shall not pass!"





#604 COOL HANDS

    Member

  • PipPipPipPipPip
  • Bad Company
  • 158 posts
  • LocationMilwaukee Wisconsin

Posted 09 August 2014 - 11:16 AM

You know when I read comments that say " I wish those numbers from the test never came out ". It lets me know that the guys who own the clan mechs were never really totally upfront about the advantages the clans had over the is mechs. Especially when people were QQing about them. I never started threads about it myself but I did have my suspicions about issues with them. There's really no point in getting worked up about it. The numbers are out there, it is what it is. :huh:

#605 Phaeric Cyrh

    Member

  • PipPipPipPipPip
  • Mercenary
  • Mercenary
  • 123 posts

Posted 09 August 2014 - 11:16 AM

View PostIraqiWalker, on 09 August 2014 - 11:06 AM, said:

I can can understand these two points, however:

Saying that clan LRMs are only good against opponents that don't mount AMS is like saying Flamers are good against opponents that don't put armor on.

The weapon shouldn't be considered OP because the playerbase refuses to counter it. That is flawed reasoning.

In close ranges (under 180 meters), they are better.


I don't think anyone said that. What I did say is that the stream style launch is only affected by AMS... That is not at all the same thing.

The reason I don't equip AMS on most of my mechs is that I mostly drive lights and mediums, and in that class of mech I can generally evade missles and locks before I take any damage, especially if I have radar derp equipped. But this is equal for all types of LRM's and not specific to clans tech.

In general we can pretty much all agree that LRM's are not terribly great weapons for either side, and are probably right where they should be at as a tactical support role.

That in mind, I would rather have clan LRM's because one way that they can be used effectively is as a supplement to med to short range brawler build without having to face the same kind of weight/slot sacrifices as the IS counterpart, and with no minimum range the basic tech is far more affective at this.

Edited by Phaeric Cyrh, 09 August 2014 - 11:17 AM.


#606 Gyrok

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 5,879 posts
  • Twitch: Link
  • LocationPeriphery of the Inner Sphere, moving toward the core worlds with each passing day.

Posted 09 August 2014 - 11:19 AM

View PostxxXKryotech OneXxx, on 09 August 2014 - 11:16 AM, said:

You know when I read comments that say " I wish those numbers from the test never came out ". It lets me know that the guys who own the clan mechs were never really totally upfront about the advantages the clans had over the is mechs. Especially when people were QQing about them. I never started threads about it myself but I did have my suspicions about issues with them. There's really no point in getting worked up about it. The numbers are out there, it is what it is. :huh:


No, if Clans had trial mechs, the numbers would be far more legitimately accepted. As it sits, you have pros on one side, and 5 day noobs on the other...who would you expect to win those matches?

They say the difference in ELO variance went from 40 to 90. Then they downplay the fact that is 225% more variance than they originally had...if matches are supposed to be 50/50 at 40% ELO variance, then a 125% increase in variance could very easily tip the scales to 90/10.

Get it yet?

Edited by Gyrok, 09 August 2014 - 11:20 AM.


#607 Phaeric Cyrh

    Member

  • PipPipPipPipPip
  • Mercenary
  • Mercenary
  • 123 posts

Posted 09 August 2014 - 11:26 AM

View PostGyrok, on 09 August 2014 - 11:19 AM, said:


No, if Clans had trial mechs, the numbers would be far more legitimately accepted. As it sits, you have pros on one side, and 5 day noobs on the other...who would you expect to win those matches?

They say the difference in ELO variance went from 40 to 90. Then they downplay the fact that is 225% more variance than they originally had...if matches are supposed to be 50/50 at 40% ELO variance, then a 125% increase in variance could very easily tip the scales to 90/10.

Get it yet?



Where do you get these numbers?

Edited by Phaeric Cyrh, 09 August 2014 - 11:32 AM.


#608 IraqiWalker

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 9,682 posts
  • LocationCalifornia

Posted 09 August 2014 - 11:35 AM

View PostPhaeric Cyrh, on 09 August 2014 - 11:26 AM, said:



Where do you get these numbers?

90 is 225% 40. It's basic math.

40 x 225% = 90

Edited by IraqiWalker, 09 August 2014 - 11:35 AM.


#609 Doctor Proctor

    Member

  • PipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 343 posts
  • LocationSouth Suburbs of Chicago, IL, USA

Posted 09 August 2014 - 11:37 AM

View PostGyrok, on 09 August 2014 - 11:19 AM, said:


No, if Clans had trial mechs, the numbers would be far more legitimately accepted. As it sits, you have pros on one side, and 5 day noobs on the other...who would you expect to win those matches?

They say the difference in ELO variance went from 40 to 90. Then they downplay the fact that is 225% more variance than they originally had...if matches are supposed to be 50/50 at 40% ELO variance, then a 125% increase in variance could very easily tip the scales to 90/10.

Get it yet?


Yes, and if they expected an ELO variance of 1 and it was instead 5, then it would be a 500% VARIANCE!!!! I can play with numbers too. The percentage of the variance is absolutely meaningless. What matters is how much of a difference that variance is expected to make. They have tons of data on how ELO variances affect the actual odds of match wins, so the fact that he largely glossed over the variance would lead me to believe that it doesn't make up for the 150% variance in win rate (90/60, see, more fun with scarily large numbers that don't mean much!).

Edited by Doctor Proctor, 09 August 2014 - 11:38 AM.


#610 Phaeric Cyrh

    Member

  • PipPipPipPipPip
  • Mercenary
  • Mercenary
  • 123 posts

Posted 09 August 2014 - 11:41 AM

View PostIraqiWalker, on 09 August 2014 - 11:35 AM, said:

90 is 225% 40. It's basic math.

40 x 225% = 90


******* duh!

Maybe I read the statistic wrong, and I have to admit that I don't know the ins and outs of the ELO system.. But I took the PGI tweet to mean that the ELO difference was between 40-90 points higher for the clans in the matches, and not that the variance increased from 40 to 90 in the test (a 50 point increase) as compared to the mixed matches.

If the former is true, it would be important to know the ELO scale, which I do not.. Hence my confusion as to how the 225% number was arrived at.

Edited by Phaeric Cyrh, 09 August 2014 - 11:42 AM.


#611 IraqiWalker

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 9,682 posts
  • LocationCalifornia

Posted 09 August 2014 - 11:50 AM

View PostPhaeric Cyrh, on 09 August 2014 - 11:41 AM, said:


******* duh!

Maybe I read the statistic wrong, and I have to admit that I don't know the ins and outs of the ELO system.. But I took the PGI tweet to mean that the ELO difference was between 40-90 points higher for the clans in the matches, and not that the variance increased from 40 to 90 in the test (a 50 point increase) as compared to the mixed matches.

If the former is true, it would be important to know the ELO scale, which I do not.. Hence my confusion as to how the 225% number was arrived at.


Your analysis is not that far off to be honest.

The main problem is that they mentioned the variance. Which is pretty much useless.

One team could have an average of 500 but have it be distributed so it's extremely lopsided (with most of it in 2 players, and the other 10 being near brain dead), while another team has 500 as well, but it's distributed evenly. The team with the 2 super players doesn't really stand a fair chance.

Throw in a variance that could be going into the favor of the clans, and it's pretty much a disaster in number crunching.

#612 Mazzyplz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,292 posts

Posted 09 August 2014 - 11:51 AM

View PostJoseph Mallan, on 07 August 2014 - 08:54 PM, said:

What... you know players will complain cause I will bring a body pillow vs their king bed pillow and demand a Range Nerf! :D



i bring throw pillow, but the dropoff on that range is atrocious :huh:

they just nerfed the speed

#613 -Natural Selection-

    Member

  • PipPipPipPipPipPipPipPip
  • Liquid Metal
  • Liquid Metal
  • 1,631 posts
  • Locationdirty south

Posted 09 August 2014 - 11:54 AM

View PostPhaeric Cyrh, on 09 August 2014 - 10:41 AM, said:



Sure.. Complaining just to complain... Right.

Nevermind the fact that I have never complained about anything in this game except clan tech.. Ever.

Also it's not like we have actual DATA supporting our position, while you have.... What exactly?

So who exactly is being closed minded here and staking a position simply out of ego?


More silly people who cant read. What exactly is "MY" position? Was it the part where I thought some of the quirk suggestions to limit "clan" mech builds sounded good? Get off your soapbox, I said nothing but the outcome of some of "MY" matches were bunk.

BTW, what clan tech feels out of whack to ya? What did you "data" tell you? A single ppc? a single gauss? A single erLL? A single AC? What build have you used that fells OP compared to others? And what was its downfall?

Funny as I only see people derping about nerfs, but don't single out a weapon, but more mechs and mech builds with nothing but "they can killz me!" as a complaint.

#614 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 09 August 2014 - 11:57 AM

View PostIraqiWalker, on 09 August 2014 - 11:35 AM, said:

90 is 225% 40. It's basic math.

40 x 225% = 90


Not quite for what we're talking about. For an average for a team of 1200, a 40 pt difference is about 3% of 1200. Your k-factor (how many points you gain or lose for wins/losses) is actually as much as 50 pts, though generally you're swinging 15-30 per match.

So that 50 pt Elo difference? Your personal Elo probably swings by more than 50 pts up and down in any given night. The Elo variance for teams isn't that big a deal.

I get the desire to blame 'trials'. The cold, hard brutal truth is that most players (Clan or IS) just aren't that good. Any benefit experience gives them is hamstrung by consistently poor habits. You'll notice we don't see general Elo stats anymore; I will bet you cash money that's because most 'experienced' players are actually statistically *below* average. That's just how it tends to go. You recall they also dropped the starting Elo for new players. It was quiet but it was done.

The Elo difference of 40-90 is absolutely marginal. I know Kiiyor tried to average them out to get performance stats but the reality is that it's as much akin to removing Heros as anything else. Plenty of people have bought C mechs for personal use; they don't denote new players. I constantly play matches as I'm sure we all do with clan mechs who don't break 100 on either or both sides.

Clan mechs are OP. We've tried to tell ourselves that 'yeah, Clan mechs do more damage but IS mechs are better at doing it all in one place' when, in fact, no. No they're not.

PPFLD is a significant advantage for probably 5% or less of total players who can consistently, accurately and intentionally hit the same location with each shot. For the other 95% or more that just means that their total DPS is lower because those shots that miss miss completely while a Clan weapon can be walked into target and on to location more easily.

We need to be honest that all damage is good damage. Blown off locations reduce fighting effectiveness overall and are not 'wasted'. Clans do a LOT of damage and they pour it out like a fire hose. I think it's awesome; I enjoy it a lot. Outside of top tier competitive players though the IS PPFLD weapon setup for ACs as their only 'advantage' just isn't much of one. Also, the majority of IS mechs are terrible. Terrible, in fact, on purpose. Designed to be 'hard to grind' because at some point that was considered a good design choice.

That's a bigger factor. There are no crap Clan variants you need to grind. Every single Timber Wolf is likely loaded well, or at least well compared to the competencies of the player. Not so with IS mechs. Some variants are just garbage. Some (like the BM 1G) get pointless torso twist nerfs to go with their Awesome-sized hitboxes, because otherwise they wouldn't be BAD, and you can't have an IS mech that isn't mostly designed of BAD, right?!?

That's the bigger issue. QD, Awesome, BM and several other IS mechs need resized down. Plenty of IS mechs need a hitbox fix. They all, every single one, need quirks and rebalances to scrub out all that 'well, this variant is trash but you need to get basics on it to get master on the one variant that isn't ****' trash.

Also, if IS engine flexibility is supposed to be an actual perk than lift the engine caps. It's no sort of 'perk' compared to Clan XL. that's another big factor between them - that whole 'I'm pretty much a STD engine with XL weight' thing Clans do? That's a massive advantage that can't be overstated.

This balancing needs done by un-nerfing IS mechs. I'm not going to say 'buffing', I'm going to say un-nerfing, because that's what it was.

IS mechs were mostly designed to be a struggle to level up and have at best 1 good variant per chassis. That's a big crux of all of this. Clan mechs were designed to be fun - IS mechs were designed to be an un-fun grind you had to complete to get to the fun. One of these is a better design choice and it's not hard to tell which.

#615 Hikyuu

    Member

  • PipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 238 posts

Posted 09 August 2014 - 11:57 AM

View PostMister D, on 07 August 2014 - 09:27 PM, said:


3/3/3/3 fixed things up for the IS matches, but it can't possibly account for the clan firepower, as I mentioned above, each Clan mech is 1 weight class more powerful than IS equivalent and then some.



Smartest thing mentioned on this thread. just so you guys know. someone nailed it right here.

3/3/3/3 is fine when you're dealing with mechs on the same tech level, for clan mechs they need to create a different system, where mediums get graded in a seperate tier as regular mediums, their heavies get graded in tiers other than IS heavies, and so on.

it's arguable on their lights however, because although theyre fairly heavy and durable, they're still lights and are slower and more fragile. they just don't work like IS lights

#616 IraqiWalker

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 9,682 posts
  • LocationCalifornia

Posted 09 August 2014 - 12:00 PM

View PostHikyuu, on 09 August 2014 - 11:57 AM, said:

Smartest thing mentioned on this thread. just so you guys know. someone nailed it right here.

3/3/3/3 is fine when you're dealing with mechs on the same tech level, for clan mechs they need to create a different system, where mediums get graded in a seperate tier as regular mediums, their heavies get graded in tiers other than IS heavies, and so on.

it's arguable on their lights however, because although theyre fairly heavy and durable, they're still lights and are slower and more fragile. they just don't work like IS lights


See, that's the thing though. Right now, our IS mechs are jihad-era in terms of tech level. All we're missing is just some of the weapons, and plenty of the unnecessary nerfs gone.

#617 BloodStar

    Member

  • PipPip
  • Philanthropist
  • 23 posts

Posted 09 August 2014 - 12:01 PM

Apologise for what? People with XL engines brawling then crying when they die? Or people not loading out their mechs for DPS and weapon cycling rather than imitating the Clans high alpha.

Sorry, but this is literally l2p. Easily going 500 damage per game in a Victor, and I've been playing for a week. Shouldn't be hard for the vets to cotton onto the gameplay style and counter them.

If you want to complain about anything, complain about LRM firing arcs, chain firing and general spam, or maybe even beam time on certain weapons or lack of viable IS alternative.

The only think this win statistic shows is that the lead designer has no idea what he's doing and will listen to the idiots calling for unnecessary nerfs instead of buffing the underpowered stuff to actually be competitive.

#618 Hikyuu

    Member

  • PipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 238 posts

Posted 09 August 2014 - 12:07 PM

by that standard, you would see more Clan Lights and mediums and a few Heavies and Assaults. making them less ridiculous.

#619 Phaeric Cyrh

    Member

  • PipPipPipPipPip
  • Mercenary
  • Mercenary
  • 123 posts

Posted 09 August 2014 - 12:08 PM

View PostIraqiWalker, on 09 August 2014 - 11:50 AM, said:


Your analysis is not that far off to be honest.

The main problem is that they mentioned the variance. Which is pretty much useless.

One team could have an average of 500 but have it be distributed so it's extremely lopsided (with most of it in 2 players, and the other 10 being near brain dead), while another team has 500 as well, but it's distributed evenly. The team with the 2 super players doesn't really stand a fair chance.

Throw in a variance that could be going into the favor of the clans, and it's pretty much a disaster in number crunching.


Can you direct me to a source that confirms that this is how it works?

Why do you think the MM is intentionally assembling a bunch of terrible IS players offset by a couple elite players while the clans have a relatively homogenous assembly of talent? It doesn't make sense really, except to try and explain the results in a way that more easily fits in with your opinion.

Also, this is a variable that is so terribly easy to see and understand that it took the forum all of 10 minutes to begin harping on yet the guys at PGI are completely ignorant of it? It's possible, but it just comes off as arrogance and a bunch of wannabe game designers thinking they are smarter than everyone else.

Edited by Phaeric Cyrh, 09 August 2014 - 12:12 PM.


#620 Hikyuu

    Member

  • PipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 238 posts

Posted 09 August 2014 - 12:11 PM

View PostIraqiWalker, on 09 August 2014 - 12:00 PM, said:


See, that's the thing though. Right now, our IS mechs are jihad-era in terms of tech level. All we're missing is just some of the weapons, and plenty of the unnecessary nerfs gone.


right but following timeline we would only have pre-invasion technology, it's not the technology that's wrong, it's the formula.

I hate to refference WoT, but this is why they have a tier system to their tanks. people don't like to admit that an Atlas is better than a Victor, but by it's stats it's a far more intense assault mech, this is what's happening with Clan assaults. they're far more intense than IS assault mechs and they're not in the same class, treating them in the same class is going to make them overpowered, nerfing isn't the solution, changing how matchmaking works is.

most of this wont matter unless it's CW play, so my suggestion is that CW grade clan mechs differently than IS mechs in terms of battle value, making clans focus more on smaller lighter team setups. while the IS can take heavier mechs on the field.

for pugging it'll probably stay Free For All so that helps balance it out (clans and IS on both sides.)





3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users