15.0 Alliance Wars Update Discussion Thread

Thestoryteller6 wrote: »

DNA3000 wrote: »

hurricant wrote: »

linux wrote: »

Max_ wrote: »

We have adjusted our game playing as a group we focus on feature 4 and 5 start champs. It’s nice to see different champs in AW defense. The MD is no longer and issue. Thank you. I hardly come across it now.

The new system is not the best but it’s playable.

But the problem with taking slow action is that Kabam will wind up losing all the players who liked competitive AW in favor of those who just want another easy chore like AQ.

Like someone said earlier, they could ruin war in one big change, but suddenly fixing it takes months

Unfortunately, this is typical for how MMO development works. When it works it works, when it doesn't work it just doesn't work.

A project like this generally starts with a problem statement. Alliance War needs work because blank. Greatly simplifying things, blank turns into a list of problems, the list of problems turns into a set of metrics that can quantify the problem, a redesign is created that is intended to address those problems, the design is implemented and pushed out. Then the game is run with the new design and data collected on those very same metrics to see if they are improving. If they get to where the devs want them to go, they are done (as done as anything is ever done in an MMO). If not, the tweak parts of the design and go again: test, measure, repeat.

Typically, the major design and implementation work happens in one long phase under development conditions completely different from what's going on when the devs are just tweaking things, so they don't just redesign things. It is always design, implement, datamine, review, tweak, datamine, review, tweak. This loop doesn't go back to design often, and datamine generally takes a significant amount of time: at least a week, sometimes a month or more.

The fact that Kabam has gone relatively quiet tells me there's a good chance they are in a tweak, datamine, review, tweak loop. There's nothing to talk about while data is being collected. And since the tweak loop doesn't generally make major design changes (there are complex operational reasons this is generally true) and since I know only design changes are going to address the problems I perceive in the current version of war, I'm unwilling to simply wait and see.

Keep in mind: I defended the devs when it came to the equally controversial balance changes to Dr. Strange. I thought they cut too deep, but I said then and I continue to say that sometimes the devs design/tweak/datamine/tweak methods are the only way they can really get to where they want to go, and Strange was an example. I think it was obvious his healing was cut too drastically, but that's something you can iterate to a better solution. I wish it could be done better and faster, and many people still think he's cut too deep, but no process can guarantee the devs will agree with the players on what needs to be changed. The point is when you're tweaking numbers anyway, iteratively tweaking numbers is a legitimate way to try to find the right ones.

But here, tweaking numbers is not the right way to solve the problems AW has. So datamine-driven iterative tweaking isn't going to get there no matter how many iterations and no matter what metrics are being monitored.

Maybe they've gone back to the drawing board and that's why they are quiet. I suppose it is possible. But if they were doing that, there's no reason to keep that a secret. The most logical reason for being quiet is they still believe that they can datamine their way to a solution and they are just waiting for the good news to come from the game data.

You think it is frustrating not knowing what's going on. It ain't less frustrating when you have a pretty good guess.

Just want to mention that reports from NYC seem to imply that there is a large change being worked on. Also they are to imply that everyone on the front lines of Kabam is aware of the fundamental issues with AW. I agree with you completely, why not just come out and say it and save a lot of grief? Well, I don't know but since it's Kabam I'm willing to believe they are working on an overhaul and just decided not to communicate that.

That's what I've heard also, from public statements. I'm willing to see if the next iteration of AW is clearly an iterative number tweak - and you can't disguise those from me - or an actual design change. If it is an actual design change, even a bad design change, I would consider that progress if it means they are reopening the design.

GroundedWisdom wrote: »

PlasmaKing wrote: »

GroundedWisdom wrote: »

Which means the current solution would be Ranking or boosting in other ways. Without forming an opinion, that would be the answer to optimizing efficiency. We are going to lose Wars now and then in whatever system we play.

Sorry again you miss the point alliance war isn't fun and isn't war. It's a cheap diy AQ right now. You don't need to actually fight to win, you know before starting the result in the main.

War needs to allow the ability for weaker teams to win and tbh that's through skill and dying less. There are many cases of real battles were strategy and the captains ensuring they kept their troops alive longer while killing the enemy faster won the day.

If you're in an alliance and you don't like dying then don't play AW, but war has to be war otherwise call it short AQ because right now at best that's what it is.

What you're talking about is penalizing the opponent through trying, and having a penalty for attempting has very little to do with skill at all. It's not necessary to have Defender Kills in order to have a skill component. That can be achieved through adding some other form of difficulty. By having metrics for Defender Kills that become the main focus, you're actually creating a situation that removes the ability for Offense. Thereby making it Defense Wars. I'm sorry, but I do not agree that Defender Kills are absolutely necessary for a skill component. The only time that is used as an argument is in reference to the opposing team. What it really means is people are upset that they can't win by causing the other team to try itself into a Loss.

What is necessary for skill to be meaningful in any game is it must be measurable. Two different performances must contribute differently to the outcome of the game. In 14.0 offensive skill was measured in terms of the balance between node exploration and kills verses defensive kills, which are offensive defeats. Measuring progress verses cost for that progress is a valid measure of offensive performance. In 15.0, offensive kill is measured only in terms of map progress, and against quantitatively weaker defenses. Because the current system is easy to saturate (i.e. it is possible for both sides to fully explore the map) 15.0 does not meaningfully use offensive skill as a metric for winning the war.

You are correct that defender kills literally are not the only way to measure offensive skill. However, you have yet to address any of the alternatives I've proposed to replace defender kills with an alternate offensive side metric. You've simply parroted the idea that all other offensive metrics are basically identical to defender kills, which is patently false.

nuggz wrote: »

If they wanna make the nodes ridiculously hard then we just end up going back to placing hard defenders with a little less diveristy and hope we aren't facing whales.

A part of me wishes Kabam would implement this for a couple rounds of AW just to show to the people who think this is a good idea that it is not a good idea.

Eliminating defender kills didn't make all defenders equal. It eliminated all shades of grey between defenders (a caveat I made sure to repeat often). It is still true that Magik is better than Spider Gwen. It is just that at the moment there are two kinds of defenders: those that can literally kill multiple players dead (quantitatively, they have the theoretical capability of getting more than four defeats when placed on a particular node) and those that can't (which, given the current way AW works, are all basically equally marginally useful). If you change the nodes, you can change things in one and only one way: you can move the dividing line up or down. You can make more top-tier defenders or more ****-tier defenders. And if you make the nodes strong enough, you can even return the illusion of 14.0 AW: every node is so hard that the right defender placed there can stop a player dead in his or her tracks.

And when the nodes get that strong, we can all ignore diversity and rating points completely and go back to placing an entire map of Magiks, Dormammus, Icemen, and so on. But this time, players won't be giving up after attacking once. Some alliances will look at the maps and give up at the starting gate before even attacking once.

I don't think anyone really wants that. But in my darker moments I sometimes think it would be fun to take that implementation and swing it around until 95% of the alliances beg for mercy.

Grufty wrote: »

Kabams reason to remove defence kills was as follows: we didn't want players with healthy champions to stop moving due to fear of giving points away.

FYI: in our most recent war we reached 100% and killed all three bosses. I get the impression that at least one whole battlegroup in the opposing alliance saw that, saw they were behind, and just gave up trying.

In the current AW system, among the many very bad and very predictable problems they created, I don't think they even solved the problem they were trying to solve. While removing defender kills might now be encouraging players to fight on that were giving up before, they have also created all new ways for entire alliances to give up a war when they realize there is no hope for winning on points, because the other side already has too much points.

You can see how diverse your opponents are in theory as you play through. You know how diverse you are. If you are not maximally diverse or likely to out-diverse your opponent and they reach 100%, you're dead right then and there, no matter how many attackers died to reach 100%. You can't catch up even if you have infinite offensive skill.

As I keep saying, the problem wasn't eliminating defender kills. The problem was failing to understand the role defender kills was playing in AW, and not recognizing the need to fill that role with something else.

This is really important, and I hope the devs understand this in the next iteration of AW. Contrary to what some players have been saying in this thread, ON BOTH SIDES, defender kills were never about defenders. They were about measuring the skill of the offense. And the reason why it is CRITICAL to make the most important variation in points be offensive performance is that offensive performance IS THE ONLY THING THAT HAPPENS DURING THE WAR.

Every other source of points is static and is determined the moment the war starts. From that point on, the only thing that "happens" is offense. How the players play the game is the only thing that changes from minute to minute during the war, and we only play on offense. If the deciding metric is measuring that performance, then (outside of extreme overkill situations) the war isn't decided until the end of the war. But if it is decided by anything other than offense, then it was decided before the war started.

Removing defender skills removed one way to reward better offense. Adding defender diversity (and adjusting the nodes) lowered the offensive difficulty making it harder to tell the difference between offensive performance. These two things are called "defensive" things but both actually cripple the ability for the game to judge and reward good offense. That shifts the decisive factor to defensive points, which are static and fixed throughout the war. Which makes war intrinsically more predictable and it makes it feel more passive.

Frankly I don't really care on a fundamental level if defender diversity is rewarded or if defender kills are eliminated. These are cogs in a larger system that is broken not because those cogs were added or removed, but because the machine wasn't redesigned to accommodate the new parts list. Most cars don't need or use carburetors anymore but if you remove a carburetor from a 1972 Camaro you can't replace it with some paperclips and a fidget spinner and expect it to work. It is wrong to argue that all cars need carburetors or carburetors are bad and should be replaced. That argument misses the point entirely.

DNA3000 wrote: »

GroundedWisdom wrote: »

PlasmaKing wrote: »

GroundedWisdom wrote: »

Which means the current solution would be Ranking or boosting in other ways. Without forming an opinion, that would be the answer to optimizing efficiency. We are going to lose Wars now and then in whatever system we play.

Sorry again you miss the point alliance war isn't fun and isn't war. It's a cheap diy AQ right now. You don't need to actually fight to win, you know before starting the result in the main.

War needs to allow the ability for weaker teams to win and tbh that's through skill and dying less. There are many cases of real battles were strategy and the captains ensuring they kept their troops alive longer while killing the enemy faster won the day.

If you're in an alliance and you don't like dying then don't play AW, but war has to be war otherwise call it short AQ because right now at best that's what it is.

What you're talking about is penalizing the opponent through trying, and having a penalty for attempting has very little to do with skill at all. It's not necessary to have Defender Kills in order to have a skill component. That can be achieved through adding some other form of difficulty. By having metrics for Defender Kills that become the main focus, you're actually creating a situation that removes the ability for Offense. Thereby making it Defense Wars. I'm sorry, but I do not agree that Defender Kills are absolutely necessary for a skill component. The only time that is used as an argument is in reference to the opposing team. What it really means is people are upset that they can't win by causing the other team to try itself into a Loss.

What is necessary for skill to be meaningful in any game is it must be measurable. Two different performances must contribute differently to the outcome of the game. In 14.0 offensive skill was measured in terms of the balance between node exploration and kills verses defensive kills, which are offensive defeats. Measuring progress verses cost for that progress is a valid measure of offensive performance. In 15.0, offensive kill is measured only in terms of map progress, and against quantitatively weaker defenses. Because the current system is easy to saturate (i.e. it is possible for both sides to fully explore the map) 15.0 does not meaningfully use offensive skill as a metric for winning the war.

You are correct that defender kills literally are not the only way to measure offensive skill. However, you have yet to address any of the alternatives I've proposed to replace defender kills with an alternate offensive side metric. You've simply parroted the idea that all other offensive metrics are basically identical to defender kills, which is patently false.

What I said was that there are other ways to add a degree of difficulty and skill besides making it punitive and limiting. There's no question that the old metrics were accumulating to the point of being the definitive aspect. Again, I never debated your suggestions.

@DNA3000 What is it specifically that you want me to address in terms of your suggestions? When I gave examples for Offense, I really wasn't trying to parrot anything. Just displaying a point. It's been a long Thread. It would be easier if I knew what you wanted me to comment on.

RagamugginGunner

GroundedWisdom wrote: »

What I was addressing is that Defender Kills are absolutely necessary. Which they're not.

And you are 100% wrong.

LegionDestroier

System still full of lunacy. We can see earlier when to give up. A lower rated higher skilled team cannot win as the tie break is the defender rating. Still so broken. When are you going to fix it Kabam?

GroundedWisdom wrote: »

@DNA3000 What is it specifically that you want me to address in terms of your suggestions?

Any specific element of them, really.

Kabam, please let us know what the future looks like for allaince wars. You know how we feel.
When you do respond, do so appropriately, in that our concerns are addressed in a manor that we have been asking for. Whether or not we like the response just give us answers. Then we can decide for sure if we would like to continue playing or not based on the answers.

Again be very specific in your plan of action moving forward

I'm sick of using all my hard earned rank up material to satisfy diveristy and rating thereof. I had plans for my recourse. I know I'm not the only one here that is in this situation. The champs I ranked up for diversity don't even get kills actually they are lucky to take a hundred health off the opponent attacking it. BUT it gives us 50 points in aw. Woot. Oh, where else do I find myself using these newly ranked champs?.....

NO WHERE!

Thanks for forcing us to play the game in way that not only we don't like, but hinders the champs that we need to rank up to finish other material in the game

DNA3000 wrote: »

GroundedWisdom wrote: »

@DNA3000 What is it specifically that you want me to address in terms of your suggestions?

Any specific element of them, really.

Let me scroll back when I have some time and I'll process.

Jon8299

The AW changes was the last straw for me and many in my alliance. I haven't played in weeks and I'm not sure this Boss Rush challenge will be enough to bring me back. I'll try it, I just have low expectations for it.

Members are leaving and honestly it was hard enough to recruit before the AW changes. Now you can't have life first alliances cause you need everyone to fill up the BGs and everyone to join on the first day of the attack or everyone will have conflicting schedules. Oh I can't attack tomorrow until the AW is almost done cause I have this, that and the other to worry about.

Theses AW changes hurt life first alliances. Maybe not all but mine and others, probably a good chunk of life first alliances. How am I supposed to fill up my alliance now? It's no longer possible to play the AW if you have a busy week, so if you played or went at a snails pace in other game modes, you might as well just not play anymore.

They stripped the AW of everything that made it about strategy and skill and replaced it with a boring tedious chore to do. Where you spend all of your starting energy on 0-1 fights, maybe 1-2 but only if you come in late.

All we need is a plan of action and end result for what kabam has in mind. For better or worse, doesn't matter just need to know whether I'll continue playing or not.

DNA3000 wrote: »

LocoMotives wrote: »

Well said @DNA3000. I really think the simplest solution to fix AW is to make diversity points equal to a single defender kill. Then there are actual choices to strategize and risk/reward incorporated in choosing what kind of defense to run (full diversity, strongest available, or a mixture of both). I can't see how to fix the idea of AW without having d kills involved in some manner.

I don't like the idea of diversity points in general. What if we implemented a "conservation of ninjitsu" principle (sorry for the tvtropes reference)? Suppose that if an alliance placed more than one copy of the same champ, each successive copy would be weaker. So the first Magik you place would be full strength. But the second one would be only 80% of the strength or something. The next one would be 60%. The fourth would be 50%. Pick the right numbers, something that eventually leveled off. And the game would choose which one was "first" by node number. The highest node number would be the "top" one. The second highest would be "second" as so forth. That way the bosses and miniboss nodes would get priority.

This way nobody gets "points" at the start of the war for doing anything in particular. The incentive and disincentives for placing a diverse defense are built into the map and the combat, not in the scoring. This serves two purposes. First, instead of an abrupt transition where the first Dormammu is worth a ton of points and the second one is worth almost zero points (under the most recent revision a kill equals his placement, plus or minus a few points for rating), we now have a more gradual penalty where the first one is worth full points, and the second one is worth the same amount of points but is easier to kill, etc.

This only works if we bring back some kind of defense side points, otherwise we will end up with too many ties. But we want to disguise defense points so people don't think they are getting penalized for trying and failing to kill the node. So what if we change the attacker points so that instead of just handing the attacker points for the kill, we give the attacker more points if they are a good attacker, which also means we give less points of the defender is a good defender. One way I can think of to do that is to use time. Suppose that we give the attacker points based on how much time it takes to defeat the node? This automatically factors in defensive kills in a sense, because if it takes multiple attackers to kill the node the tendency is for that time to be longer.

Suppose we give the attacker 100 points per kill, but we divide that by the number of minutes it takes to kill the node. And for mathematical numberish scorey reasons lets make the first minute free. So if you kill the node within the first minute, you get 100 points. If you kill it in two minutes, 50 points. If you die in two minutes and then try again and kill it in 30 seconds with your second attacker, 40 points (100 / 2.5 minutes).

This reincentivizes placing strong defenders because strong defenders in effect take points away from the attackers. But placing the same defender over and over makes them weaker, which hands the attacker points. Even in the very top tiers of war, even in tiers where everyone gets 100% complete, there is still a way to distinguish between good attackers and bad, between good defense and bad.

This would be the second purpose: to return tactical decision making to the players. Instead of being told what to place, which the diversity system essentially does, this system hands the players a set of pros and cons and asks them to find their own balance point that trades them off against each other.

And instead of the Nash equilibrium being at one extreme or the other, it is actually in a fuzzy middle of the game where you want diverse defenders, but you also want the strongest defenders, and there's no easy way to calculate the perfect balance between them. So different players and different alliances will likely compute different "optimal" strategies. Which means you could get diverse defense placement. And that's I think the important diversity. Not diversity of different champions. Diversity in unexpected map placements.

I think this could be tweaked into a reasonable balance, although I admit I would want to think about the precise numbers to use. I'm just tossing out numbers to illustrate the principle at the moment. I would want to have top tier alliances playtest this for best balance between the attacker scoring and the defensive penalty. Maybe a minute is too long, and the defensive penalty almost certainly needs to be tweaked.

This idea isn't perfect, but it does reward good attackers without penalizing defensive kills directly, it encourages diverse defense placement without awarding point advantages, and it strongly encourages direct attacker/defender competition again. I wouldn't say it "fixes" AW, but it shows it is possible to fix it even within the parameters that Kabam appears to be operating under and what I perceive the players' strongest complaints about 15.0 are. I also think it is less likely to encourage degenerate player behavior. Placing a bunch of 3* champs for diversity gets significantly punished.

So, keeping in mind it's 5:30 A.M. here and I'm skimming with the Search Engine, I can offer my thoughts on this.
I can see what you're suggesting with the diminishing strengths for multiples placed, but I'm not sure I could get behind that as a viable encouragement for Diversity. Inevitably, people will place repeats regardless. Of course this comment may have been made before the Diversity Points were lowered, but I'm assuming you still feel the same way about having them in general. I actually agree with having them, for a few reasons.
The idea on Points for timing are interesting. I'm just curious how Suicides would play into that. They would no doubt become the norm, and that doesn't necessarily translate to skill to me as much as Masteries.
I agree that an element of engagement for Defense would help. One thing you mentioned in another Thread (may have been this one, I was scrolling Search Results) that is sticking to my mind is the idea of alternating Nodes. That could be interesting. If they were set on a randomizer and hidden from the opponent, at least partially, that would be interesting. Perhaps Nodes that are useful for two or three Champs to encourage using a selection.
Another idea that occurs to me is Traps. This would encourage Placement as well as provide challenge for Attack. Traps could be hidden until the Map is engaged. For example, let's say a Trap is Heal Block, or Degen, or 1% Poison. If there are a certain number of Traps allowed to be used, it could make it interesting with placing different Traps with different Champs on various Nodes.
To be clear, I wasn't ignoring any of your ideas. I've just been focused on certain aspects of the conversation. I understand the point you're making. We may not always agree on things, but I still respect your views.