Jump to content

Team Balancing Based On Mech Cost Instead Of Just Weight.


48 replies to this topic

#21 A Man In A Can

    Member

  • PipPipPipPipPipPipPipPip
  • 1,594 posts
  • LocationRetired

Posted 20 July 2013 - 02:02 PM

View PostKhobai, on 20 July 2013 - 01:35 PM, said:


Its not a start. All that accomplishes is adding a new way for players to abuse the game. I can make a low cost Stalker with a standard engine that will obliterate any mech twice its cost easily.

This issue happened even with the BV system as players found ways around it to load up on more firepower, so its nothing new.

Plus, it's not just the mech we're balancing here. It's the player's skill with the mech complete with engines modules weapons etc in conjunction with soon to be 11 other people on your team and 12 more on the other side.

Edited by CYBRN4CR, 20 July 2013 - 02:09 PM.


#22 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 20 July 2013 - 02:28 PM

To clarify I ment the system would use elo + tonnage + cost to balance matches. And the formula could be tweeked to find the best balance. You could also figure in the pilot trees for even a better matching system.

As for a mech costing a lot but not being good well that is also true for weight. You will always have some people spending a lot on a bad build. No system will ever be perfect. But the question is will adding cost into the formula help or hurt. I believe it will help balance matches not hurt balance.

Here are some reasons why it will help....

1. Custom mechs are almost always better than stock mechs.

2. New players mostly use stock mechs and then cheaper mechs even when they start using custom mechs. For example they normally do not have the money even with the bonus for a custom Atlas with seismic and all the other modules etc. And if you figure in the pilot trees you can even get a better match. So it will help match new players against other new players.

3. Multiple ways of matches teams will give a finer control. Elo + Cost + Tonnage + Pilot Tree.

What are some of the potential problems.....

1. Gaming of the system to make a low cost but effective mech. For example a Stalker with Standard engine. ---- This will still not count the same as a stock stalker. As and mentioned you will always have some people that build expensive mechs that do not perform well. Just like there will be heavy mechs that do not perform well. Also we can tweek it so ELO and Weight and or Pilot tree etc count for more or less than cost.

#23 Nik Van Rhijn

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,905 posts
  • LocationLost

Posted 20 July 2013 - 02:38 PM

The other thing I can't understand is why start people on a median Elo. This means that a total newcomer in a trial mech will be placed with "average" players in customised mechs. Even if Sulla's system isn't perfect it would be a hell of a lot better.
People when starting out for the first time should be near the bottom of the bell curve.

#24 Tor6

    Member

  • PipPipPipPipPipPip
  • 270 posts

Posted 20 July 2013 - 02:47 PM

Terrible terrible idea because the costs in this game are utterly divorced from balance consideration.

#25 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 20 July 2013 - 03:58 PM

View PostUnbound Inferno, on 20 July 2013 - 01:17 PM, said:

BattleValue can't really be put into MW:O due to the differences in how each weapon operates and their effective use.

It'll be too big of a problem to balance.


Now, mech worth, C-Bill cost... that has promise.


So, you see no problems with assigning numerical values to different mechs/loadouts and balancing the game by those values as long as there's "c-bill" displayed after the number, yet same exact numbers with "BV" displayed after them "can't really be put into MW:O"? Care to elaborate on your reasoning?

#26 Mystere

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 22,783 posts
  • LocationClassified

Posted 20 July 2013 - 04:02 PM

View PostXX Sulla XX, on 20 July 2013 - 02:28 PM, said:

To clarify I ment the system would use elo + tonnage + cost to balance matches ...


And matchmaking becomes even more complicated than it already is.

Why not just keep things extremely simple and go back to random matchmaking? I had no issue with that, even while being stomped from time to time. It actually helped me want to do better.

Edited by Mystere, 20 July 2013 - 04:03 PM.


#27 Unbound Inferno

    Member

  • PipPipPipPipPipPipPipPip
  • 1,168 posts

Posted 20 July 2013 - 04:03 PM

View PostIceSerpent, on 20 July 2013 - 03:58 PM, said:


So, you see no problems with assigning numerical values to different mechs/loadouts and balancing the game by those values as long as there's "c-bill" displayed after the number, yet same exact numbers with "BV" displayed after them "can't really be put into MW:O"? Care to elaborate on your reasoning?

View PostUnbound Inferno, on 20 July 2013 - 01:32 PM, said:

No. The Battle Value system was based around the inaccurate Battletech game system.

The numbers for what determined BV would be different in MW:O if the same things were taken into consideration.


Thus we need something else.


#28 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 20 July 2013 - 04:16 PM

View PostUnbound Inferno, on 20 July 2013 - 04:03 PM, said:

No. The Battle Value system was based around the inaccurate Battletech game system.

The numbers for what determined BV would be different in MW:O if the same things were taken into consideration.


Thus we need something else.


Nobody (to the best of my knowledge) ever suggested a verbatim copy of TT BV values. When people ask for BV, the idea is to have a number that represents how good this mech/loadout is and use that number in matchmaking.

#29 Unbound Inferno

    Member

  • PipPipPipPipPipPipPipPip
  • 1,168 posts

Posted 20 July 2013 - 04:23 PM

View PostIceSerpent, on 20 July 2013 - 04:16 PM, said:


Nobody (to the best of my knowledge) ever suggested a verbatim copy of TT BV values. When people ask for BV, the idea is to have a number that represents how good this mech/loadout is and use that number in matchmaking.

And that is fine. Just want to make sure PGi doesn't think that if they read this too.

#30 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 20 July 2013 - 04:50 PM

View PostUnbound Inferno, on 20 July 2013 - 04:23 PM, said:

And that is fine. Just want to make sure PGi doesn't think that if they read this too.


I hear you. Unfortunately, PGI tends to do whatever voices tell them to do regardless of what we ask for here.

#31 Gaan Cathal

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 2,108 posts

Posted 20 July 2013 - 09:30 PM

So this would 'weigh less' than this?

#32 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 20 July 2013 - 11:08 PM

No it would take into consideration weight and ELO and mentioned.

#33 EvilCow

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 2,243 posts

Posted 21 July 2013 - 01:37 AM

I like the idea but probably finding a balance between weight and cost would be even better. For example something like:

value = cost * (tons / 50)

Potentially also other elements could be factored in, not just cost and tons.

#34 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 21 July 2013 - 08:13 AM

Yes exactly you would have to come up with a formula that factored in ELO, Cost, Weight and possibly other things like pilot tree etc.

#35 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 21 July 2013 - 09:53 AM

Using cost is problematic because cost increase doesn't necessarily correlate with greater effectiveness of the build, i.e. an Atlas with big XL engine and a couple of LBX-10s would be very expensive and very bad.

#36 Bagheera

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,920 posts
  • LocationStrong and Pretty

Posted 21 July 2013 - 11:11 AM

View PostKhobai, on 20 July 2013 - 01:06 PM, said:

Absolutely horrible idea. Do you really want noobs with LB10Xs and XL engines on their Atlases to drag down the whole team? Just because a component is expensive doesn't mean its good.


This.

At best a case could be made for including a L1 vs L2 tech consideration in matchmaking, but that still doesn't necessarily mean "better." I guess an L1 Tech only queue could be interesting, but there's just not enough players for that sort of thing atm. Its pretty difficult to have a solid, accurate matchmaker with a somewhat small population set from which to choose.

This is just a theory of mine.

#37 wickwire

    Member

  • PipPipPipPipPipPipPip
  • 741 posts
  • LocationIgnoring The Meta Since 2012

Posted 21 July 2013 - 11:24 AM

CBill value divided by tonnage would be a good compromise

#38 Erata

    Member

  • PipPipPipPipPipPip
  • 285 posts
  • LocationGoro Company Dropship MK1, Long live Lord Shang Tsung.

Posted 21 July 2013 - 11:39 AM

View PostXX Sulla XX, on 20 July 2013 - 12:22 PM, said:

Just thinking about a good way to balance teams better. Weight would help. But a stock Atlas is very different that a fully custom Atlas. So yo still end up with the balance being. But if the cost of the mechs including custom items like engines and modules is factored in it makes much more even teams mech wise. Although I would think at the moment there are not enough people playing to do this and not have to wait a long time for games.


Mech cost isn't good enough because, like others have said, expensive weapons that are TERRIBLE in the game right now, like LBXes and to a lesser extent ER LL would throw the metric off entirely.

A better way to weight mechs after tonnage would be by a common build's average damage output. It would put PPCs on top, followed by Medium Lasers.

So there's another reason the point system would appear strange: Medium lasers are one of the best weapons in the game due to flexibility in practice, causing a mech like the HBK 4-P to appear to an automated system to be one of the BEST MECHS GOD HAS EVER KNOWN!

And it isn't. It's subject to a boating nerf that it didn't deserve. Mediums are already largely irrelevant in this game due to their low health vs the firepower heavies and assaults can carry, and now mechs like the 4-P are unable to punch back as fiercely due to heat scale penalties.

edit: There's no real way to weight mech builds unless you actually sat down and began identifying good mechs. Like fighting game tiers, "what's best" would change as time passes, and in a game like this, as weapons are altered and especially when more maps are added to the rotation.

The way tournaments are structured does not help to identify which builds are most-valuable because the builds people use during the tournaments are designed for less alpha-lethality anyway, vs the current meta that still favors alpha damage in spite of the heat penalty.

Finally, the tournaments give bad results because they tend to focus entirely on a single mech instead of normal play, where most players will make their mech as strong as possible.

I'm not sure where PGI is pulling their data from, but they should be looking at 8 v 8 drops to ID the top ELO builds if they want to make the ~best matchmaker ever~. If they ever add a pre-game lobby, then they'll need to look at the system again because people will be able to make specialized teams instead of mechs designed for generalized situations.

Edited by Erata, 21 July 2013 - 11:44 AM.


#39 ArmandTulsen

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 1,184 posts

Posted 22 July 2013 - 01:27 AM

FFS, add tonnage limits.

500 tons.

#40 Purlana

    Member

  • PipPipPipPipPipPipPipPip
  • 1,647 posts

Posted 22 July 2013 - 04:06 AM

You do realize that anything will an XL engine will totally mess up your balancing system right?





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users