Arcanaville

Honestly, that specific part threw me as well. Nowhere is it established that he sees this quest of his being about "freedom" in any real sense: its about achieving ultimate power. So its a bit weird to cry out "is it so wrong" because if he's crying out to a hero player, the answer to "is it so wrong to become a murderer to achieve ultimate power" is yeah, and if he's crying out to a villain player, the answer to "is it so wrong to become a murderer to achieve ultimate power" the answer is no, which is why I'm stomping on your weasel face, and if its just to the universe itself, then the answer to "is it so wrong to become a murderer to achieve ultimate power" is "I'm sorry, your call cannot be completed at this time, please check the number and try your whine later."

In other words, he should not expected understanding from a hero, mercy from a villain, or anything from a universe he just tried to use a cheat code to.

Fair enough, but then what you're talking about when you refer to elitism bears no resemblance to what I understand elitism to be or to any reasonably common definition of elitism. Its something mostly unique to you. By your definition of elitism, any expression of preference other than player skill - something we cannot judge or measure prior to actually running content with the player - is elitism, and by that definition I'm elitist. I sometimes express a preference for teaming other than player skill. Like if the player is a jerkwad, I might choose not to play with them. I'm an elitist when it comes to dipsticks.

Also, certain tasks like the Avatar fight at the end of the Underground are extremely difficult to accomplish without significant aggro control. Without a few tankers or number of brutes with taunt, the likelihood of a random assemblage of players being successful is low. Its not impossible, but I'm far less likely to want to spend an hour or two of my time waiting for such a league to form and then attempt it, unless I have some special interest in seeing that attempt succeed. If a friend asked me to fill it to make the attempt, I probably would. But otherwise, I would check my elitist butt out of there and go run a DA arc.

The way most people define elitist (within the context of teaming in this game) is either to demand the best even when good enough is good enough, or else judge actual players based on superficial accomplishments and not on the merits of whether they are sufficiently capable of running the content. By that definition, I'm not elitist. By yours, I am. I'm ok with that. I'm still not going to self-identify as an elitist just because your definition is so expansive that it includes me, because I don't acknowledge that definition as being particularly representative of what the word means to the vast majority of players.

I'm pretty sure I didn't say any of that. Now would you like to discuss this completely different thing, or the original topic about why THB's data is at best an experiment, and not statistical data. I'm good either way.


Also: if you go by reward generation and leveling speed the blaster is not doing better: if I'm interpreting THB's data correctly the Blaster is leveling slightly slower (about 2% slower, with fast travel and no debt) from level 3 to level 9.

I don't think that's an entirely fair characterization. I believe the dev team in all its incarnations has had a vision for the game, and that vision includes their interpretation of what would be the best mix of casual entertainment and challenging entertainment that most games have to balance.

I think the biggest misinterpretation of the devs' motives relates to the fact that the difference from Geko to Castle to Black Scorpion is that the dev teams have over time become more aggressive in their design philosophies. That has tended to be misinterpreted as being more attentive to the players, when the dev team has always been attentive to the players. And its not that they are simply doing what the players want more than before, its that their design reach has tended to expand over time to include many of the things that prior dev teams would not have done.

That has not come at zero cost to the game. Its good in some ways, not so good in others. Mostly, its just different. Its much more likely that this dev team will do something that makes the players go wow than in the old days, but its also more likely that the current dev team will do something that makes the players go wtf as well. That's the tradeoff when you have a more aggressive design philosophy: more fireworks going off overhead, and more fireworks going off in your face.

The thing I think we should all remember about Jack is that in retrospect, knowing everything about the original dev team, their reach and capabilities, and the original design target for this game, I believe had Jack not intervened with a much more simplified and rigid vision for the game, we wouldn't have one now. Maybe we outgrew that vision, but at one time that vision was the only thing that saved the game from oblivion.

Perhaps we just got lucky: we had a conservative team at just the right time when the game needed one, and we had a more evolutionary team when the game could afford it, and now we have a more aggressive team when the goal is to spice up the game in its middle age. And then again, maybe that's not luck: maybe that's the dev team collectively being smart enough to know what the game needed when it needed it more than we players give them credit for. The game design is not a Nemesis plot: they don't do everything perfectly and in just the right way at just the right time, but they may see which way the wind is blowing and trend in the right direction in the long run. Maybe a little of both.

If Scrappers are vanilla ice cream, then Brutes are a cookie dough blizzard. But I don't think that means Scrappers need something different sprinkled into them. Some people would rather just have ice cream. Particularly when its extremely good ice cream. Rolling the top in M&Ms doesn't make them equal to Brutes, it just makes them distasteful to people who hate M&Ms.

Personally, I like Mint Chip. Which is doubly bad for throwing extra junk into. I picked a mint primary and a chocolate chip secondary, and now I'm done. If I wanted Fury or Gauntlet, I would go to Cold Stone.

That's interesting, because I would presume most people hospital back more often than not when solo. I cannot easily tell how often missions are taking in your statistics because there seems to be just time entries and not start and stop times, but it might make sense to keep track of how much time is spent returning to a mission when dying. The presumption made on death was that the debt was only part of the problem, and the extra travel time was as much a factor in slowing down blasters as the incurred debt, and also why there was a performance gap below level 10 when debt doesn't exist. I cannot easily tell what the penalty for dying was in the mission it occurred in.

Do I link to a definition of statistics, or just post a facepalm. Decisions, decisions.

Rularuu nose hairs.

Of course, you only got them after you stopped asking for several years. So I don't think the clock has started on this idea yet.

Instead of a moon zone, we should have a Rularuu zone where we literally fight on Rularuu and defeat him by taking out his chakra points or something.

Then again, given how crotch-fixated the playerbase is, perhaps that's not the best idea.

No debt until level 10, but how are you rezzing? Inspiration, Return to Battle, or Hospital?

Definitely not. Probably not.

Your agenda is obvious. But the way you change portraying it over time is not. You've often stated that you don't want Tankers to have the highest damage, but you've then used arguments that stated there's no justification for Tankers to have *lower* damage than any other melee. You've said the problem is that this game doesn't represent tankers correctly, and then you use as examples of that characters with both the highest defense and the highest offense as examples - then you say you don't need the highest offense. You say that Brutes stole fury from Tankers, using as evidence statements from someone you then selectively ignore when the overwhelming evidence states his original vision of tankers was to be a low damage archetype.

You obfuscate what you want so you can ask for everything, then claim you're not asking for anything in particular, take credit for every change you want while also claiming those changes don't go far enough.

Its precisely *because* your motives are transparent that no one trusts what you say, at least where it involves Tankers. Everyone knows you believe Tankers to be the thing that hits the hardest and is the strongest both offensively and defensively, and Scrappers and Brutes very existence prevents you from having what you want. So you quibble about how to chip at the margins, when everyone knows the margins are not what you're really interested in.

And yeah, that's coming from me. I'll put my reputation for transparency against yours any day. The day I lose that one is the day I become a Tibetan monk.

Although blasters are easier to play at the lower levels, they ironically also don't have the damage edge they are ultimately intended to have against other archetypes because as you say, AT damage modifiers don't fully establish themselves until level 20. Prior to level 20, the blaster 1.125 ranged modifier is effectively lower than that relative to other archetypes.

For example, at level 15 the Blaster ranged modifier is only 48% higher than the Defender one, and only 65% higher than the Controller one. And while that may seem high, consider that solo Defenders have a 22.5% damage buff at a time when most players are not heavily slotting damage yet (the Vigilance buff also scales with level until 20). And Controllers have containment. The actual modifier edge may be less than 25% against defenders and actually a wash against controllers (having more attacks will likely still give the blaster an edge, but not nearly as much of one as the numbers would suggest at a glance).

By the time the Blaster modifier fully asserts itself at level 20, the threat level has also become high enough to erase that modifier advantage.

Blasters and Defenders share the same mitigation caps. Should they then share the same damage caps?

Defenders and Corruptors share both caps. Should Corruptors then be allowed to have base damage buffs that bypass the damage strength caps?

How about this one: even at the mitigation caps Tankers will still have more health than Brutes. So should Brutes have higher damage in direct proportion?

The caps argument is a smoke screen, because J_B also believes that at or near the caps defensive differences are irrelevant. So even if Tanker mitigation caps were higher, it would not mean that the offensive cap of brutes would be justifiably higher to J_B, because that difference would be irrelevant. Its only meaningful because it supports his assertion that Tanker offense should not be significantly lower than any other melee archetype. The argument must be valid both ways, or its not valid.

Except that's not what I'm talking about, and this is the third time I'm having to say it. The statistics can point to an existence of a problem. It won't always, but when it does you can't ignore that data. That data won't always give you the right information to optimally address the problem, which is where modeling comes in. Modeling *must* match the existing data, or the model is wrong. *Fitting* the data doesn't mean the model is correct, models become increasingly trustworthy when they are capable of predicting future data collection without significant modification. Models have to be tested by experiments, and in a sense THB's comparison could be an experiment to test a model.

But THB doesn't have a model to test at the moment, at least not an acknowledged one. His thesis is that Blasters don't have the performance gap that the data shows, because he doesn't trust the data, and he wants to generate comparison runs to see if those comparison runs demonstrate that the performance gap doesn't exist.

For sufficiently opaque definitions of "clear."

Funny, because I was thinking it would be hilarous if the entire thing turned out to be a Nemesis plot to manipulate Wade into merging with Rularuu so that the players would defeat him and allow his mind to scanned for Rularuu's secrets so that Mender Silos could use that information to fight the Coming Storm.

I mean, the players would burn down Paragon Studios so I would never actually get to see the Coming Storm, but it would still be hilarious.

Actually, the problem was slotting with extreme invention builds. And technically speaking, the performance of that build is still not actual data yet, in the sense of being analyzable.

Extremely unlikely on Triumph. The players tend to avoid it, the server culture tends to ridicule it, and the population numbers can't support it.

As I said, an analysis of the details can give you greater insight into the best way to address the problems, but they aren't as good a starting point for determining *if* there is a problem because it can be impossible to know how those details fit together holistically.

First we find out that there is a problem, and then we break the problem down into its component parts. If Blasters underperform, they have a problem by definition. If they are getting mezzed far more often, that suggests that mez is a component of the problem. But going the other way around suggests that if Blasters get mezzed second most often, and controllers, say, get mezzed even more often, that implies controllers have an even bigger problem than Blasters do. But if that mez issue doesn't translate into a performance issue, then it may not be a problem given all the other weapons at controllers' disposal.

What few people possess, and I assert no one possesses, is what things actually constitute real problems. The fact that they could underperform by the enormous margins they did, moreso than any other archetype, with no one being able to prove it and few people even daring to express it, suggests that the statistics are extremely important to inform everyone's intuition. Because everyone's intuition collectively failed Blasters. Even I thought they underperformed, but not as severely as they turned out to.

Using individual anecdotes will likely tend to reinforce everyone's preexisting notions of blaster performance, and prior to Defiance 2.0 everyone's preexisting notions of blaster performance were wrong.

If your model of Blaster performance doesn't predict I10 Blaster performance, saying you don't trust the data and so are going to invent your own suggests that you'll reject any data that contradicts your intuition. At that point, all the modeling in the world is not likely to generate a reasonable solution, because the data its based on is strongly skewed. Models are constructed from data. And the statistics of the entire playerbase are a perfectly good starting point for that model, and probably better than a single player's experience.


No, but it does tell you that there's a problem with sniper blasts, no matter how many players you ask who say they are fine. If someone says that they love snipes and everyone they know loves them, and the datamining showed that less than 1% of blasters actually took them, I would know that the player was either lying or a strange exception. But no amount of polling would convince me snipers were fantastic, if only 1% of players took them.

Conversely, when the data says Blasters earned XP, inf, and other rewards slower than other archetypes, that they did so in basically all security level ranges, that they did so while solo and teamed, and that they had more debt, died more often, were mezzed more often, and died while mezzed a high percentage of the time, *that* statistic does tell you something. It says no matter how much people say that mezzing is not a problem because of insps or Clarion or anything else, and no matter how many people start bragging about their own blaster performance, you know that blasters are vulnerable, they die, mez is often the cause, and tools to counter that will almost certainly help, if players can make good use of them. That statistic overrides anecdotes to the contrary.

I myself have personally spoken out frequently and often about players getting excluded from teams for all kinds of reasons. In my opinion, you should play whatever you want and if the team is any good, most of the time whatever you want is going to be good enough. The team has to be pretty shaky to begin with if one scrapper without mez protection or a petless mastermind brings the whole thing down.

But I do make an exception for tasks which have a proven track record of failing when certain power levels are not met. If you want to attempt a four player Eden trial, and you find three other daredevils, that's cool. But by the same token if those four players decide that isn't going to work so its not worth the effort, that's cool also.

Asking for lots of level shifts and incarnate powers in BAF, or Lambda, or even Keyes is probably overkill. I've seen them succeed many times without them. But MoM, TPN, and Underground all have aspects to them that can make them impossible to complete with leagues below a certain power level. And in some cases, you won't know until you've spent a long time getting to that point. Given that, if league leaders decide to only lead leagues that satisfy certain minimum requirements, I don't consider it elitism if they feel they are severely impacting the odds of success by not adhering to those standards.

You can run Apex with a team full of non-incarnates, but you're going to lose. Its not elitist to ask for players to be at least Alpha slotted. For Underground, MoM, TPN, and DD I don't think its as bad, but I do think its a judgment call on the league leaders.

Only if you want to get a "neutral" view of the archetype; sort of a how would it perform if everyone was forced to play it at gunpoint.

But I don't think that's all that useful in practice. I think in practice its much fairer to say that how good an archetype is is directly related to how much players actually squeeze out of it. If they get more out of Scrappers or Controllers than Blasters, how they do is not as important as the fact that they do.

There is a train of thought that suggests its actually possible to construct two archetypes that are "fair" in the sense that they have equal intrinsic power, but A underperforms B because people don't know how to get the most out of A. I think that line of thought only has merit when A and B are both options for which people have had a limited amount of exposure to.

But when A and B have been around for eight years, its likely that the average player is not going to suddenly get any better at either. Individual players may get better at A or B or both, but the average player probably stays mostly the same.

In that circumstance, I believe that A and B are equal if players actually get equal results from them, regardless of how they do. Its basically judging the two archetypes in the grand totality of their intrinsic power, their gameplay, their build options, and the degree of difficulty they present.

An analysis can show why one thing performs better or worse, but it won't dictate values. If what this game values is presenting players with a choice of archetypes, each of which presents approximately the same chance to succeed at earning in-game rewards, then measuring in-game reward earning is the best way to know if they actually present that choice correctly. Player behavior can skew the statistics in certain directions, in particular by altering performance at different security levels or by teaming at different frequencies. But when the statistics factor those things out, the only remaining player behaviors are the things we don't want to factor out. If players slot Scrappers better than Blasters, the question is why should Blaster encourage that behavior. If players learn to play Controllers faster than Blasters, again the question is why are Blasters so much harder to learn. But most importantly, even if we cannot figure out why these things are true, if we intend to present the proper gameplay choice to the players, the ultimate question is do we need to know. And the answer is no, we don't. It would be nice to know, but in the absence of knowing we can still try to compensate for those failings.

All I really assume is that most players try to have fun, and within that context try to do well. Every archetype gives them a different set of options for accomplishing that, and there's no need to normalize for that fact. The different options people exercise within each archetype is part of the value of the archetype we're trying to measure.

Convincing you doesn't help me that much. Convincing the devs helps me more. For an experiment like this to have any chance of outweighing the devs' own data, you'd probably need to get data from at least a few hundred players, all with objective evidence that they were performed with reasonable care.

That's ironic, given that you are attempting to use a single example to prove false something that has been determined by simply measuring the performance of everyone that plays the game. That Blasters underperform is true when you take into account all the people that play the game. Its actually others that keep trying to come up with excuses for why that assertion doesn't actually represent the real performance gap between Blasters and other archetypes. The devs datamining counts everyone. You're implying counting yourself is better than counting everyone.

The best you can do is determine whether you are representative of the average player or not, and by extension whether your impressions are representative of the average player experience. That would be useful information both to me and to the devs, although probably not in the way you intend.