33 replies to this topic

[DjPush]

What is battle value and how is it used to balance a game?

Edited by DjPush, 13 July 2015 - 04:12 PM.

[Burktross]

When in tabletop, units-- more specifically components-- have battlevalues denoting their power. For instance, a locust has a BV of 432, but its 20t clan brother, the Phirana has a BV of 998. This is due to superior clantech equipment. Equipment upgrades in general, such as upgrades like endo/DHS raise BV. Bigger mechs also have bigger BV. So in essence, BV makes it such that you have many small, low power units, or few high power units.

Rule of thumb is that clanner mechs have higher BV than their equal weighted counterparts-- thus the IS was able to field a number advantage to counter the clan power advantage.

We don't have that here, so clantech and IStech must be balanced

[MischiefSC]

Battle Value is a largely arbitrary point system where point values are assigned to components and mechs. 1,000 pts of one this is worth 1,000 pts of another. So a 1,000 pt mech that's 50 tons but endo and good weapons is equal to a 75 ton mech with crappy gear.

In reality it doesn't really work. There is no actual way to actually objectively assign values to things in this sort of environment. Synergy, positioning, the fact that an XL may be brilliant in one build and suicide in another, that on a Spider FF is worthwhile while on an Atlas it's stupid, etc. etc.

It's like the idea of utopia. A great idea but it doesn't actually work once you involve people.

[Roadkill]

Think of it as a rating system for Mechs, sort of like Elo is a rating system for players. Except that BV is based on weapons, hardpoint locations, etc, instead of performance in battle.

What it sounds like PGI is trying to do is assign a point value to various weapons, high mounted hardpoints, perks, etc. They'll then aim for a particular value when balancing a Mech under the theory that two Mechs with the same BV are equally good.

I've yet to see a BV system that's accurate for BattleTech, and they've tried like 3 times that I can think of. So... we'll see.

[Night Thastus]

Even if they implemented a BV system, how would it be used for balance? Would matches have unequal number of players on each side, but roughly equal BV? Would they have the same number of players (like 12v12) but the BV's would still be roughly equal?

How would this work in game? Would a 'mech have a BV cap? Would teams have a BV cap, like dropships have a tonnage cap?

[MechaBattler]

Night Thastus, on 13 July 2015 - 04:20 PM, said:

Even if they implemented a BV system, how would it be used for balance? Would matches have unequal number of players on each side, but roughly equal BV? Would they have the same number of players (like 12v12) but the BV's would still be roughly equal?

How would this work in game? Would a 'mech have a BV cap? Would teams have a BV cap, like dropships have a tonnage cap?

I don't think it's a matchmaker tool. It's more a way of determining the base value of a mech and then balancing from there. That's the impression I got from the townhall meeting. They're not using it the way the table top version does.

[Alistair Winter]

From what I've read and gathered from the scraps of information scattered across the interwebs, it seems like PGI's final solution to the balance problem is to assign every mech a value from 'terrible' to 'great'. So if you run a 'terrible' mech, matchmaker will balance this by putting a 'terrible' mech on the enemy team. And if you run a 'great' mech, MM will put a 'great' mech on the enemy team. And I guess Elo will be factored in somehow.

So essentially, the Locust won't be as good as the Firestarter. But you won't have 4 Locusts vs 4 Firestarters, for example. You'll have 2 Locusts and 2 Firestarters vs 2 Locusts and 2 Firestarters.

Even if we assume that PGI will nail the BV and find a system that accurately predicts the actual value of every mech and every build... it still begs the question of why anyone will want to play the 'terrible' mechs. Because you can still end up in a match where 11 of the 12 enemy mechs are better than yours. And then what do you do? Go looking for the one mech equally 'terrible' as yours?

I don't get it. Maybe I've misunderstood their idea.

MischiefSC, on 13 July 2015 - 04:15 PM, said:

It's like the idea of utopia. A great idea but it doesn't actually work once you involve people.

I don't think you're saying what you think you're saying. But I get what you're saying.

[DjPush]

MechaBattler, on 13 July 2015 - 04:26 PM, said:

I don't think it's a matchmaker tool. It's more a way of determining the base value of a mech and then balancing from there. That's the impression I got from the townhall meeting. They're not using it the way the table top version does.

Do you think they are going to put a "battle value" cap on a mech limiting the load out of a mech?

[Alistair Winter]

MechaBattler, on 13 July 2015 - 04:26 PM, said:

I don't think it's a matchmaker tool. It's more a way of determining the base value of a mech and then balancing from there. That's the impression I got from the townhall meeting. They're not using it the way the table top version does.

So it's a system to determine which mechs are Tier 1, Tier 2, etc? And then assign quirks accordingly?

I don't see how this is a huge improvement from what they had already. It's basically the same as before, but now they've got some algorithm doing the balancing instead of a person looking at statistics and feedback from the players. Which, unless they've got Hari Seldon working for them, probably won't work out very well.

[Escef]

Burktross, on 13 July 2015 - 04:15 PM, said:

Equipment upgrades in general, such as upgrades like endo/DHS raise BV. Bigger mechs also have bigger BV. So in essence, BV makes it such that you have many small, low power units, or few high power units.

This is a little misleading. Endo Steel, in and of itself, does not raise BV. However, the extra armor, heat dissipation, weapons, and/or miscellaneous equipment it frees up tonnage for will increase BV. Bigger mechs usually have larger BVs, but not always, it depends upon the build. For example, the stock Tallman Thunderbolt, aka Top Dog, has a higher BV than many 70 and 75 ton mechs of comparable technological advancement. The Jagermech JM6-S has the lowest BV of any canon 3039 era heavy mech, 901; which is topped by the JR7-F Jenner's BV of 1011.

Edited by Escef, 13 July 2015 - 04:35 PM.

[MischiefSC]

Alistair Winter, on 13 July 2015 - 04:31 PM, said:

I don't think you're saying what you think you're saying. But I get what you're saying.

No, that says what I wanted to say. BV is a great concept. It works great on paper. It works terribly when you actually have people try to execute it.

[MechaBattler]

Alistair Winter, on 13 July 2015 - 04:33 PM, said:

So it's a system to determine which mechs are Tier 1, Tier 2, etc? And then assign quirks accordingly?

I don't see how this is a huge improvement from what they had already. It's basically the same as before, but now they've got some algorithm doing the balancing instead of a person looking at statistics and feedback from the players. Which, unless they've got Hari Seldon working for them, probably won't work out very well.

We'll just have to wait and see.

DjPush, on 13 July 2015 - 04:32 PM, said:

Do you think they are going to put a "battle value" cap on a mech limiting the load out of a mech?

They haven't really implied anything like that. It would be interesting. But I think a lot of people would cry that it's another restriction to their customization freedom.

[Alistair Winter]

MischiefSC, on 13 July 2015 - 04:34 PM, said:

No, that says what I wanted to say. BV is a great concept. It works great on paper. It works terribly when you actually have people try to execute it.

Yeah, I got that part. It was more about you comparing it to utopia

[Throat Punch]

MischiefSC, on 13 July 2015 - 04:34 PM, said:

No, that says what I wanted to say. BV is a great concept. It works great on paper. It works terribly when you actually have people try to execute it.

Unfortunately BV doesn't even really work good on paper (TT) that's why they revamped it from Combat Value to Battle value to Battle value 2.0, to a proposed in house BV3 to scrapping that and just adding to BV 2 to create BV 2.1.

[SilentSooYun]

MischiefSC, on 13 July 2015 - 04:34 PM, said:

No, that says what I wanted to say. BV is a great concept. It works great on paper. It works terribly when you actually have people try to execute it.

It gets worse in a real-time sim because you have dozens of intangible factors at work that never occur on paper, like weapon positions, projectile velocities, recharge times, hit boxes, etc. And no BV system yet takes into account pilot skill and experience.

[Ovion]

MischiefSC, on 13 July 2015 - 04:15 PM, said:

Battle Value is a largely arbitrary point system where point values are assigned to components and mechs. 1,000 pts of one this is worth 1,000 pts of another. So a 1,000 pt mech that's 50 tons but endo and good weapons is equal to a 75 ton mech with crappy gear.

In reality it doesn't really work. There is no actual way to actually objectively assign values to things in this sort of environment. Synergy, positioning, the fact that an XL may be brilliant in one build and suicide in another, that on a Spider FF is worthwhile while on an Atlas it's stupid, etc. etc.

It's like the idea of utopia. A great idea but it doesn't actually work once you involve people.

Give me a few days, and I could do a coherent, comprehensive and effective pointing system taking these things into account and more.
Including placement, XL friendly or not, potentially allowing for pilot skill.
It'll be incredibly complex though

Of course that's a lot of work, and I'm not being paid, so~
Maybe I will, maybe I won't.

Edited by Ovion, 13 July 2015 - 05:09 PM.

[Jaeger Gonzo]

BV while maybe not perfect and need some update it works in TT.
Don`t see why something similar could not in MWO.
You got your elo now right? So Elo x Mek bv= final your personal bv in this particular mek.
Or lets say in CW we can just don`t bother at all with elo and pilot skill level, as we don`t now.
Nothing more easy.

[TELEFORCE]

It's a shame the BattleTech developers haven't spent more time on tweaking BV, as MegaMek is really useful for testing changes in the BV system.

For example, under the current BV2.0 system for BattleTech, 'mechs that carry weapons without the heat sinks to effectively fire them all are over-BV'd. Also, 'mechs that have both MASC and jump jets are also over-BV'd as they are treated as being able to use both systems at once, which isn't possible with the table top rules.

[Wintersdark]

MischiefSC, on 13 July 2015 - 04:15 PM, said:

Battle Value is a largely arbitrary point system where point values are assigned to components and mechs. 1,000 pts of one this is worth 1,000 pts of another. So a 1,000 pt mech that's 50 tons but endo and good weapons is equal to a 75 ton mech with crappy gear.

In reality it doesn't really work. There is no actual way to actually objectively assign values to things in this sort of environment. Synergy, positioning, the fact that an XL may be brilliant in one build and suicide in another, that on a Spider FF is worthwhile while on an Atlas it's stupid, etc. etc.

It's like the idea of utopia. A great idea but it doesn't actually work once you involve people.

You've done a really good job of succinctly stating what I've been harping on about in huge walls of text for ages, I tip my hat to you good sir.

It's long frustrated me, because people keep going on that Battlevalue would Solve All The Game's Balancing Woes, and always with some handwavium like "They just need to simply assign values to all the equipment, add it up, add in Elo, and BAM! Done!"

Because "coming up values" for disparate items is hard/impossible. How do you assign a value to a Seismic Sensor as compared to a PPC; do you consider the value of a PPC in a Blackjack's upper mounts vs. one in a Commando's arm? What if a mech has PPC quirks? What happens if PGI decides to increase the velocity of a PPC bolt by 100m/s? The answer is always handwaving, just "make a formula that takes all that into account". *twitch*


But really, this is all moot.

The "Battlevalue" system Russ was talking about was an entirely internal system to rate mechs for comparison; essentially what may as well be tiers. It's not going to be used in matchmaking at all - and that's good. If players figured out what the values where for equipment, they'd exploit the ever-living crap out of it.

And look, I went all off on this yet again. *sighs*

[Zordicron]

The TT system on all accounts had big flaws, as many have already pointed out.

IMO, a system BASED on the BV system could have some potential, but would require some real hashing out. For instance, each weight class would need it's own ranking system, where things like Endo and FF would have a higher value on lights then say, on assault class mechs, due to the nature of loadouts for them.

Thats just a very basic example. Really, a system could be made to work much better then our current 3-3-3-3 silly system. Making a MWO specific BV system would take DAYS of dev time though, prolly WEEKS. Trying to hash through all the nuances of the mechs in game and configure a system to accomodate them, PLUS then try to make it future proof would be a daunting task.


Overall, it is likely not worth it, unless ELO was simultaneously 100% overhauled in function and purpose. One thing MWO has: the pilot makes a lot more of a difference then the mech. N00bcad3t running ubermeta TBR is still ganked by a good pilot in a Spider 5V. And that, that right there is why a BV system can't be a stand alone system, OR even the bigger part of the system. At best, a class specific BV system would simply be a modifier to the "average team ELO" based on one mech being summarily inferior to another, like a Spider 5v vs meta FS9 or a Huggin.

I wouldnt bother with a BV honestly. not until we see how Russ' ideas for an ELO/matchmaker overhaul work out. Maybe, maybe, if it turns out like he wants a BV system could be the final tweak to ensuring a match is as fair a fight as can be, at least at the start.

Edited by Eldagore, 13 July 2015 - 06:22 PM.