EVE Search - Do We Need More than Databases to Determine Balance?

All Channels

EVE General Discussion

Do We Need More than Databases to Determine Balance?

» Click here to find additional results for this topic using Google

Monitor this thread via RSS [?]

Pages: 1 2 :: [one page]
Author	Thread Statistics \| Show CCP posts - 0 post(s)
Kuga Sacred Templars DARKNESS. 22	Posted - 2015.03.09 23:46:38 - [1] - Quote TL;DR: using database figures alone may not offer a comprehensive overview of game activity, or indeed may return misleading figures through the inclusion of confounding factors or failure to recognise significant contributions. Overall community sentiment should play a substantial role in decision making and help guide developers to issues where their database information alone may be inadequate. Firstly I will begin by saying databases are immensely powerful resources which, through directed scrutiny, can reveal detailed patterns, relationships and forecast the probability of events within the confines of statistical confidence. The crux of the system however, is that the greater the number of interacting parameters within the database, the larger the number of contributory factors to each parameter. Having worked significantly in the area, I have had extensive experience of this. The EvE database is much more complex than I have dealt with, amplifying this issue. For example, if we take the Eve database and say we want to look at the impact of Ishtars on null sec PvP, we can simply go to our database and see how much damage is being dealt by Ishtars. Simple! But is it the full story? So then we consider what the Ishtars are doing: is it PvE or PvP? Then we think further on the issue - what precisely are the Ishtars hitting? Before long the list of variables defining Ishtar effect on null sec PVP starts to look like this: PVP or PVE Structures or ships. What structures, what ships? How does their damage break down against each of these? How fast do they deal damage? How often do they die? What do they die to? Is their role in PvE influencing their use in PvP? What regions are they used in? How many Ishtars are on the average kill mail? How does their damage application and limitation of alpha damage per target limit this? Where are they being built? How are they moved around? How many jumps do pilots make in them through null sec? And the list goes on. Within a multifaceted system like the EVE universe it quickly becomes an impossibly Herculean task to include every variable in data analysis. Even if it could be done, how would you collate and present the data in a comprehensible, concise format? The answer is you don't. The statistician will select from the database the information which they feel, from knowledge of the system and personal statistical experience, the data they feel is most relevant. This will never narrate the whole story, but a select 'chapter', in this case of the impact Ishtars have on null sec PvP. The most recent example of such a clash between community and developer was the information released by CCP Rise on the role of battleships in PvP. Trusting in this data, CCP has chosen to take no action and now stands clearly at odds with the community sentiment on the issue. I will not comment on whether this may be the right or wrong decision: this is up to the discretion of the reader. Has the excessive usage of databases been clouding the judgement of those responsible for making decisions to the point where issues are being overlooked or eschewed in favour of balancing spreadsheets?
Serene Repose 2358	Posted - 2015.03.09 23:48:20 - [2] - Quote Statisticians use statistics much like a drunk uses a lamp post; more for propping themselves up than for illumination. Treason never prospers. What is the reason? Why, if it prospers, none dare call it "treason."
Hasikan Miallok Republic University Minmatar Republic 1419	Posted - 2015.03.10 00:15:09 - [3] - Quote no no .. balance works like this ... 1) introduce new feature/module/ship which is useful 2) nature of EVE is if something is even 1% better it becomes a doctrine and everyone uses it 3) forum whiners immediately start saying xxx is overpowered and quote all the doctrines using it as evidence 4) CCP responds by nerfing new item to the point of being useless
Kiandoshia Applied Anarchy SpaceMonkey's Alliance 2225	Posted - 2015.03.10 03:02:44 - [4] - Quote How dare you make so much sense! Someone lock this quickly.
Dun'Gal Myriad Contractors Inc. 206	Posted - 2015.03.10 03:45:08 - [5] - Quote In many ways I have to agree with the OP on this. "Everyone loves graphs," has become all to common a thing when balance comes into question, particularly when the thing being rebalanced is controversial. The "here's how much damage x ship is doing in PVP" becomes misleading when dealing with larger hulls, because as a rule they do much more damage anyway. Saying they are statistically being used more is... wishy washy at best. As you said it doesn't take other factors into account that could skew the numbers. Edit: an example off the top of my head is careless people taking there battleships and battlecruisers into hostile space and eventually being killed, they may well be doing damage to there asailants, but it doesn't mean that they are either in a good place, or being used more for PVP purposes.
Sibyyl Gallente Federation 23777	Posted - 2015.03.10 04:53:54 - [6] - Quote The only alternative to databases is checking for fate in chicken bones or consulting your totem animal. The right metrics need to be considered but not using metrics at all (like you seem to be suggesting) will lead to even more catastrophic game changes. If you mean "experience", we need enough data to drown out the anecdote factor that is so popular in politics and law. Friendship is the best ship Sabriz for CSM go go go
Serene Repose 2362	Posted - 2015.03.10 05:13:45 - [7] - Quote Dun'Gal wrote: In many ways I have to agree with the OP on this. "Everyone loves graphs," has become all to common a thing when balance comes into question, particularly when the thing being rebalanced is controversial. The "here's how much damage x ship is doing in PVP" becomes misleading when dealing with larger hulls, because as a rule they do much more damage anyway. Saying they are statistically being used more is... wishy washy at best. As you said it doesn't take other factors into account that could skew the numbers. Edit: an example off the top of my head is careless people taking there battleships and battlecruisers into hostile space and eventually being killed, they may well be doing damage to there asailants, but it doesn't mean that they are either in a good place, or being used more for PVP purposes. Aren't they applying the Stupid People Filter? Treason never prospers. What is the reason? Why, if it prospers, none dare call it "treason."
Unsuccessful At Everything The Troll Bridge 20321	Posted - 2015.03.10 05:59:32 - [8] - Quote Quote: Do We Need More than Databases to Determine Balance? No, we need more than balance to determine databases! *Since the cessation of their usefulness is imminent, may I appropriate your belongings?* Vote Sabriz Adoudel for CSM 10!
Sibyyl Gallente Federation 23783	Posted - 2015.03.10 06:29:15 - [9] - Quote Serene Repose wrote: How in the world did we get along before databases? Those tens of thousands of years we seemed to develop just fine without them, just by passing information along? How did we get along? By dying of dysentery and typhus and believing that the sun went around the earth. Being uninformed isn't a guarantee of failure, but it's a good way to ensure you don't make more intelligent decisions. Data is important. Also the OP is ostentatiously a "they don't listen!!!" whine. Don't you see that? Friendship is the best ship Sabriz for CSM go go go
Pok Nibin Filial Pariahs 590	Posted - 2015.03.10 07:00:55 - [10] - Quote Sibyyl wrote: Serene Repose wrote: How in the world did we get along before databases? Those tens of thousands of years we seemed to develop just fine without them, just by passing information along? How did we get along? By dying of dysentery and typhus and believing that the sun went around the earth. Being uninformed isn't a guarantee of failure, but it's a good way to ensure you don't make more intelligent decisions. Data is important. Also the OP is ostentatiously a "they don't listen!!!" whine. Don't you see that? "We?" The Mayans tracked Venus while the West believed Aristotle. And, dysentary/typhus...anthrax/cholera? People are still dying of that today. We all didn't. What you call "uninformed" Serene is calling different information. You infer data trumps the anecdotal, she disagreed, you infer she's ignorant and by so doing prove her point. Well done. +1 for math. Dont fight it; Rejoin your Amarrian patriarchs; You know you want to.
Vyl Vit 1083	Posted - 2015.03.10 07:05:34 - [11] - Quote Language is a virus from outerspace. Hearing your name is better than seeing your face. By the way, "whine" is a word that gets used a lot these days. Same with "troll". We tend to pull those out for those with whom we disagree, no? Anyone with any sense has already left town.
Sibyyl Gallente Federation 23783	Posted - 2015.03.10 07:10:48 - [12] - Quote Pok Nibin wrote: And, dysentary/typhus...anthrax/cholera? People are still dying of those today, as well as a litany of things that were unheard of pre-high tech. Arguing we've made no medical progress? We have a higher life expectancy as a result of technology. You can argue about diseases till you're blue in the face but the argument is disingenous. Quote: We all didn't. That's significant, but you managed to not mention that fact. What you call "uninformed" Serene is calling different information. You infer data trumps the anecdotal, she disagreed, you infer she's ignorant and by so doing prove her point. Well done. Data does trump the anecdotal. Informed decisions on a large scale are not made on anecdotes. That is by definition poor decision making. And I'm not sure how disagreeing with her is the same as calling her ignorant. Defensive much? Friendship is the best ship Sabriz for CSM go go go
Pok Nibin Filial Pariahs 591	Posted - 2015.03.10 07:14:02 - [13] - Quote Sibyyl wrote: Pok Nibin wrote: And, dysentary/typhus...anthrax/cholera? People are still dying of those today, as well as a litany of things that were unheard of pre-high tech. Arguing we've made no medical progress? We have a higher life expectancy as a result of technology. You can argue about diseases till you're blue in the face but the argument is disingenous. Quote: We all didn't. That's significant, but you managed to not mention that fact. What you call "uninformed" Serene is calling different information. You infer data trumps the anecdotal, she disagreed, you infer she's ignorant and by so doing prove her point. Well done. Data does trump the anecdotal. Informed decisions on a large scale are not made on anecdotes. That is by definition poor decision making. And I'm not sure how disagreeing with her is the same as calling her ignorant. Defensive much? Oh yea of the Heavily Indoctrinated. Your religion is as invalid as all the rest. Dont fight it; Rejoin your Amarrian patriarchs; You know you want to.
Herzog Wolfhammer Sigma Special Tactics Group 6399	Posted - 2015.03.10 07:21:43 - [14] - Quote Some may argue that databases may not be the sole source. But I don't think who complains the loudest on the forums is a good source either. Bring back DEEEEP Space!
stup idity 85	Posted - 2015.03.10 07:24:44 - [15] - Quote [quote=Kuga] Overall community sentiment should play a substantial role in decision making and help guide developers to issues where their database information alone may be inadequate. /quote] Measuring "overall community sentiment" would, if done in any way that allows drawing educated clonclusions from it, result in statics. Everything that can be quantified can be put together into a statistic. And a question for you: how do want to know what the "overall community sentiment" is anyway? By looking at the forum posts or even reddit? A very minor percentage of EVE players are actively or passively using the forums, an even smaller percentage is looking at a specific topic and only the smallest amount of players (and each with their ten to twenty alts) is posting their "sentiment". Add the fact that people are more inclined to post when they want to complain. (And all those numbers will result in statistics, again.) You are right, however, by saying that you have to use your brain when interpreting data. But that's about the same level as having to pull your pants down when sitting on a toilet to take a dump. I am the Herald of all beings that are me.
Benny Ohu Chaotic Tranquility Warp to Cyno. 4310	Posted - 2015.03.10 07:24:56 - [16] - Quote observation/data point: Serene Repose wrote: "Well, Ug did it and he's dead now." hypothesis: ug's dead because he did that goal: determine if doing that killed ug and if doing that will kill others method: a trial group consisting of the tribe's least sexy cavepeople will keep doing that. a control group consisting of the handsomer cavepeople won't do that observations: a lot of the trial group died after doing that. however, this may have been due to the violent clubbing used to convince them to start doing that findings: inconclusive notes: this was an excellent way to get rid of that bastard ur
Serene Repose 2363	Posted - 2015.03.10 07:27:20 - [17] - Quote Benny Ohu wrote: observation/data point: Serene Repose wrote: "Well, Ug did it and he's dead now." hypothesis: ug's dead because he did that goal: determine if doing that killed ug and if doing that will kill others method: a trial group consisting of the tribe's least sexy cavepeople will keep doing that. a control group consisting of the handsomer cavepeople won't do that observations: a lot of the trial group died after doing that. however, this may have been due to the violent clubbing used to convince them to start doing that findings: inconclusive notes: this was an excellent way to get rid of that bastard ur I love this one. Ug dies jumping off a cliff. To determine if that was the "actual" cause of death, send half the village over the cliff and get a statistical analysis. Sounds like how the priesthood grabbed ahold of the real estate market - appropriating all the empty huts...in the name of "Science"! Sibyyl, cool your jets. I was just farggin' wichoo. Treason never prospers. What is the reason? Why, if it prospers, none dare call it "treason."
Benny Ohu Chaotic Tranquility Warp to Cyno. 4310	Posted - 2015.03.10 07:29:42 - [18] - Quote whoa whoa whoa it wasn't half the village there weren't very many sexy cavemen to use for the control group. most of them were involved in the trial
Benny Ohu Chaotic Tranquility Warp to Cyno. 4310	Posted - 2015.03.10 07:31:19 - [19] - Quote and that's the story of how humanity is as damn fine-looking as it is today the end
Sibyyl Gallente Federation 23784	Posted - 2015.03.10 07:42:14 - [20] - Quote Serene Repose wrote: Sibyyl, cool your jets. I was just farggin' wichoo. Geez, woman. Nothing but love in my words.. nothing but love. Friendship is the best ship Sabriz for CSM go go go
bigman111222 Internet Terrorists SpaceMonkey's Alliance 16	Posted - 2015.03.10 08:25:15 - [21] - Quote Serene Repose wrote: Statisticians use statistics much like a drunk uses a lamp post; more for propping themselves up than for illumination. This is an amazing comment It's often thought the most common lie is statistics.
bigman111222 Internet Terrorists SpaceMonkey's Alliance 16	Posted - 2015.03.10 08:26:01 - [22] - Quote fix this double posting stupidness ccp
Thebriwan LUX Uls Xystus 216	Posted - 2015.03.10 08:37:36 - [23] - Quote Kuga wrote: Has the excessive usage of databases been clouding the judgement of those responsible for making decisions to the point where issues are being overlooked or eschewed in favour of balancing spreadsheets? Simple answer: No. The problem is not the data or the statistics. The problem lies in the question what the "community" feels. - This is something that can not be quantified. For one thing it is NEVER "The Community (tm)". Because you will find always someone who disagrees. Sometimes it is the majority. But mostly it's a loud and visible minority that wants this or that change, or not. To conclude: I think the actual approach from the devs is the right one: Fact and numbers. Not feelings.
Scipio Artelius The Vendunari End of Life 34631	Posted - 2015.03.10 08:54:42 - [24] - Quote Thebriwan wrote: To conclude: I think the actual approach from the devs is the right one: Fact and numbers. Not feelings. To a degree I agree with you. Objective data is the basis of good decision making for a great number of things. However, this is a game that we as players invest in emotionally in many respects. So while the feeling of an individual should give way to data, the feeling of the community as a whole is important data also. Opinion polls are an important thing for this very reason. The data produced helps quantify feelings and is extremely important in forming policy, in measuring success and identifying areas where better communication is needed. At the end of the day, CCP can use whatever data they want, but individually we all use our emotions to decide how much we invest in something. It's silly to ignore that when implementing change (or choosing not to change). Come Win At Eve - Join The Vendunari
Falin Whalen GoonWaffe Goonswarm Federation 760	Posted - 2015.03.10 13:33:05 - [25] - Quote There are three "lies," in the world: Lies, Damn lies, and Statistics. "it's only because of their stupidity that they're able to be so sure of themselves." The Trial - Franz Kafka-á
Vyl Vit 1083	Posted - 2015.03.10 13:44:33 - [26] - Quote Take a lemming population, for instance. Just because simply everybody is doing something, doesn't mean it's a good thing. Anyone with any sense has already left town.
Daerrol Furtherance. 56	Posted - 2015.03.10 13:45:27 - [27] - Quote Yes we do need data. Yes it must be the right data. The right data may be unnaccessible. the wrong data can hurt more than no data, but I don't think CCP is using the flat out wrong data. (IE: Using Industrial Activity to balance Lasers) I think it's worth mentioning in the new age that is comming, the one without 1000 man super cap fleets and multiple sov fights happening at once, combined with the slow but hopefully steady nerfs to T3 (Both Cruiser and Destroyer) the batlteship MAY seen a new role. We should wait for that to all come into effect, let it go a few months for metas to develop, then evaluate the position of the oft loved BS. I imagine a Napoc fleet sitting on a TCU will be be plenty hard for those pesky Tengus to deal with in contests. The 250km range on the Entosis however... that's another issue all together.
Kiandoshia Applied Anarchy SpaceMonkey's Alliance 2227	Posted - 2015.03.10 13:52:54 - [28] - Quote Sibyyl wrote: The only alternative to databases is checking for fate in chicken bones or consulting your totem animal. The right metrics need to be considered but not using metrics at all (like you seem to be suggesting) will lead to even more catastrophic game changes. If you mean "experience", we need enough data to drown out the anecdote factor that is so popular in politics and law. The OP isn't saying to ignore data but to look for more data that will tell you more than just numbers. Plain and simple numbers really don't tell 'the whole story' a lot of times. There are tons of other factors that should be looked at when attempting to balance anything. If all the want to to is look at numbers and go from there, CCP has to start giving us a meta rather than letting us figure one out every time.
Sibyyl Gallente Federation 23794	Posted - 2015.03.10 14:45:03 - [29] - Quote Kiandoshia wrote: The OP isn't saying to ignore data but to look for more data that will tell you more than just numbers. Plain and simple numbers really don't tell 'the whole story' a lot of times. There are tons of other factors that should be looked at when attempting to balance anything. If all the want to to is look at numbers and go from there, CCP has to start giving us a meta rather than letting us figure one out every time. Ki, I don't think CCP should ever design to "overall community sentiment" that OP wants. The vocal minority always ends up skewing things too far. Kuga wrote: Overall community sentiment should play a substantial role in decision making and help guide developers to issues where their database information alone may be inadequate. Friendship is the best ship Sabriz for CSM go go go
2Sonas1Cup 29	Posted - 2015.03.10 15:02:21 - [30] - Quote If I was CCP I would balance the game VERY easy. Heres what I would do: Pick 10 people, 4 the most active in pvp and FC leaders in the major null allainces with capitals experience (1 goons, 1 PL, 1 russian, 1 black legion) 2 of the same quality of people but from highsec (1 rvb (mangala?), 1 code (james?)) 1 person of the same quality from K space 1 person of the same quality from industry) Invite them over for a week to iceland with everything paid, sit dow together with the balance devs get it done. I guarantee you the mix and discussion and conclusion from these people on what to balance would be perfect.
BoBoZoBo Paragon Fury Tactical Narcotics Team 522	Posted - 2015.03.10 15:06:00 - [31] - Quote Yes. We also need good analysis to go along with it. The problem with too much data is that it becomes not only a crutch, but a cage. CCP needs to take a bit os a step back, they can't see the forrest through the trees; or even the damn tree through the leaves. Primary Test Subject GÇó SmackTalker Elite
Sibyyl Gallente Federation 23794	Posted - 2015.03.10 15:06:07 - [32] - Quote Design by null committee. No thanks. Friendship is the best ship Sabriz for CSM go go go
Lykouleon Trust Doesn't Rust Triumvirate. 1514	Posted - 2015.03.10 17:59:45 - [33] - Quote 87% of statistical analysis are actually interpreted wrong by the people that take them. 63% of all studies that relied heavily on statistics were later to be found to be fatally flawed in some way. 73% of statistics are made up on the fly anyway. Lykouleon > CYNO ME CLOSER so I can hit them with my sword
2Sonas1Cup 29	Posted - 2015.03.10 19:17:57 - [34] - Quote Lykouleon wrote: 87% of statistical analysis are actually interpreted wrong by the people that take them. 63% of all studies that relied heavily on statistics were later to be found to be fatally flawed in some way. 73% of statistics are made up on the fly anyway. Which proves you are statistically 72.3337870% precisely wrong, and your statistics have a 67% chances of being 99% incorrect.
ISD Ezwal ISD Community Communications Liaisons 4020	Posted - 2015.03.10 23:33:59 - [35] - Quote Kiandoshia wrote: Someone lock this quickly. As I see no reason to comply to this request, I'm afraid I will have to say no to that. ISD Ezwal Vice Admiral Community Communication Liaisons (CCLs) Interstellar Services Department
Slavealt Sheep Can Hear A Zipper From A Mile Away 13	Posted - 2015.03.11 05:25:00 - [36] - Quote CCP has failed hard at this for as long as I can remember. They constantly pull some graph or number out, while excluding tons of other factors, and buff/nerf/do nothing based on this. It usually results in OP staying OP for ridiculous lengths of time, or garbage staying garbage for the same time. It is really annoying, especially in cases where it is very clear that the number they're using is flawed.
Glathull Warlock Assassins 990	Posted - 2015.03.11 07:47:47 - [37] - Quote One of the things I consistently like about the data CCP chooses to show us is exactly how little it's been doctored up. We have some basically pretty raw numbers, some means here and there, and some rolling means. These kinds of numbers are not the numbers people generally choose to display when they have an agenda. This is grade-school type stuff. There is nothing going on here that I would call even remotely obfuscatory. If anything, I would like to see CCP doing more (or perhaps I should say showing more to us) of things that are perhaps a bit more advanced. I'd like to see a factor analysis around some of the variables mentioned in the OP to see if there are latent causes behind the raw data that could explain some of the variance, for example. I'd like to see some regressions to understand how much of the changes in the numbers can be explained by natural cyclic variability, and how much of the change is actually related to mechanics changes. But in any case, there are a couple of things you have to take into account. When people are using statistics to hoodwink you, you have to consider 1) the level of certainty claimed by the alleged perpetrator, and 2) the obscurity of the mechanism used to arrive at the supporting data. If someone comes to you and says they know for absolute certain that x and y are true (where x and y are totally contrary to intuition), and you can tell it's all true because of this new method of analysis that they just made up for this one purpose: guess what! You're getting lied to!!! If, on the other hand, people use very straightforward methods to arrive at some conclusion that's relatively weak, well . . . there's little room for duplicity there. CCP's use of numbers is closer to the latter case, in my opinion. First off, I don't think it matters at all what statistical methods they are using when their claims are so weak. All of anything that they ever say amounts to: "When we look at these numbers, we feel good about them. When we look at these other numbers, we don't feel so good about those." In seven years of EvE, I can't remember a dev blog that really made a stronger claim based on "the numbers." So I have to wonder, if no one is claiming any particular truth, where is the lie that these alleged statistics are telling? CCP has always struck me as a company with a philosophy-first approach to design that sometimes looks to the numbers to make sure they are not completely out there. It has never struck me as a data-driven company at all. And I don't mean that in a negative way at all. Facebook and Zynga are data-driven companies. They are companies that tweak every aspect of every interaction to maximize addictiveness based on behavioral psychological data. Blizzard is somewhere between what I see as two extremes with EvE and things like Dwarf Fortress on one end of the spectrum and Farmville and Candy Crush Saga on the other end. Sibyyl has already made what was going to be my second point far more eloquently than I was going to, so I'll leave that alone. OP sounded an awful lot to me like he was saying, "CCP, I want you to ignore the numbers I don't like and design the game around what I feel." But I will make a third point. CCP does already do a certain amount of what the OP most charitably could have meant. CCP appears, from my point of view, to design the game around a) a large, reasonably consistent philosophy, b) a strong desire to troll, and c) some numbers here and there to serve as big-picture guideposts. If that's not what the OP is asking for in terms of "extra-database-ical-ishy" data, then I can't imagine what reasonable thing is. I just think the OP doesn't like the consequences of CCP's design philosophy or the way that CCP loves to troll and is using the weakest possible intellectual approach to criticize Fozzie and Rise because when you throw around hardcore words like database, you can't be wrong. Right? I honestly feel like I just read fifty shades of dumb. --CCP Falcon Shut up, Anslo. --everyone
Rain6637 GoonWaffe Goonswarm Federation 30825	Posted - 2015.03.11 09:56:59 - [38] - Quote The problem I think EVE devs have is using statistics to alter the game, and then using statistics to validate those changes. So basically, they're not balancing a game, they're balancing statistics. You gain a sense of a thing when you use it in context (of the game). In contrast, statistics are disconnected from their context. I think this is why devs are out of touch. Because context. Help, I can't download EVE

Pages: 1 2 :: [one page]

COPYRIGHT NOTICE
EVE Online, the EVE logo, EVE and all associated logos and designs are the intellectual property of CCP hf. All artwork, screenshots, characters, vehicles, storylines, world facts or other recognizable features of the intellectual property relating to these trademarks are likewise the intellectual property of CCP hf. EVE Online and the EVE logo are the registered trademarks of CCP hf. All rights are reserved worldwide. All other trademarks are the property of their respective owners. CCP hf. has granted permission to EVE-Search.com to use EVE Online and all associated logos and designs for promotional and information purposes on its website but does not endorse, and is not in any way affiliated with, EVE-Search.com. CCP is in no way responsible for the content on or functioning of this website, nor can it be liable for any damage arising from the use of this website.