145 replies to this topic

[mekabuser]

so in case i missed it, how do you propose to balance from the top down. ?

[Stanton Langley]

Some good points regarding ELO. Not sure that I subscribe to all of it, but I agree there should be unranked matches available. I don't mind getting steamrolled when I test a new build, but it shouldn't skew my or others' ratings. Visible ratings, even in the form you propose, have the potential to be a negative impact on the game as people will game the system to achieve their own goals.

[WolvesX]

Stanton Langley, on 16 March 2013 - 11:20 AM, said:

Some good points regarding ELO. Not sure that I subscribe to all of it, but I agree there should be unranked matches available. I don't mind getting steamrolled when I test a new build, but it shouldn't skew my or others' ratings. Visible ratings, even in the form you propose, have the potential to be a negative impact on the game as people will game the system to achieve their own goals.

Why would you get steamrolled in unranked matches with Elo ?

[Hobietime]

Great discussion. I agree that the top certinly needs more weight from a balance standpoint. However, I do recall either on Kutaku or another interview of the LoL dev team that for a while, they had focused too much on top tier balancing and had made it a mess for lower tiers because of this.

Extra Credits actually has a great video on this subject.






Basicly, we need things like BoomCats and Splats, but we also need higher skill, more powerful weapon combinations. Something that gives a great player a slight advantage.

I would suggest having the different weapon systems complement each other. So that hitting a component with a laser after a ballistic gives you a slight damage bonus. The same would go for missiles and lasers, and missiles and ballistics. Then you would have to carefully choose varied load outs and consistently score hits on the same target within all of your weapon systems' optimum range to get just a bit more damage out of your 'mech.

This won't chage the game at all for low skill players and will add huge amounts of depth to the upper level meta game.

Edited by Hobietime, 17 March 2013 - 10:28 AM.

[XIRUSPHERE]

ELO won't change the ghetto during USTZ primetime, it's the ghetto no matter what game you play if you are solo dropping.

[WolvesX]

XIRUSPHERE, on 17 March 2013 - 09:59 AM, said:

ELO won't change the ghetto during USTZ primetime, it's the ghetto no matter what game you play if you are solo dropping.

I don't understand waht you just said.

[Jonathan Paine]

Ironically, weapon combos are already in the game. Missiles do large chunks of damage, but tends to be spread out. Following up with a more concentrated weapon, like a PPC or a large caliber AC, can then take out a previously damaged area. For those with lesser aiming ability (me!), simply revers the process: PPC first to create a weakened area, then missiles or an LBX to maximize the chance of hitting the already damaged area.

Back to Elo. Perhaps players of a high Elo rating could drop unranked in trial/stock mechs? I also like the idea of having at least two options for dropping into a game. The current " just give me a game already" quick and dirty drop, and perhaps an " I am OK waiting five minutes for two well balanced teams"

[WolvesX]

Jonathan Paine, on 17 March 2013 - 02:03 PM, said:

Ironically, weapon combos are already in the game. Missiles do large chunks of damage, but tends to be spread out. Following up with a more concentrated weapon, like a PPC or a large caliber AC, can then take out a previously damaged area. For those with lesser aiming ability (me!), simply revers the process: PPC first to create a weakened area, then missiles or an LBX to maximize the chance of hitting the already damaged area.

Back to Elo. Perhaps players of a high Elo rating could drop unranked in trial/stock mechs? I also like the idea of having at least two options for dropping into a game. The current " just give me a game already" quick and dirty drop, and perhaps an " I am OK waiting five minutes for two well balanced teams"

Same here, balanced teams would be great!

[WolvesX]

Elo should be not for classes, but for single mechs.

Explaination:

AWE-M9 -> Player X got exactly the same build as player Y.

AWE-M9 with player X got a rating of 3000.
AWE-M9 with player Y got a rating of 1700.

This makes every match fair, regardless of the equip or the mech.

AWE-M9 with 6 small laser, could have a rating of 3000, IF player X can do as much good as a an other player with 4 PPC.
AWE-M9 with 5 LLs 20 DHS..., could have a rating of 20, IF my 4 years old cousin plays it.

Totaly balanced and fair.

[WolvesX]

Hobietime, on 17 March 2013 - 09:25 AM, said:

Great discussion. I agree that the top certinly needs more weight from a balance standpoint. However, I do recall either on Kutaku or another interview of the LoL dev team that for a while, they had focused too much on top tier balancing and had made it a mess for lower tiers because of this.

Extra Credits actually has a great video on this subject.






Basicly, we need things like BoomCats and Splats, but we also need higher skill, more powerful weapon combinations. Something that gives a great player a slight advantage.

I would suggest having the different weapon systems complement each other. So that hitting a component with a laser after a ballistic gives you a slight damage bonus. The same would go for missiles and lasers, and missiles and ballistics. Then you would have to carefully choose varied load outs and consistently score hits on the same target within all of your weapon systems' optimum range to get just a bit more damage out of your 'mech.

This won't chage the game at all for low skill players and will add huge amounts of depth to the upper level meta game.

Extra Credit... so informative, every time!

[WolvesX]

Any other opinions / critics / analysis on this topic?

[MadBomberMan]

Broceratops ftw lol XD

[Glythe]

WolvesX, on 13 March 2013 - 04:40 AM, said:

I think a game should be balanced around HIGH Elo, around competitive environments and

not around PUG games.

Therefore, what's really needed is a balance around HIGH Elo gameplay, like in LoL, Dota, Starcraft.



Yup! I also don't like to play my best builds all the time, some times I want to play a 4 MG spider not matter how ridiculous it is. Therefore we need:

Ranked and Unranked games.

Also an option like: Searching for Games OR Searching for FAIR Games, resulting in a longer search, would be great.


EDIT:

Elo should be not for classes, but for single mechs.

Totaly balanced and fair.

I agree with all of this.

In short we need ranked (elo matching) and unranked play (random matching with balanced weight/ecm).

WolvesX, on 13 March 2013 - 04:40 AM, said:

Elo is a must in both to have a fair match, but in ranked games, the mindset is: I want to win! As opposed to the casual mindset I want to shoot mechs.


As a matter of balance and fairness:

We don't need open stats, but we need small "shields" near our name, like:

GREEN = Rookie
BLUE = Mechwarrior
ORANGE = Veteran 25%
RED = Elite 5%
(GOLD = Legendary 1%)

and for a matter of balance, or example Elites should no matched against Rookies and if the MM system does match them, you know there is a problem (ingame).

But this is just wrong. Unless you reognize my name from either the forums or from another encounter you shouldn't know if I am the best player or the worst player on the team. World of tanks allowed you to see (with a mod) who was the best player on the team and it ruined the game if you were one of those people. You can imagine why if you have ever been in a game with Garth Erlam and everyone guns for him first. Now then... after the game if it showed everyone's score in a ranked game that would be fine.

MWO is not chess. As such you can't hope to apply Elo. This matchmaker is the worst to date (even worse than 8 man premade vs 8 randoms). Balance weight and balance ECM and let a random MM sort out the rest. With less waiting there will be more matches. In all likelihood there is a higher probability of finding real better matches (because what we have is terribly broken).

It is nearly impossible to try and match people with different weights and different skills. Assuming 0 is worst and 1 the best at what point does an Atlas equal a centurian? Is a .8 centurian worth a .6 Atlas? Since there is apparently no accounting for weapons loadouts, heat effeciency, or firepower all mechs of the same type are treated as equals (with possible user data tweaks based on win percentage). This is only going to make things worse.

[Joseph Mallan]

As I have said, I do craptastic and teh team wins, How is that a reflection on my performance in game. I ROTFSTOMP and the team loses. How is that a reflection of My performance? If individual performance was tracked cool, but a team's success or failure is not based on the individual's performance.

[xRaeder]

If you want all chassis to be viable then what needs to happen is a serious rework of the chassis quirk system.

Take the Hunchback 4G for example... Before the Highlander patch (when all the poptarting started), I was averaging 600-1000 damage with 1AC20 and 2MLs + 250std engine and the rest ammo. This was even before the stat rewind patch where I was consistently hitting lights through the lagshield.

After the poptarting started, and everyone was either piloting a Highlander or heavy alpha strike Stalker my damage went down to 200-500 per match... which ended up being mostly because the RT was so big and everyone had a huge alpha that would take out the AC20 in one shot(s).

I played about 20 games in my trusty 4G to find that it wasn't so trusty anymore. I then switched to a poptart Cataphract-3D and in my first game I did 1000... and in 15 more games I averaged around 700-1000 damage just like before (I'm usually in a non-min/max Mech because the high alpha cheese builds that many high ELO players use makes things too easy - in other words I could be up there with you guys... but I choose not to be... the problem is that I'm not a typical player, and these issues will frustrate a lot of players... new and old, who will/have stopped playing).

So, I've basically shelved my 4G and am using the YLW which is the same basic concept but much more viable.

Now... to the quirk system. The 4G should be immune to critical hits on the RT, and should be able to hold more ballistic ammo per ton than other chassis. That would potentially make the 4G more relevant than it is now.

Also... regardless of what your ELO ranking is... the high alpha strike must have that this game currently encourages makes the game to be pretty lame. Penalties for certain Mechs to boat weapons needs to be examined.

My own personal suggestion it to look at adding heat penalties for boating too many weapons in chassis that were not intended for it (see the Awesome and the Hunchback 4P as examples).

But I've also seen a suggestion taken from the Battletech books in that even the biggest fusion engine could only put out so much power and that you couldn't alpha strike too many weapons at once without getting a firing delay as the engine prioritized power... so if you fired off 6 PPCs only 3 of them would fire with a delay of 2-3 (or some number) seconds for the rest of them... so each weapon would have a power rating, and if your current fire exceeded the engine's maximum amount you'd get fire delay.

Hunchback 4G Stats:
HUNCHBACK HBK-4G 77 52 24 2.17 93 37 2.51 28,102 50,543 06:29:53


Cataphract 3D Stats:
CATAPHRACT CTF-3D 44 22 22 1.00 63 22 2.86 20,966 26,968 04:10:20

Edited by xRaeder, 14 May 2013 - 04:42 PM.

[WolvesX]

Its not about chassis, its the value of the player in that specific chassis that matters for team balance.

[xRaeder]

WolvesX, on 15 May 2013 - 04:18 AM, said:

Its not about chassis, its the value of the player in that specific chassis that matters for team balance.

Then explain why there are so few Hunchbacks (ALL variants) vs Stalkers or Highlanders. I rarely see Trebuchets anymore, far fewer Atlases, etc. Current weapon and hardpoint loadouts have made many chassis non-viable.

[Deathlike]

xRaeder, on 15 May 2013 - 06:00 AM, said:

Then explain why there are so few Hunchbacks (ALL variants) vs Stalkers or Highlanders. I rarely see Trebuchets anymore, far fewer Atlases, etc. Current weapon and hardpoint loadouts have made many chassis non-viable.

I've seen them on the field just fine.. although some more than others (I'd see more 4P and 4SPs than I would a 4G). The only bucket that seems to be on the field is the 7M for whatever the reason, and I still see plenty of Fatlases of all kinds (even the stray K).

Perhaps my ELO is bad, but honestly most mechs are still being fielded... some variants more than others for obvious reason.

[WolvesX]

Deathlike, on 15 May 2013 - 06:31 AM, said:

I've seen them on the field just fine.. although some more than others (I'd see more 4P and 4SPs than I would a 4G). The only bucket that seems to be on the field is the 7M for whatever the reason, and I still see plenty of Fatlases of all kinds (even the stray K).

Perhaps my ELO is bad, but honestly most mechs are still being fielded... some variants more than others for obvious reason.

I think most players know that your elo isn't bad.^^

[Kitane]

That video from Extra Critic was informative and reminded me why I don't like the concept of balancing from the top, because it's often deceiving. It's a story from another simulation genre, but it is still valid.

More than decade ago I used to spend a lot of time in online WW2 flight simulator called Warbirds and the best example of a perceived balance was a power balance between Spitfire and Bf-109. Various versions of these planes shifted balance from side to side, but the most significant part of the perceived "OPness" was based on a skill level of involved players.

Essentialy:

Situation 1: two newbie players, barely able to hold themselves in air without entering vicious spins.
- Fairly random, they are clueless, they can't keep track of their enemy's position, they can't shoot...it might go to either side.

Situation 2: two more experienced players who already know how control their planes and can turn at low speeds without stalling
- Spitfire almost every time, unless it is surprised. It can turn faster and that's the most instinctive action new players do in air combat. At this point, people started to cry about OP Spitifire and Spitdweebs. 109 pilots learned to fly together to cover each other from more agile enemies.

Situation 3: Frustrated 109 pilot starts to learn about energy and how to preserve his energy advantage (speed and altitude ftw)
- Mostly 109 wins. Spitfires find themselves on defensive, dodging endless passes of fast uncatchable 109s who take advantage of their superior climb rate and acceleration to avoid close knife-fights that favor Spitifres. 109s hold the initiative through the whole fight and they are untouchable, unless they seriously f*ck up or get hit by a lucky bullet from long range.

Obviously at this point Spitfire pilot started to complain about at OPness of 109 and lameness of vertical Boom'n'Zoom tactics. 109 pilots disregarded that, because they were lulled by their self-perceived sense of superiority, conveniently avoiding the fact that Spitfires were at disadvantage in engine power contest.

Situation 4: The Spitfire pilot is fed up with arrogant Luftwaffles and focuses fully on understanding of the real core of the air-to-air combat (too long for this post). He becomes efficient, aware of his energy state, minimizes energy expenditure on maneuvers to bare essentials and learns how to capitalize on their relative positions and current energy levels. Ordinary 109 pilots become easy fodder for him because he can preserve enough energy to catch them and kill them and unless their flight skills aren't top notch, they will be unable to get advantage over him again, they are too inefficient.

And when he meets a 109 pilot who has the same knowledge, the match becomes balanced or almost balanced, a thinking man's match, because human mistakes have far greater impact than relative strengths of airframes.

As the pilot skill grows, the so-called balance changes, often drastically. Yet many people fall off from that climb, leaving with bitter memories of being overpowered by OP enemy.

If we had ELO back then and planes were balanced by requirements of top players, newbies and intermediate players would perceive the balance as horribly flawed, veterans would have a vague understanding of underlying concepts but they would still frequently and vocally disagree. And a small group of elite players would be satisfied, isolated from the rest by having a completely different understanding of what actually matters.

It's nice to have a true balance, but it will not help the game if 99% of the players are not skilled enough to see it.

Edited by Kitane, 15 May 2013 - 08:01 AM.