TL; DR - MWO would be massively improved by even a rudimentary Battle Value-like system for matchmaking. It would make it easier to have evenly matched teams, and also allow for an easier way to make balance changes to weapons (by changing their numerical ranking, not how they function).
Intoduction:
For those who weren't aware, in table top there is a game concept called battle value. Basically there is an equation that takes into account all the equipment a mech (or tank, or infantry, or fighter, or dropship) has, factors in some other stuff like speed/mobility. A weapon's BV is determined by its damage and its range. There is a discount provided to any weapons added to a mech that are beyond the number of weapons it can fire and remain cool.
Stock CPLT-K2: 1,319 BV
Jagermech JM6-DG (Dual Gauss): 1,661 BV
Already we see in tabletop that the dual gauss 65 ton mech was considered significantly better than the dual PPC 65 ton mech (and that was under tabletop heat and rate of fire rules already)
The system isn't perfect in table top, with a number of designs (notably over-gunned clan mechs) that have a significantly higher BV than they probably should. However it is fine for rough estimates.
How this helps balancing matches and weapons:
The usefulness in balancing random pug games is obvious. Rather than matching weight classes 1 for 1, you try to match the total value of the two teams. This means that if your team has a 20 ton Flea with machine guns, you don't feel like you got screwed when the other team brings a laser Jenner. If your team has a trial Awesome, that is ok because it has a lower BV than a tricked out custom Atlas.
It also allows for easier balancing of weapons. Let's consider gauss rifles:
Right now everyone agrees gauss are better in general than AC20s. They were better in tabletop too, and their BV reflected that. Rather than buffing the AC20 or nerfing the gauss, we can instead merely tweak the BV value that is applied to the weapon. PGI can probably see how often a weapon is being used (and how often they show up on the winning team). If, for example, gauss cats (and gaussaphracts) massively out number the PPC users, just keep tweaking upwards the gauss rating until equilibrium is reached.
Additionally, this allows us to include some options (like double heat sinks and, in tabletop, LBX and Ultra ACs) because you aren't being penalized for bringing weaker loadouts.
But it isn't a perfect system! It has problems!
Well duh, of course it does. I never claimed it was perfect. But have you ever spawned, looked at the chassis on your team, then seen what the enemy had and felt you were at a disadvantage already before a single shot was fired? This would at the very least make that less often, and make the differences between teams smaller.
An imperfect solution to a problem is better than no solution to a problem.
3
Reviving An Old Topic: Battle Value
Started by Kobold, Nov 30 2012 12:17 PM
39 replies to this topic
#2
Posted 30 November 2012 - 12:27 PM
Kobold, on 30 November 2012 - 12:17 PM, said:
TL; DR - MWO would be massively improved by even a rudimentary Battle Value-like system for matchmaking. It would make it easier to have evenly matched teams, and also allow for an easier way to make balance changes to weapons (by changing their numerical ranking, not how they function).
Intoduction:
For those who weren't aware, in table top there is a game concept called battle value. Basically there is an equation that takes into account all the equipment a mech (or tank, or infantry, or fighter, or dropship) has, factors in some other stuff like speed/mobility. A weapon's BV is determined by its damage and its range. There is a discount provided to any weapons added to a mech that are beyond the number of weapons it can fire and remain cool.
Stock CPLT-K2: 1,319 BV
Jagermech JM6-DG (Dual Gauss): 1,661 BV
Already we see in tabletop that the dual gauss 65 ton mech was considered significantly better than the dual PPC 65 ton mech (and that was under tabletop heat and rate of fire rules already)
The system isn't perfect in table top, with a number of designs (notably over-gunned clan mechs) that have a significantly higher BV than they probably should. However it is fine for rough estimates.
How this helps balancing matches and weapons:
The usefulness in balancing random pug games is obvious. Rather than matching weight classes 1 for 1, you try to match the total value of the two teams. This means that if your team has a 20 ton Flea with machine guns, you don't feel like you got screwed when the other team brings a laser Jenner. If your team has a trial Awesome, that is ok because it has a lower BV than a tricked out custom Atlas.
It also allows for easier balancing of weapons. Let's consider gauss rifles:
Right now everyone agrees gauss are better in general than AC20s. They were better in tabletop too, and their BV reflected that. Rather than buffing the AC20 or nerfing the gauss, we can instead merely tweak the BV value that is applied to the weapon. PGI can probably see how often a weapon is being used (and how often they show up on the winning team). If, for example, gauss cats (and gaussaphracts) massively out number the PPC users, just keep tweaking upwards the gauss rating until equilibrium is reached.
Additionally, this allows us to include some options (like double heat sinks and, in tabletop, LBX and Ultra ACs) because you aren't being penalized for bringing weaker loadouts.
But it isn't a perfect system! It has problems!
Well duh, of course it does. I never claimed it was perfect. But have you ever spawned, looked at the chassis on your team, then seen what the enemy had and felt you were at a disadvantage already before a single shot was fired? This would at the very least make that less often, and make the differences between teams smaller.
An imperfect solution to a problem is better than no solution to a problem.
Intoduction:
For those who weren't aware, in table top there is a game concept called battle value. Basically there is an equation that takes into account all the equipment a mech (or tank, or infantry, or fighter, or dropship) has, factors in some other stuff like speed/mobility. A weapon's BV is determined by its damage and its range. There is a discount provided to any weapons added to a mech that are beyond the number of weapons it can fire and remain cool.
Stock CPLT-K2: 1,319 BV
Jagermech JM6-DG (Dual Gauss): 1,661 BV
Already we see in tabletop that the dual gauss 65 ton mech was considered significantly better than the dual PPC 65 ton mech (and that was under tabletop heat and rate of fire rules already)
The system isn't perfect in table top, with a number of designs (notably over-gunned clan mechs) that have a significantly higher BV than they probably should. However it is fine for rough estimates.
How this helps balancing matches and weapons:
The usefulness in balancing random pug games is obvious. Rather than matching weight classes 1 for 1, you try to match the total value of the two teams. This means that if your team has a 20 ton Flea with machine guns, you don't feel like you got screwed when the other team brings a laser Jenner. If your team has a trial Awesome, that is ok because it has a lower BV than a tricked out custom Atlas.
It also allows for easier balancing of weapons. Let's consider gauss rifles:
Right now everyone agrees gauss are better in general than AC20s. They were better in tabletop too, and their BV reflected that. Rather than buffing the AC20 or nerfing the gauss, we can instead merely tweak the BV value that is applied to the weapon. PGI can probably see how often a weapon is being used (and how often they show up on the winning team). If, for example, gauss cats (and gaussaphracts) massively out number the PPC users, just keep tweaking upwards the gauss rating until equilibrium is reached.
Additionally, this allows us to include some options (like double heat sinks and, in tabletop, LBX and Ultra ACs) because you aren't being penalized for bringing weaker loadouts.
But it isn't a perfect system! It has problems!
Well duh, of course it does. I never claimed it was perfect. But have you ever spawned, looked at the chassis on your team, then seen what the enemy had and felt you were at a disadvantage already before a single shot was fired? This would at the very least make that less often, and make the differences between teams smaller.
An imperfect solution to a problem is better than no solution to a problem.
Once ELO matchmaking/ pilot rankings make it in, We'll probably see BV of some sort come into play. BV 2.0 actually balances decently with the current ERA we have... so it would work out quite well.
#3
-
-
- Bad Company
- 925 posts
Member
- LocationMyrror
Posted 30 November 2012 - 12:50 PM
I wholeheartedly agree that BV is the most logical way to make matches. It isn't an easy quick-fix type solution, so I don't think we'll be seeing it anytime soon, but it would be nice to know PGI was looking that way for the future.
#4
-
-
- The Nimble
- 5,136 posts
Member
- LocationPlymouth, MN
Posted 30 November 2012 - 12:52 PM
YES PLEASE!
In addition if your Mech's BV is calculated once you have saved it in the mechlab you can then use such things for other matchmaking such as Premade vrs Premade. In this manner you can actually drop with mismatched team size and still have "fair" battles. Say you can choose to run a BV match of 5,500 to 6,000. This could all be taken up by 4 mechs, or spread across 8 and technically should be a fair match. It works well in TT and I think the system would work well here.
In addition if your Mech's BV is calculated once you have saved it in the mechlab you can then use such things for other matchmaking such as Premade vrs Premade. In this manner you can actually drop with mismatched team size and still have "fair" battles. Say you can choose to run a BV match of 5,500 to 6,000. This could all be taken up by 4 mechs, or spread across 8 and technically should be a fair match. It works well in TT and I think the system would work well here.
#5
-
-
- Elite Founder
- 2,930 posts
Member
- LocationChicago, IL
Posted 30 November 2012 - 12:54 PM
Warrax the Chaos Warrior, on 30 November 2012 - 12:50 PM, said:
I wholeheartedly agree that BV is the most logical way to make matches. It isn't an easy quick-fix type solution, so I don't think we'll be seeing it anytime soon, but it would be nice to know PGI was looking that way for the future.
I know that months and months ago Paul mentioned they were looking into it, but it has been disappointingly absent from all their most recent public statements about their plans for matchmaking and community warfare.
I hope they haven't given up on it, in large part because it has the trickle down effect of forcing them to gimp things like double heat sinks and gauss rifles simply because they are better than other options, rather than just giving those things the higher BV they deserve.
#6
-
-
- Legendary Founder
- 1,341 posts
Member
- LocationAstral Space
Posted 30 November 2012 - 12:55 PM
I think with selective equipment placings something like this is going to be needed, just to avoid ECM/non-ECM matchups.
It does have excellent elements to it, and some rebalancing would be needed to what MWO has changed from tabletop.
It does have excellent elements to it, and some rebalancing would be needed to what MWO has changed from tabletop.
#7
-
-
- The Nimble
- 5,136 posts
Member
- LocationPlymouth, MN
Posted 30 November 2012 - 12:57 PM
Kobold, on 30 November 2012 - 12:54 PM, said:
I know that months and months ago Paul mentioned they were looking into it, but it has been disappointingly absent from all their most recent public statements about their plans for matchmaking and community warfare.
I hope they haven't given up on it, in large part because it has the trickle down effect of forcing them to gimp things like double heat sinks and gauss rifles simply because they are better than other options, rather than just giving those things the higher BV they deserve.
"He has 3 Gauss Rifles on that thing!?!"
"Yeah but his team had to bring two stock Hunchbacks instead of two other Heavies to make up for the BV."
#8
-
-
- The Messenger
- 1,215 posts
Member
Posted 30 November 2012 - 01:00 PM
Jade Kitsune, on 30 November 2012 - 12:27 PM, said:
Once ELO matchmaking/ pilot rankings make it in, We'll probably see BV of some sort come into play. BV 2.0 actually balances decently with the current ERA we have... so it would work out quite well.
Agreed, BV 2.0 is the way to go.
#9
-
-
- Elite Founder
- 2,930 posts
Member
- LocationChicago, IL
Posted 30 November 2012 - 01:02 PM
Obviously there will need to be significant tweaking and trial and error.
#11
-
-
- Elite Founder
- 2,930 posts
Member
- LocationChicago, IL
Posted 30 November 2012 - 01:08 PM
Taryys, on 30 November 2012 - 01:02 PM, said:
They are looking into it.
I have suggested a combo of stat and BV based match making in my
thread-which-shall-not-be-named which is my sig.
I have suggested a combo of stat and BV based match making in my
thread-which-shall-not-be-named which is my sig.
Given the slew of "balancing" threads that are around (DHS, streaks, gauss, and ultimately ECM as well) I figured it was worth bringing BV back to the forefront. Many of the current players weren't necessarily around back when these topics were first being tossed around.
Also I agree that there should ideally be a blended system. Imagine two teams with identical mechs, and identical players, but team A has their best players in their best mechs, and team B has their worst players in their best mechs, while the best players are in the worst mechs. Team A likely has an advantage.
The tabletop BV had the skill level of the pilots be a multiplier on top of the BV of the unit they are driving. You'd probably want a similar thing here for PUG matchmaking (but not for organized competitive play).
#12
-
-
- The Nimble
- 5,136 posts
Member
- LocationPlymouth, MN
Posted 30 November 2012 - 01:34 PM
I guess I can only post to further agree with you Kobold. A great deal of current balance issues could be addressed with BV aiding in matchmaking.
The multiple comments about people hating dropping with a specific Trial mech on their side assuming there will be a tweaked out build that it is matched up with by the existing system really show why this system would be good for the game. If a Trial Mech was dropped on your side in a random match-up it would mean that a similarly weighted combination of mechs would end up on the other side so your teammate would no longer be giving you a perceived disadvantage. Having it modified by player stats would go a long way towards balancing our matches despite which mech we chose to launch in or our level of expertise. This would make the community more inclusive and less exclusive.
The multiple comments about people hating dropping with a specific Trial mech on their side assuming there will be a tweaked out build that it is matched up with by the existing system really show why this system would be good for the game. If a Trial Mech was dropped on your side in a random match-up it would mean that a similarly weighted combination of mechs would end up on the other side so your teammate would no longer be giving you a perceived disadvantage. Having it modified by player stats would go a long way towards balancing our matches despite which mech we chose to launch in or our level of expertise. This would make the community more inclusive and less exclusive.
#13
-
-
- Knight Errant
- 524 posts
Member
Posted 30 November 2012 - 04:36 PM
I think since you're not restricted to certain variants in MWO, you can pretty much work with tonnage as a base value.
Everyone can optimize his mech. You're not stuck with the one you randomly got, therefore the system doesn't have to compensate for better or worse variants.
The only things that really make a difference are tech level and upgrades. Mainly upgrades.
So you could simply modify the tonnage with a certain multiplier for ES, FF, DHS, Artemis, XL engine and/or being a Clan mech. Star League equipment would be optional in this calculation, depending on whether or not those weapons are superior or balanced.
Example:
Hunchback 4G: 50
Hunchback 4P: 50
Hunchback 4P +ES +DHS: 50 *1.1 *1.1 = 60.5
Jenner 7D: 35
Jenner 7D +ES +FF +DHS +XL: 35 *1.1 *1.1 *1.1 *1.1 = 51.2
Those numbers are just examples. For simplicity I chose a 10% modifier for every upgrade.
-edit-
Trial mechs on the other hand.. could be matched against trial mechs only, same mechs on both sides. Maybe boring, but ultimately fair
Apart from that, of course a proper BV would be the best solution. It's just not easy to get it right and definitely a lot more work. A flawed BV wouldn't be better than simple tonnage.
Everyone can optimize his mech. You're not stuck with the one you randomly got, therefore the system doesn't have to compensate for better or worse variants.
The only things that really make a difference are tech level and upgrades. Mainly upgrades.
So you could simply modify the tonnage with a certain multiplier for ES, FF, DHS, Artemis, XL engine and/or being a Clan mech. Star League equipment would be optional in this calculation, depending on whether or not those weapons are superior or balanced.
Example:
Hunchback 4G: 50
Hunchback 4P: 50
Hunchback 4P +ES +DHS: 50 *1.1 *1.1 = 60.5
Jenner 7D: 35
Jenner 7D +ES +FF +DHS +XL: 35 *1.1 *1.1 *1.1 *1.1 = 51.2
Those numbers are just examples. For simplicity I chose a 10% modifier for every upgrade.
-edit-
Trial mechs on the other hand.. could be matched against trial mechs only, same mechs on both sides. Maybe boring, but ultimately fair
Apart from that, of course a proper BV would be the best solution. It's just not easy to get it right and definitely a lot more work. A flawed BV wouldn't be better than simple tonnage.
Edited by John Norad, 30 November 2012 - 04:45 PM.
#14
-
-
- 602 posts
Member
- LocationSF Bay Area, California
Posted 18 December 2012 - 08:59 AM
Instead of creating my own thread on this topic, I found this thread that had been well written and constructed already.
I wanted to bring this discussion back to the surface. Thoughts gentlemen?
I wanted to bring this discussion back to the surface. Thoughts gentlemen?
#15
-
-
- 727 posts
Member
Posted 18 December 2012 - 09:08 AM
Awsome idea, can't fault this would make the matches a lot better. Would be nice as well in the future for the premades to see their sides total BV as well and allow them to build setups to fit into certain tiers of match kinda like how boxing is divided into weight categories i guess.
#16
-
-
- 602 posts
Member
- LocationSF Bay Area, California
Posted 18 December 2012 - 09:09 AM
Class matching is a start, but very flawed. Team weight balancing is on the right track. But BV team balancing is ultimately the best idea. It's a proven system.
#18
-
-
- The Nimble
- 5,136 posts
Member
- LocationPlymouth, MN
Posted 18 December 2012 - 09:13 AM
I'm still for it. BV is the way you balance things in TT. It just is. You can use economy to help drive that balance but in the end the GM should still be pulling out his tables and charts and figuring out if the fight is a good challenge for his players or not.
Obviously it will need tweaks but a Jenner with 2 flamers and no other weapons or upgrades is not equal to a CMD-2D with ECM, 3 Streaks and a MLas. Right now the Matchmaker things they are.
Obviously it will need tweaks but a Jenner with 2 flamers and no other weapons or upgrades is not equal to a CMD-2D with ECM, 3 Streaks and a MLas.
Right now the Matchmaker things they are.
#19
-
-
- Bad Company
- 46 posts
Member
Posted 18 December 2012 - 09:14 AM
this would be much more fair, hope they move to it soon
#20
-
-
- Knight Errant
- 259 posts
Member
Posted 18 December 2012 - 09:21 AM
You need a bigger pool of players to pull off BV matchmaking, or any kind of matchmaking for that matter.
MWO is losing players with each day. I don't see this happening in any future.
MWO is losing players with each day. I don't see this happening in any future.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users
