367 replies to this topic

[Kirkland Langue]

Big Tin Man, on 15 August 2016 - 10:10 AM, said:

...and I forgot to mention the +15 CT structure on all variants, as well as other structure and mobility quirks on variants that aren't the KDK-3, clan endo/ferro, 45.5 tons of space with full armor + max engine, masc that can go above 90, huge number of hardpoints, shield arms large enough to shield, humanoid form. Take your pick. Nothing even gets close to this perfect storm of positive mech traits in an assault.

The only negative thing I can come up with against the Kodiak is the CT hitbox being somewhat large. But it has a quirk to balance that.

Curious what you think my logic flaw is here...

Logic flaw?
"One particular player appears 4 times on the leaderboard with his main, that proves how OP one of those mechs is".
I was being overly generous to say there is a flaw with that logic.

As for the Kodiak - Your answer shows why PGI's method for balancing matches is abysmal, and they really should have gone with BV. It isn't too late for them to create a BV system - but Russ doesn't give a damn about MWO any more so zero chance that he wants to think about MM.

[KodiakGW]

BigBenn, on 15 August 2016 - 10:12 AM, said:

If it is truly the pilot and not the mechthen we'd see a lot more different mechs on the battlefield. There are a whole host of mechs we simply do not see. The Trebuchet, Dragon, and Quickdraw quickly come to mind. When was the last time you saw a Mist Lynx? Orion? The crutch mechs are what they are (Cheeter and Kodiak are the two most prominent, atm). Sure, there are different mechs with different abilities, and players with different abilities, but certain mechs will always be favored due to their offensive power, defensive might, etc. Other mechs, like the Dragon simply cant cutit due to hit boxes and weapon location.

It is what it is.

I don't know what queue you are in, but I saw a couple of Mist Lynx this weekend. Always on my team for some reason. Plenty of Quickdraws and Trebs during the medium mech event, and I pilot a Dragon quite regularly. Maybe you didn't see my video last month?

There is a reason why I took the Kodiak. I wanted to finish of the KMDD part of the event. I also got three from taking an EBJ three games prior. Both those got mechs got me over 10 so I could stop playing. I wouldn't have been able to otherwise because there was nothing but stomps.

Edit: You really got to love someone who calls out that he doesn't see Quickdraws when my screen shot SPECIFICALLY HAS A QUICKDRAW BEING PILOTED BY A 228 MEMBER GETTING HIGH SCORE. Pay attention next time.

Edited by KodiakGW, 15 August 2016 - 10:35 AM.

[Revis Volek]

KodiakGW, on 15 August 2016 - 10:18 AM, said:

Evidentially you didn't see my earlier post saying I saw two 228 and an EmP player all o the same team that produced a 1-12 stomp. I was in a heavy that time. This time I took a mech I know I could get at least one KMDD before losing. I got two this round, and finished off two before dying. That, and the work of the other three non-scrubs is the only reason why it was even close this time.

So tell me, what is the statistical probability that out of 9 games I played I played between that 1-12 stomp (including it) and the one posted, over a two day period, that I would see three players of some of the top tier comp teams all on one side, and not once see them teamed with me?

I know statistics, and the probability of that happening is pretty damn high.

The Elo code for making "underdog" teams is still running on their matchmaker server. Couple that with PSR being an XP bar, and you get lousy games. As well as inflated tournament scores for those lucky enough to not to be on the "CARRY HARDER" queue.

You got it all figured out, PGI, 228, EmP and everyone else is against you damn scurbs because you are bad at game so we stack teams against you.


And where did you get hard evidence tha PGI was cabaple of pragraming and UNDERDOG code into Elo? ELO doesnt exist and doesnt have an UNDERDOG code to make it work. You really dont understand how Elo or MM works in general it seems.



Tournament scores dont get inflated because you CARRY SO HARD WTF BBQ OP they get inflated because people dont have lives and just plays this game alllllll weekend long farming good games and naturally you will get 10 games of better scores then someone who only played a normal few hours a day.

Again, your lack of understanding is the problem here.

[Big Tin Man]

Kirkland Langue, on 15 August 2016 - 10:26 AM, said:

Logic flaw?
"One particular player appears 4 times on the leaderboard with his main, that proves how OP one of those mechs is".
I was being overly generous to say there is a flaw with that logic.

As for the Kodiak - Your answer shows why PGI's method for balancing matches is abysmal, and they really should have gone with BV. It isn't too late for them to create a BV system - but Russ doesn't give a damn about MWO any more so zero chance that he wants to think about MM.

Would you prefer to just simply compare how much higher the Kodiak scores are on average compared to every other chassis instead? I was taking away the 'It's not the mech, it's the pilot' argument, since it's clear as day the top 75 scores show a significant statistical advantage to the Kodiak and people were denying that, saying people aren't using other capable mechs to their full potential. So here's the same very capable pilot, 1300 point spread between his kodiak and the next mech.

As for the MM, across this many matches, I simply cannot believe that there are enough completely one sided matches to create the sheer volume of high scores. 12-0 stomps usually end up with the winning side all having 1-2 kills per person and ~400 damage per person. That doesn't qualify for these leaderboards.

[KodiakGW]

Revis Volek, on 15 August 2016 - 10:36 AM, said:

You got it all figured out, PGI, 228, EmP and everyone else is against you damn scurbs because you are bad at game so we stack teams against you.

And where did you get hard evidence tha PGI was cabaple of pragraming and UNDERDOG code into Elo? ELO doesnt exist and doesnt have an UNDERDOG code to make it work. You really dont understand how Elo or MM works in general it seems.

Maybe you should check the history of this game before making claims.

http://mwomercs.com/...ted-april-19th/

Specifically:
"We discovered a bug that was causing player Elo values to become inflated. Players that would win a game against a weaker opponent were being rewarded as though they were the underdog in that match. This caused Elo values to increase by a lot when they should have increased by only a little, skewing the player Elo distribution so that it became top-heavy."

There were a number of posts around this time where players, who are long gone, found out that the matchmaker was making "underdog" teams so that their version of an Elo system could work this way.

And I never said it was against scrubs, it is making top heavy teams that lead to stomps. So take that tin foil hat and file it.

Edit: Yeah, done. Plenty of other threads about stomps. If you all don't want to come to a realization that something is wrong, than you'll never learn. Laters.

Edited by KodiakGW, 15 August 2016 - 10:59 AM.

[Big Tin Man]

..psst: PGI scraped Elo and put the PSR system in it's place. Elo isn't a thing anymore.

edit: and yes, PSR has flaws. Big ones, but different flaws.

Edited by Big Tin Man, 15 August 2016 - 11:12 AM.

[Kirkland Langue]

Big Tin Man, on 15 August 2016 - 10:41 AM, said:

1300 point spread between his kodiak and the next mech.

One of the problems with this leaderboard is that it doesn't distinguish between chassis.
I "suspect" that most of the Kodiak leaderboard is full of KDK-3 builds.
but that is also because I "suspect" that almost all of those top scores were done with Dual Gauss and Dual cERPPCs.

[Kubernetes]

"Underdog" in that case means the team with lower ELO. Despite best efforts, MM probably could not mix 24 players to create an exactly matched 12v12, so the lower ranked team is the underdog. Under ELO, your gain upon victory depends on your opponents' ELO. In the post above, PGI is simply saying that ELO was incorrectly assuming that the stronger team was the underdog. That's it, no conspiracy, no stacking.

And regarding the match with EmP and 228 pilots, the MM has no idea if these guys are "regular T1s" or "elite T1s." It just knows that they're T1 and will try to match other T1s against them.

Edited by Kubernetes, 15 August 2016 - 11:21 AM.

[Quicksilver Aberration]

KodiakGW, on 15 August 2016 - 10:49 AM, said:

Maybe you should check the history of this game before making claims.

Maybe you should check what is being used by the matchmaker before using 3 year old announcements? This game doesn't use ELO, at all. Once they added PSR, ELO was thrown out.

Kubernetes, on 15 August 2016 - 11:17 AM, said:

And regarding the match with EmP and 228 pilots, the MM has no idea if these guys are "regular T1s" or "elite T1s." It just knows that they're T1 and will try to match other T1s against them.

This, even within tier 1, there is a giant skill gap.

Edited by Quicksilver Kalasa, 15 August 2016 - 11:23 AM.

[Manei Domini Krigg]

Kodiak is best for fight VS OTHER ASSAULTS but it not so good vs lights and fast mediums/heavys.

[Big Tin Man]

Kirkland Langue, on 15 August 2016 - 11:14 AM, said:

One of the problems with this leaderboard is that it doesn't distinguish between chassis.
I "suspect" that most of the Kodiak leaderboard is full of KDK-3 builds.
but that is also because I "suspect" that almost all of those top scores were done with Dual Gauss and Dual cERPPCs.

Now this is funny. His chassis that is 1300 points lower is a Dire Wolf. That can run IDENTICAL weapons loadouts to any kodiak. Except it's slow, can't turn, is shaped like a cube, has low weapons mounts, can't arm shield...

but keep telling me that the kodiak isn't OP compared to the DWF. I'm listening. Maybe the magical Mk VII targeting computer will make up for all of that.

[Quicksilver Aberration]

Kirkland Langue, on 15 August 2016 - 11:14 AM, said:

but that is also because I "suspect" that almost all of those top scores were done with Dual Gauss and Dual cERPPCs.

I'm not sure why you suspect that, while I prefer that build, it doesn't farm the damage like dakka (which is not gone). I'm willing to bet the top builds are actually using 2 UAC10/2 UAC5/1 ERPPC to get there.

[Revis Volek]

Kirkland Langue, on 15 August 2016 - 11:14 AM, said:

One of the problems with this leaderboard is that it doesn't distinguish between chassis.
I "suspect" that most of the Kodiak leaderboard is full of KDK-3 builds.
but that is also because I "suspect" that almost all of those top scores were done with Dual Gauss and Dual cERPPCs.

Wut?

and i suspect you have no idea what you are talking about.

2x10's and 2x 5's, take a look at other mechs loadouts once and a while and you will what what they are running and you wont have to SUSPECT anything anymore!


Man, target info is so OP.

Quicksilver Kalasa, on 15 August 2016 - 11:25 AM, said:

I'm not sure why you suspect that, while I prefer that build, it doesn't farm the damage like dakka (which is not gone). I'm willing to bet the top builds are actually using 2 UAC10/2 UAC5/1 ERPPC to get there.

I also heard 2 10's and 2 LPL's wasnt half bad

[Kirkland Langue]

Big Tin Man, on 15 August 2016 - 11:24 AM, said:

Now this is funny. His chassis that is 1300 points lower is a Dire Wolf. That can run IDENTICAL weapons loadouts to any kodiak. Except it's slow, can't turn, is shaped like a cube, has low weapons mounts, can't arm shield...

but keep telling me that the kodiak isn't OP compared to the DWF. I'm listening. Maybe the magical Mk VII targeting computer will make up for all of that.

I never once said that the Kodiak isn't OP.

Mechs are difficult to balance. Mechs have different Hardpoints located in different spots. Mechs have different sizes and shapes. I asked you why the Kodiak was OP, and your answer was "EVERYTHING!". We all see that the Kodiak scores are much much higher than the other Assaults - yet none of us can really point to a specific reason why this is.

How exactly should the Kodiak be nerfed such that it's score is in line with the others?
Do we really consider the scoring system to be fair in the first place?

The entire discussion is very frustrating - because almost none of this stuff matters in the slightest; or, rather, it shouldn't matter. It's impossible to completely balance the mechs - let's get that clear right from the get-go. But that shouldn't be a problem. Why in the world do so many players think that it's important for the mechs to be balanced? Go play Call of Duty or some other shooter, where all players are identical, just wearing different skins, if you want perfect balance.

The problem isn't that mechs aren't balanced - the problem is that PGI decided, in their great wisdom, to reward performance identically for all mechs. The problem is that PGI decided, in their great wisdom, to build a matchmaker around the notion that weightclasses could be balanced within itself.

And the players are just as stupid as PGI. So many players worried about Mech Balance (which is impossible, might as well be crying about the inevitable doom of our planet as the sun will one day devour the planet) --- instead of being worried about team balance or game balance or roles.

I'd keep going, but it'd just devolve into rage because I hate MWO even though I want to like it. Sure, it's okay to jump into games and smash some mechs - but in reality, it's an awful game developed by an awful company, and full of awful players. I just want to like it because I like mechs and the game has so much potential. It's just a pity that neither the players nor the developer are even slightly interested in trying to achieve that potential.

[Quicksilver Aberration]

Revis Volek, on 15 August 2016 - 11:43 AM, said:

I also heard 2 10's and 2 LPL's wasnt half bad

Not nearly as good as full dakka because the 10s don't really play well with hitscan weapons, that's why they use the ERPPC instead of the LPL on the 2 UAC10/2 UAC5 build.

[Quicksilver Aberration]

Kirkland Langue, on 15 August 2016 - 11:44 AM, said:

Why in the world do so many players think that it's important for the mechs to be balanced?

Why would you want to play as a pawn if something as potent as a queen is available? That's exactly the reason, because no would give a **** about anything but assaults in this game without special restrictions (like 3/3/3/3 or tonnage limits).

If you think that having mechs being equal means that they are just different skinned version of the same thing, I suggest you look at Overwatch which offers characters with very different playstyles/roles and stop relying on worthless strawmen as part of an argument.

Edited by Quicksilver Kalasa, 15 August 2016 - 11:50 AM.

[Kirkland Langue]

Quicksilver Kalasa, on 15 August 2016 - 11:25 AM, said:

I'm not sure why you suspect that, while I prefer that build, it doesn't farm the damage like dakka (which is not gone). I'm willing to bet the top builds are actually using 2 UAC10/2 UAC5/1 ERPPC to get there.

Heck, we don't even know that it's the KDK-3 that people are using on the leaderboard.
But we all "suspect" that it is.

I agree that the dakka builds are very strong. But the gauss builds are what the top players are running in competitive play and I've seen far far too many 1200 damage games using gauss builds.

[Templar Dane]

Big Tin Man, on 15 August 2016 - 10:41 AM, said:

Would you prefer to just simply compare how much higher the Kodiak scores are on average compared to every other chassis instead? I was taking away the 'It's not the mech, it's the pilot' argument, since it's clear as day the top 75 scores show a significant statistical advantage to the Kodiak and people were denying that, saying people aren't using other capable mechs to their full potential. So here's the same very capable pilot, 1300 point spread between his kodiak and the next mech.

As for the MM, across this many matches, I simply cannot believe that there are enough completely one sided matches to create the sheer volume of high scores. 12-0 stomps usually end up with the winning side all having 1-2 kills per person and ~400 damage per person. That doesn't qualify for these leaderboards.

The more people playing the chassis the harder it is to get top 5, and the higher the score gets. There was also a lot of damage to be farmed because of all the assaults.

[Quicksilver Aberration]

Kirkland Langue, on 15 August 2016 - 11:50 AM, said:

Heck, we don't even know that it's the KDK-3 that people are using on the leaderboard.

If it isn't that doesn't really bode well for the Kodiak since the 3 is the best variant, even after the nerfs. It has the most ballistic mounts out of all the Kodiaks and that is the key for any assault that isn't blessed with energy hardpoints like the Wubmaster/Wubshee.

Kirkland Langue, on 15 August 2016 - 11:50 AM, said:

I agree that the dakka builds are very strong. But the gauss builds are what the top players are running in competitive play and I've seen far far too many 1200 damage games using gauss builds.

As I said, dakka is better for farming stats, which is why it is highly likely that dakka is the build being used. As for top players using ERPPC/Gauss vs Dakka, honestly it seems like a split, they both have their uses.

Edited by Quicksilver Kalasa, 15 August 2016 - 11:54 AM.

[Deathlike]

Quicksilver Kalasa, on 15 August 2016 - 11:53 AM, said:

If it isn't that doesn't really bode well for the Kodiak since the 3 is the best variant, even after the nerfs. It has the most ballistic mounts out of all the Kodiaks and that is the key for any assault that isn't blessed with energy hardpoints like the Wubmaster/Wubshee.

It's not really just about "the most", it's really about "how high" those mounts are located.

Even if we somehow magically created an Atlas with dakka in both torsos (assume the LT is a mirrored RT), the biggest problem the Atlas has is the low mountedness of those hardpoints. Even if we inflated it to insane levels (like having a 4B LT like the Banshee-3E), it would still likely to keep a shameful level of low mounts. A Mauler would still be a better dakka option than the Atlas.

Unless the Atlas and other low torso weapon mounted mechs got some sort of #torsolift (think plastic surgery for mechs to lift their weapon mounts), a mech that has higher mounts will be preferable for all ranges and builds... which almost always increases the potential for better damage dealing on the whole.

Even the Awesome's weapon nipples gives itself a shot of "sucking less". I don't want to talk about lurms or the Pretty Baby though.