A Dynamic Self-Balancing Battlevalue System
#21
Posted 05 March 2014 - 05:35 PM
For example im very good with srm (even with hit detection being wonky) Thusly I could probly be in an unfavored mech and as long as it has missle slots, I would probly do very well. Prefferably missle slots and some jump jets.
I simply dont see a reason not to include this no matter how you work it. Weapon systems and basically players ratings with them need to be included in the calculation.
#22
Posted 05 March 2014 - 05:55 PM
Varent, on 05 March 2014 - 05:35 PM, said:
I see nothing wrong with what you describe here.
If you are able to take a mech that is rarely used by the population at large, and make good use of it.. that's awesome. That's exactly the kind of thing that we want to encourage.
#23
Posted 05 March 2014 - 06:17 PM
If you can't set up a team 5-11 for example, then do you expect PGI to even think about putting into
practice virtually any of these intelligent ideas?
They will just simply money wise find this a very hard pill to swallow. It is not going to happen.
This Is in no way meant to take away from all your good ideas and hard work but is not a board game
and even if it makes it. Board games such as this will not fly in the online world. The closest we have ever
gotten or will ever get is Battletech 3025. It was balanced, had a universe that everyone could play, It also
made you want to play it. Also, even in early beta it was in overall game play far ahead of what we have now.
That was over twelve years ago. PGI learned nothing from what worked. Microsoft was so afraid of it that
they killed it knowing that MW4 would be killed if it went head to head against it. They made EA an offer
they could not refuse.
Wish I was wrong.
#24
Posted 05 March 2014 - 06:34 PM
shotokan5, on 05 March 2014 - 06:17 PM, said:
If you can't set up a team 5-11 for example, then do you expect PGI to even think about putting into
practice virtually any of these intelligent ideas?
Well, one of the reasons why I bothered to put together this thread was specifically because I was so deeply disappointed in their decision to reverse their plans to enable larger groups once again.
My hope is that by presenting a system like this, which could be quickly and easily implemented, it may provide them with a solution to some of the problems that they cited as reasons for not allowing larger groups.
I figured that rather than just cursing them (which was my first response), that if I provided a concrete solution in detail, that it might actually result in them changing their minds.
Quote
I don't see how this would be detrimental to them at all, from a monetary point of view.
There are actually a few reasons why this system would actually benefit PGI's bottom line.
Quote
and even if it makes it. Board games such as this will not fly in the online world. The closest we have ever
gotten or will ever get is Battletech 3025. It was balanced, had a universe that everyone could play, It also
made you want to play it. Also, even in early beta it was in overall game play far ahead of what we have now.
Well, bear in mind here that I'm actually NOT a huge Battletech fan. I have only a passing experience with the game. My experience in this universe is almost entirely within Mechwarrior.
I like the shooter aspects of mechwarrior. I don't really want to play the TT game, nor do I want this game turned into a TT game.
While the term Battlevalue comes from the TT game, my implementation of it here is far different from the TT system. Instead of a static set of values, the solution I present here is designed to create a balancing system which works within the dynamic and fluid environment of an online competitive combat game.
Don't get me wrong, I feel your pain, and I'm pretty bitter about a lot of things going on in MWO.
But damned if I can't just give up all hope. Some part of me just keeps believing that no matter how bad things get, this game still has the potential to be what we believed it could be back when we were in beta.
The game still has potential to be one of the greatest ever... And I know I'm a sucker for clinging to hope, but I keep hanging on.
#25
Posted 05 March 2014 - 06:53 PM
- How does your system account for 'mechs that experience incidental surges in popularity? For example, Trial 'Mech variants always see fairly heavy use. So do newly released 'mechs and Hero variants. These aren't always the most effective variants from a pure BV standpoint, but they do see popularity surges.
- Does your system in any way address the difference between the heavily modified "meta" variants and their often much worse stock base variants? For example, a JM6-S fully upgraded with DHS, a pair of AC/20s and an engine swap is a fearsome opponent, but the base stock variant is a DPS plinker that lacks adequate armor or ammo to perform its intended job. Likewise, the SHD-2H is very popular at the moment, but rarely seen in its unmodified, light-on-firepower-heavy-on-ammo stock configuration.
#26
Posted 05 March 2014 - 07:16 PM
Solis Obscuri, on 05 March 2014 - 06:53 PM, said:
- How does your system account for 'mechs that experience incidental surges in popularity? For example, Trial 'Mech variants always see fairly heavy use. So do newly released 'mechs and Hero variants. These aren't always the most effective variants from a pure BV standpoint, but they do see popularity surges.
Well, the fact that BV is being calculated with in a sliding window will prevent "flash in the pan" incidents from having much impact. For instance, the weekend of Mediums vs the World, probably had more mediums that normal.
However, since BV would always be getting calculated using a whole month of data, that weekend's impact would be somewhat muted. Immediately after the event, for instance, it'll have 3 days of heavy medium usage.. but it'll have 27 days of "normal" medium usage. So, for the next month, while that incident is in the data, mediums may have a slightly higher price than normal, but nothing huge, because it would be mixed with the normal data.
The only way that prices would move signficantly, would be when the playerbase actually modifies its habits in a more focused, sustained manner.This will tend to take place over periods of a few weeks, I suspect, just due to the inertia of having a large dataset being calculated over, combined with the sliding window.
Now, for trial mechs, it's possible that their prices may be slightly inflated during the entire duration of their stay.. but I don't really suspect that their values would actually be that high anyway, given that most of the playerbase moves away from them once they've played a while... So even with an inflated value, they'd still likely be fairly cheap in terms of BV.
For that particular case though, you could deal with them as an edge case and handle them uniquely.
For instance, you could give them a BV of zero, and not track drops of trial mechs, thus they wouldn't impact the BV calculations of non-trial versions of the same variant. Extracting them from the data set would be pretty trivial, I suspect.
I probably wouldn't worry about it unless running the data actually demonstrated an issue that needed to be dealt with though.
Solis Obscuri, on 05 March 2014 - 06:53 PM, said:
- Does your system in any way address the difference between the heavily modified "meta" variants and their often much worse stock base variants? For example, a JM6-S fully upgraded with DHS, a pair of AC/20s and an engine swap is a fearsome opponent, but the base stock variant is a DPS plinker that lacks adequate armor or ammo to perform its intended job. Likewise, the SHD-2H is very popular at the moment, but rarely seen in its unmodified, light-on-firepower-heavy-on-ammo stock configuration.
No, as I've pointed out to others, the system doesn't care what you do with a mech. You don't get any bonus points for running a "bad" build. It's assumed that anyone who runs a variant will be doing so with the intent of being as effective as possible.
Different people may pursue that effectiveness in different ways, and frankly, the system doesn't care.
It merely says, "This guy thinks this variant is effective! Account for that when we calculate its BV!"
#27
Posted 06 March 2014 - 05:19 AM
In the Tabletop it was the same - first they tried it with a simple tonnage limit - they saw it didm`t work and after several years the BAttlevalue System was born. BV + ELO will work , ELO alone won`t.
Just for Reference :
http://www.sarna.net/wiki/Battle_Value
Regards,
Plizzken
#28
Posted 06 March 2014 - 05:43 AM
I like the self-adjusting "BV"-mechanism for anything we field, sounds simple, yet has the needed complexity to allow for most (if not all) circumstances this game can provide so far .
What was that old saying: You don´t need to create something overly complex if simpler would do also .
Your system has the definite advantage over straight tonnage restrictions due to it´s "BVs" being permanently derived from the actual fieldings of Mechvariants without any need of implementig some database with fixed BV values for each piece of kit which then would have to be permanently maintained and balanced against one another as well to account for "Metagaming" and it´s changes ( and we all fully know how "quick" balancing is being done around here, no insult PGI, you just have a vision which we don´t and never got talked through, so how should we understand ? ).
Now with this said, how can we talk PGI into letting us try this ?
I know one way, but that includes sh!tstorms, which is not the way of changing things for good .
Instead I will vote and spread word .
Please do the same !
Thanks to Roland !
Edited by Rad Hanzo, 06 March 2014 - 05:44 AM.
#29
Posted 06 March 2014 - 07:18 AM
Thanks, Roland!
#30
Posted 06 March 2014 - 08:48 AM
Traditional Battlevalue systems just won't work for us at all, because while the concept is great, getting that dataset in our constantly changing game ecosystem is far harder than people like to think. And then on top of that, PGI would need to devote dev time to monitoring and maintaining/balancing the BV system forever - and they already struggle to keep on top of weapon balance.
Your system totally sidesteps that problem, the huge initial design cost in coming up with a data set and the developer time maintaining it going forward, as it's entirely self contained and self adjusting.
Applying it to groups, to ensure a certain limit of "meta crush" per group is pure, unfettered awesome. It works on so many levels - keeping group/solo mixed queues (should such a thing happen) a little less hardcore, getting low-usage mechs (or even weapons/builds/etc) more play time increasing diversity, generating a self-retarding aspect to the overall meta game that works even as the current meta changes.
Seriously, this idea is just genius. There are bad ideas, good ideas, and neat ideas posted here all the time, but there are precious few that are as intriguing as this one.
Edited by Wintersdark, 06 March 2014 - 08:48 AM.
#31
Posted 06 March 2014 - 09:50 AM
Roland- I really like everything I see here.
But how would you answer the question raised upthread about Hero mechs? Being 'cash only' they will have artificially lower drop rates.
#32
Posted 06 March 2014 - 09:57 AM
for the sake of argument, lets assume the recent jump jet changes had not been put in. Since im sure there will be a dominant meta upcoming. (just a fact) and people will want to use certain weapon systems (like ppc, ac etc) on some mech chasis.
The 733c is a very dominant chasis. Not so much anymore with the current changes. You would see pretty much just that chasis out there in jump sniping. With this system it would be counted (higher) and therefore limit the amount of them you would see in game. Thusly you could just use a different highlander chasis, mount almost the same weapons on it and run it till the bracket raised, and then switch to another highlander chasis....etc...etc..... or a victor chasis... or a cata chasis... and by the time that one raised switch back yet again to the top of the list.
Or ya know... you could just track the weapons as well and make this system complete. Track the players and the stats they have with those weapon systems and frequency of use with them and include that in the overall calculations.
In other words if player a always uses ppc and has a good ratio with them, Track that and carry on that statistic overall for there gaming, so as they continue to use those weapons it raises there numbers no matter what mech they are using as long as they are using them on a mech and continues to rise as they keep winning in that mech with those weapons etc.
The only reason I can think to NOT include this is if you want to abuse the system yourself? I mean really now...
#33
Posted 06 March 2014 - 10:06 AM
Varent, on 06 March 2014 - 09:57 AM, said:
for the sake of argument, lets assume the recent jump jet changes had not been put in. Since im sure there will be a dominant meta upcoming. (just a fact) and people will want to use certain weapon systems (like ppc, ac etc) on some mech chasis.
The 733c is a very dominant chasis. Not so much anymore with the current changes. You would see pretty much just that chasis out there in jump sniping. With this system it would be counted (higher) and therefore limit the amount of them you would see in game. Thusly you could just use a different highlander chasis, mount almost the same weapons on it and run it till the bracket raised, and then switch to another highlander chasis....etc...etc..... or a victor chasis... or a cata chasis... and by the time that one raised switch back yet again to the top of the list.
Or ya know... you could just track the weapons as well and make this system complete. Track the players and the stats they have with those weapon systems and frequency of use with them and include that in the overall calculations.
In other words if player a always uses ppc and has a good ratio with them, Track that and carry on that statistic overall for there gaming, so as they continue to use those weapons it raises there numbers no matter what mech they are using as long as they are using them on a mech and continues to rise as they keep winning in that mech with those weapons etc.
The only reason I can think to NOT include this is if you want to abuse the system yourself? I mean really now...
Wouldn't this mean that medium lasers would be the most valued of all weapons?
If other Highlanders can do what the 733c can, then why do we see the 733c used so much? (Honest question here, not trolling)
#34
Posted 06 March 2014 - 10:33 AM
Davers, on 06 March 2014 - 09:50 AM, said:
Yes, they potentially would have artificially lower drop rates. Although, in reality, there's nothing "artificial" about their frequency being lower... It's actually a correct interpretation of the data. Those mech are showing up less frequently, because they are less desirable. It's just that their desirability is being influenced by a less traditional factor, in that they require real-world money to use.
As such, the system may value them lower than they would be valued if they didn't require MC to purchase. But at the same time, they would be showing up less frequently than other mechs anyway, so the impact would be somewhat minimal. Having a lower cost wouldn't really mean as much for them.
The normal impact of a mech having a lower cost, would be that its frequency would rise. While the BV for the hero mech is lower than it "should" be due to external factors, the increase in frequency would also be dampened by exactly those same factors. As such, I don't expect this to really have a significant impact on the overall game Also, let's be honest.. most of the hero mechs aren't really that great anyway.
However, like I suggested with trial mechs, if hero mechs proved to be a really destabalizing force, we could choose to deal with them specifically as an edge case. But I suspect that they constitute a small enough percentage of the overall population of mech drops that they wouldn't really need to be worried about.
Quote
This came up while I was discussing this idea with some colleagues, and it's why we introduced the idea of the sliding window. The sliding window will have the effect of smoothing price changes, to the extent that there won't really be the ability to manipulate the prices by jumping back and forth between different chassis.
The exploitation technique you describe here would only work in an environment where you only have those small set of chassis which all do exactly the same thing, and they made up the entire population of variants.
The reality is, even when you take ALL the Highlander variants, you're still only talking about 5 mechs... out of a total set of over 120. What's more, what you describe presupposes a singular market entity which is directing ALL mech purchases. Effectively, you are attempting a manipulation of the market.
But in the community we have, you can't really do that. Even if you got every person you know in the game, you would only constitute a trivial percentage of the full set of transactions taking place in the game. Thus, even if you guys all said, "Ok, we're ALL going to drive 732's this month!" the rest of the community would be doing their own thing. Some of them would take 733's, others would take some other variant, etc.
The end result is that usage would tend to be split across all of the highlander variants, and as long as their usage was higher than the Baseline Drop Average, ALL of their prices would be higher than average. No one of them might be as extremely high as the 733C's price would likely be to start out, but they'd all probably be higher than average.
And if that is ever not the case, where that means all those variants are used at the average rate.. which mean's you've achieved balance. Huzzah!
Quote
As I pointed out, the reason I'm not including those statistics is that they will complicate the system for no significant gain. Collection of the data wouldn't complicate anything, but processing it would make the system more difficult to deal with.
Quote
There's really no way to abuse the system not taking into account weapon statistics. The only effect is that a high value chassis loaded up with a poor configuration will have the same value as a configuration loaded up with a good configuration. And that's perfectly fine in my mind. It's not a problem worth worrying about... because people don't tend to intentionally bring bad configurations. People are TRYING to make a mech as effective as possible... and they choose a specific variant because it carries what they believe will be an effective loadout.
Regardless, I understand that you believe it's necessary to include individual weapons and equipment. I disagree, as the system here inherently accounts for all of that based on the fact that people select a mech variant based largely upon the weapons it can carry. Thus, drop frequency is related to weapon usage, and as such weapon usage doesn't need to be considered separately.
I certainly am not suggesting that you shouldn't go any construct the math for a system which includes all of that, if you so choose. It'd be great if you did. I'm just saying that I'm not going to spend my time on that, because I don't think it'll have value in proportion to the complexity it would add.
#35
Posted 06 March 2014 - 12:02 PM
There's a _lot_ of players, even if people like to think there isn't. Particularly in the Average Folks elo bracket, there's thousands dropping all the time.
This volume is what makes Roland's system so plausible. In order to game it, you'd need an absurd number of people working in concert over a long time (sliding scale for adjustment). You'd never get that many people to cooperate to that degree.
Edited by Wintersdark, 06 March 2014 - 12:05 PM.
#36
Posted 06 March 2014 - 01:18 PM
The Jenner K is considered undesireable in comparison to the founder jenner due to the lack of an additional streak. You could easily then allow that kenner K to bring two ppc instead (a very common thing in high elo) and since its not tracking the ppc itself and only going by the founder jenner you get an unfair advantage on your team that you can use for more BV wich the other team would be forced to meet.
Basically it makes it so that Unless you are CONSTANTLY thinking outside of the box and CONSTNATLY underbidding you always allow for a team to have an advantage.
Its actually highly exploitable.
I cant get behind a system that could so easily be exploited on that level.
#37
Posted 06 March 2014 - 01:33 PM
Quote
But it's not abusable, because the system will adapt to that, and then those "niche builds" will become more expensive.
This is exactly why this system is superior to a static BV system, in that it will automatically balance itself in response to exactly that type of action on the part of players.
Quote
What exactly is the Jenner K build that you are saying runs two PPC's and is "very common" in high elo games? Because while I've encountered such builds, they are kind of joke builds compared to other Jenner builds. 14 tons of PPC's basically means that the mech is making massive sacrifices.
Regardless, it doesn't matter, because if it's actually a good build, then it will become more widely used, and its price will increase.
Quote
Again, I have to point out that you are misunderstanding the application of this system.
Your team gets a certain BV, based on the number of members on your team. What you bring has absolutely no impact on what the other team gets. You can't force them to use some lower BV value, which you keep mentioning.
Quote
As I and others have pointed out, it's really not exploitable, and is extremely robust in this regard because it's specifically designed to adapt itself to any actions of the playerbase. I'm not sure what part of this you're missing though.
I'm thinking that you're not considering that second key part, where the actions of players will reflect itself in price changes.
#38
Posted 06 March 2014 - 01:39 PM
1. Hero mechs are less popular due to being MC-only, and a good hero variant will result in P2W situation - you would "spend" less BV than you should because you paid RL cash.
2. Popular mechs would be difficult to unlock/upgrade for new players. As soon as they buy that new "meta" variant, they will start wasting BV of the team they are on. They may lack proper engine, weapons, DHS upgrade, speed tweak, etc., yet that mech would already be treated by MM as a fully pimped high BV build. The more of those you get on the same team, the more likely it is for that team tolose the match and that would slow down XP/C-bill gains for that player.
#39
Posted 06 March 2014 - 01:55 PM
Quote
Yep, this is potentially an issue. Although for the very reason it's BV is lower than it would otherwise be, its presence on the battlefield is also limited. And at the same time, most of the hero mechs are bad anyway.
But I acknowledge that this is an issue which could result from the fact that Hero mechs have an additional aspect to them. Although I'd point out that perhaps the REAL issue is that it's kind of a bad idea to have mechs which can never be bought at all except for real world money.
Regardless, I suspect that if it does become an issue, it's such a small portion of the overall drop population that it could be dealt with separately on an individual basis. You could even choose to put in something like a modifier for hero mechs based on their MC cost.
Quote
This is true to some extent, although I would point out that there are a few "saftey valves" to prevent this from being a signficant problem.
Mainly, for solo players and small teams, they are given enough of a bonus to their PPBV that they really shouldn't have much trouble dropping in whatever they want, regardless of its BV.
So really, the main players affected by what you describe here would be players using brand new, unlocked mechs in large teams.. Which, I would guess, is not extremely common... Or at least, not commonly paired with a situation where making the most out of every BV point is of critical importance.
For a serious team playing competitively, they probably wouldn't be dropping someone in a mech that is running without upgrades and isn't completely unlocked. So this isn't an issue for them.
For large teams which aren't playing competitively, they probably can afford to let Joe "waste" some of their BV by bringing his new mech, if they are just playing for fun.
So I suspect it won't be a real big issue for anyone.
#40
Posted 06 March 2014 - 02:03 PM
You also under this system are basically encouraged to play the over under. You under bid with certain mechs and make specific meta builds out of unusual chasis (VERY DOABLE) so that you can free up more bv to do the super strong bv mechs (733c previously).
It just makes mech building the new meta..... Id rather not have to worry about this honestly myself since I honestly enjoy mech building and this would make it a grind....
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users