1. #1
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201

    [Guide] Benefit-Cost Analysis and how it helps you

    Hey Everyone,

    For me this is really a general issue, but lately I struggled very badly with identifying the profession and race I want to play in Guildwars 2. I have tried out several Quizzes in order to evaluate which character might be the right one for me. Unfortunately I haven't really been satisfied with them as they narrowed down results too quickly. So instead I browsed through the GW2 wiki, going through the traits and skills of all professions, just to switch my opinion whenever I went through another professions ability descriptions. They are just all so awesome!

    In this guide, I will try to explain a different approach to you, so that you will be able to make decision more easily. This method is applicable to nearly every situation, so it is not limited to economy and gaming. Please note: that all the figures I come up with in this thread are just examples to demonstrate the method I use. They are by no meanings reflecting the state of the game GW2(Guildwars2) nor my personal opinion.

    The BCA( Benefit-Cost Analysis)
    The BCA is a method that is mostly used in economical context. It supports your decision making with two simple steps. First, it helps you to find out your priority. Secondly it will compare your options according to this priorities. You expected more "Rocket Science"? Sorry to dissappoint you. This is basically precisely the same process our brain goes through a billion times a day. They key point here is focus.

    Evalueating your Priorities
    So for the first step, think of what are really the most important factors for you. As an example, we will apply this first step, to find out what makes a profession appealing for us. Now, please don't say; "a profession has to be fun!" Eventhough I agree with you that this is a top priority in a RPG, its not bringing us any further. The right Question to ask is "What makes a profession fun?".

    Factors could be(In no specific order): Mobility, Damage Potential, Survivability, Number of Abilities, Theme, Supportivity
    I will take these as examples. Please note that you can taky any factor that comes to your mind. Maybe it is importent for you how reliant a profession is on pets, or whether it is ranged or meele focused. It realy could be anything you can rate. I reccomend to use between 5 and 10 factors to weight. Less will make single factors too meaningful while more will just wash out the results, giving very low weighting to each and every factor.

    To compare the chosen factors, we are going to setup a table. Every factor is compared to every other one. A factor can be either less, equal ore more important than the other one he is compared to. You can either use 0, 1, 2 or -, 0, + to note this down in your table. When you have done that, you can add up the points for each factor and set the weight as percentage. Now you know your priorities.

    Mobility Damage Potential Survivability Number of Abilities Uniqueness Supportivity Total Percentage
    Mobility 1 0 0 2 2 1 6 16.667
    Damage Potential 2 1 2 2 2 2 11 30.556
    Survivability 2 0 1 1 1 1 6 16.667
    Number of Abilities 0 0 1 1 1 2 5 13.889
    Uniqueness 0 0 1 1 1 2 5 13.889
    Supportivity 1 0 1 0 0 1 3 8.333
    (Example Priority Evaluation)

    So what does this tell us now? Let us make some assumptions here. I seem to like very offensive characters, with mobility and survivability being equal, but only half as important as damage throughput. I am not so focused on the theme of my class. Even though I might have some preferences, I am okay with both, playing a bulky juggernaut or a fragile mage as long as it provides me with the most damage. I like a broad set of abilities, but a limited set is not a show-stopper. Support hardly is of any importance for me.

    Comparing Options
    Now we come to the most time consuming part, the actual comparison of options, in our example Professions. We are going to use our 6 chosen factors and their weights, to come up with a ranking for every profession. In order to make this ranking as accurate as possible, we have to base our assumptions on a ton of information. The better the information, the better the ranking. The ideal case was, to have played all the professions in all possible talent builds. But to be honest, if that was the case, you probably wouldn't want to go through this whole exercise. Instead, we will have to base our ranking on YouTube videos, articles on gaming sites, GW2 wiki and if you are lucky, some hours of beta testing. If you really want to make a well-founded decision at the end of this process that makes you happy on the long turn, invest time to do this. If you just want to try this out you can go with your assumptions.

    To make this ranking, we need another table, containing our options and the factors on two different dimensions. We will create a ranking for every single factor, multiplying it with the weighting we created in a first step. This will give you a number of points for all the professions, allowing you to identify which wants suits your requirements best.

    Elementalist Engineer Guardian Mesmer Necromancer Ranger Thief Warrior
    Mobility 4*16.667% 2*16.667% 3*16.667% 1*16.667% 5*16.667% 7*16.667% 8*16.667% 6*16.667%
    Damage Potential 7*30.556% 4*30.556% 1*30.556% 2*30.556% 3*30.556% 5*30.556% 8*30.556% 6*30.556%
    Survivability 3*16.667% 5*16.667% 6*16.667% 1*16.667% 8*16.667% 2*16.667% 4*16.667% 7*16.667%
    Number of Abilities 8*13.889% 7*13.889% 6*13.889% 5*13.889% 3*13.889% 2*13.889% 1*13.889% 4*13.889%
    Uniqueness 5*13.889% 7*13.889% 6*13.889% 8*13.889% 4*13.889% 2*13.889% 3*13.889% 1*13.889%
    Supportivity 5*8.333% 6*8.333% 7*8.333% 8*8.333% 4*8.333% 3*8.333% 1*8.333% 2*8.333%
    Total 5.52783 4.83337 4.05558 2.97219 4.38894 3.83338 5.08341 4.86118
    (Example Profession Comparison)

    This profession comparison leads us to the conclusion that a Elementalist would probably be the best way to go and that I should not start with a Mesmer. What is important to note is, that I didn't just get the Profession with the highest damage ranking . I got the one, which covers my personal priorities overall best. And that my friends, is precisely what puts this method ahead of class quizzes and so forth. It takes your personal interests into consideration.

    Please note, this method is not perfect. It has its pros and cons. Also it is not saying you : "Elementalist is the only profession that will bring you fun!" it is meant to show you tendencies, which are more reliant if you go through the whole process as precisely as possible. Be honest with your rating. Even if it means you have to downgrade a profession you otherwise like. If you are absolutely unhappy with the result, you probably have not considered the factors that really matter to you.

    Many thanks for reading. I hope this will help you to make decisions in the future. Comments and feedback are much appreciated. I apologise for grammar mistakes and my bad writing. I am still working on it.

    Reccomended Factors and Rankings
    This section contains factors and rankings that have been propsed by the members of MMO-Champion. I tryed to rank the professions based on information from GW2 wiki and polls in this forum. You may take them and apply your personal factor priority scales to it. I hope it helps you. Many thanks go to all the people who have contributet to this by adding a lot of value to a fruitful discussion.

    Mobility: This ranking is determined by the amount of spells that grant a profession stealth, swiftness, evasion or let you teleport, rollback, charge, leap or pull you character to your targets location. Also % movementspeed increases are covered.

    Number of Spells: This ranking is determined by the number of spells available to a profession. The information was taken from the GW2 wiki. Chains count as one spell per step. Abilities that grant you access to new spells are counted as one plus the amount of temporarly grantet spells. The Reason for the Ranger being ahead is the amount of spells that come with his pet choice.

    Unpopularity: This ranking is determined by your votes, given to a poll that asked for your mains profession choice after the last beta weekend.

    Supportivity: This factor is ranked by the number of skills available to a profession that have a positive direct influence to allies aka give a boon to them ore removing negative conditions from them. Traits are not added as too dificult to rate. Please note, that a Ranger is not worse a supporter than a Guardian. He is just more limited in spell choice to fill the role of a supporter.

    Elementalist Engineer Guardian Mesmer Necromancer Ranger Thief Warrior
    Mobility 5 2.5 1 6 2.5 7 8 4
    Number Of Spells 7 6 3 1 2 8 4 5
    Unpopularity 2 3 4 8 5 6.5 6.5 1
    Supportivity 5 6.5 8 6.5 3 2 1 4
    (Prepared ranked factors that can be used)


    Update Notes:
    2012.08.07 13:30 -> Exchanged the factor "Theme" with "Uniqueness" and adapted ranking accordingly
    2012.08.08 -> Added a section with reccomended factors that have been ranked propperly, based on data from gw2 wiki or MMOC polls.
    Last edited by Migrin; 2012-08-08 at 12:01 PM.

  2. #2
    Warchief Kallanna's Avatar
    Join Date
    Aug 2010
    Location
    London UK
    Posts
    2,048
    Woah this is so interesting... but so very complicated at the same time. Wouldn't it be easier to just... try them all out and see what feels right?
    Still thank you very much for a very interesting read! =)

  3. #3
    Thank you.

    Thank you very much. I took a degree in economics but i couldn't remember this one. Good call.

    I am the kind of people who likes to "numberize" everything, so this will be a good help to choose my final/main profession.

    Did your factor ranking for every profession was worked for you or it's just an example with "random" numbers?
    Last edited by promoss; 2012-08-07 at 09:45 AM.

  4. #4
    The problem with your analysis is that your profession ranking seems completely arbitrary. Elementalist and Mesmers (probably the most original class in the game) having lowest theme score and warrior (arguably the most 'standard' class) having one of the highest? Or even more ridiculous, Guardian has the lowest damage potential? I know you said that the ranks here are only examples, but they are the most important part of the story, without it the method is practically useless. And compiling such a table in a meaningful way is a huge time investment.

    An arguably more flexible method is to assign probability distributions which reflect our uncertainty about the current state to profession ranks instead of numbers. One could start an online questionnaire where players would be asked to rank classes depending to the criteria you outlined here and take the overall result as a sample to compute these distributions from. Could be an interesting project.

    BTW, your result should not be interpreted as 'you should play Warrior', but rather, 'do not play Mesmer'.

  5. #5
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201
    Hey Guys, thanks a lot for the replies. @Kallana You are absolutely right. More importantly, it would even be more fun. Unfortunately, I do not have access to the game, so gambling with numbers and buid editors is all I can do. Still I am glad you liked it.

    @Promoss: Glad you like it! This is just a set of random figures. My personal tables look different

    @mafao: Thank you for your feedback. With Theme I don't reference to the uniqueness or originality of the profession, but rather on my personal taste. In the example I rated the 3 scholar profession lowest because my fictive me, does not like cloth armor for example but digs the bulkier type of character wearing plate and wielding heavy weopons. But I agree with you, Mesmer and Enginneer would have had a much higher rating if it was about uniqueness. In gener I don't do justice to the mesmer with this example. I didn't plan to have it as one sided. Will take it up as a To-Do.

    About the ranking on the table, honestly what I put in this example table is complete bs. I totaly see your point and I would totaly love to create a more accurate ranking as this is precisely the disadvantage of this method. I hope you can understand that it is extremely hard for me to get full picture without access to the game. One possibility though would be, to base the ranking only on figures that are precisely countable.

    The number of abilities is a great example. You could also rate Suportability by the number of abilities that effect other players and so forth. Possibly even better the maximum number of abilities that effect other players in one single build. Your ideas with uncertainty are great and I thought about it too. But at the and, the factors you use for the ranking are very personal. So how could I come up with factors that are right for everyone? Making a survey about it is a great Idea though and I will keep this in mind if there will be more people that are interested in this thread, I might give it a go.

    I am very thankful for rankings from your perspective with the same or different factors. This really improves my limitations regarding information sources.

    Thank you guys
    Last edited by Migrin; 2012-08-07 at 11:20 AM.

  6. #6
    Can anyone help me please? Especially with the profession factor classifications..

    I was between Engineer, Warrior and Guardian.

    This is what i get: imgur.com/x7wQE

    Also, the special factor is my natural tendency to the profession (e.g. i like a lot the multiple conditions engineer is capable to inflict).

  7. #7
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201
    Hi promoss. If you have three professions with precisely the same amount you have to take more factors into consideration. But it seems that you like engineer the most. I mean, come on you have created a special factor for it

  8. #8
    Oh wow. Super interesting. Great OP.

    I shall reward you one day for posting awesome.

  9. #9
    Mmmm, nice... I've been looking for a new model to apply to other things.

  10. #10
    Whoa interesting read.

    buuuut XD my margin between classes is too low making me just as the same confused hahaha XD.

    Because I want all of them O.o!

  11. #11
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201
    Honestly, I can very well understand how you want to play all of them. If you want to get a more clear result, I would reccomend to a) make sure youre ranking is as accurate as it can be, b) remove your weakest factor from the model or c) add a new factor that matters to you. I hope that helps.

    ---------- Post added 2012-08-08 at 02:02 PM ----------

    The thread now also contains the feedback given by forum members on what makes a profession fun for them. Note that this is still work in progress.

  12. #12
    This is awesome. Number crunching time.

    My part in this story has been decided. And I will play it well.

  13. #13
    Your graph is severely flawed by the simple fact that all those numbers will change depending on the build, and there really isn't any objective way to actually measure those yet, besides "number of abilities".

    Here's something, you gave Thief an 8 for mobility/damage. You don't factor in the fact that a thief must sacrifice damage potential in order to be mobile (which is why they're terrible at the moment in competitive PvP for all but supportive roles, which for some reason you gave the thief a 1 on...). An Elementalist can be supportive, mobile, or have a lot of damage at the same time with minimal effects on their ability to do the other two, within the capability of their build of course. Similarly, an engineer can keep near 100% uptime on Swiftness and be able set up and burst down a person they see at a moment's notice, yet you gave the engineer and elementalist a lower score for both.

    Fact is, you can't "create a ranking for every single factor". It's just not plausible. It's also worth mentioning that in order to get accurate numbers, you sort of have to play the profession/build in question in the first place, which kind of defeats the purpose of this graph.
    Last edited by Larynx; 2012-08-08 at 07:28 PM.
    6.0 Disc Priest Guide http://i.imgur.com/QbCmDc2l.jpg

  14. #14
    I think I will simply test all the classes and see which one I find the most fun.

  15. #15
    Very interresting read.
    The BCA is an important step in the design of consumer products.
    However, I have to agree with most people on the second part of your analysis.
    To get an analysis like this as accurate as possible, you need to define your factors as clearly as possible.
    The best way to do that is to translate them into something that can be specified with units

    To take mobility as an example:
    Mobility could translate to an uptime percentage (amount of time mobility boosting abilities were up divided by the total amount of time the fight lasted times 100%)
    Of course this means you would need additional information (for example, the average duration of the fight, reliability on build etc.), but that's just a matter of gathering data (through literature studies or actual experiments).

    This can be done for most things you mentioned (although the general feel of a profession could not be specified in such a way).
    However, doing this does require some degree of insight and mathematical skill.
    Therefore, this analysis might be too complicated for some people, which is the real downside to it.

    My advice: if you want to use the BCA, give it a try, but think things through.
    There's more to those factors than just their name.
    And if you have trouble defining your factors or converting data to results, ask for help.
    There are always people around who are willing enough to help you.
    Accusing someone of breaking the forum rules is backseat moderating.
    Accusing someone of backseat moderating is backseat moderating.
    Therefore, due to its nature, backseat moderating will lead to nothing more than circular discussions.

  16. #16
    thank you for the work you put into this, its really interesting

  17. #17
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201
    Hey Guys, Thank you very much for reading and your feedback. It means a lot to me. There are some things I would like to clarify though. This thread was mainly created for those, who would like to play the game but do not have access to it. I mean seriously, why would you spend time on excel spreadsheet if you could play GW2 instead?!

    Quote Originally Posted by Larynx
    Your graph is severely flawed by the simple fact that all those numbers will change depending on the build, and there really isn't any objective way to actually measure those yet, besides "number of abilities".
    Thanks for this post. Totaly agree, that was going through my head as well. In fact, before I created this thread, I startet using it to determine what profession to play. Afterwords I started comparing different builds of the same profession. It is absolutely possible to do this if you have all the figures. It is just a bloody hell of lot of effort. That is why I decided to go the easier way and add descriptions to each factor. Mobility is not really saying how mobile a profession/build is. It only says how many abilities a profession has in total to achieve mobility in a build. Same goes for Suportivity and so forth. This ranking does not say "Guardian is more supportive than a Thief". It says "A Guardian has a larger amount of abilities he could add to his build to support his allies". This is a big difference. But still, for someone who really is keen on playing a supportive character or build, this indicates that the Guardian might be a better choice, because the variation for support builds is probably bigger than it was as a Thief.

    Quote Originally Posted by Larynx
    Fact is, you can't "create a ranking for every single factor". It's just not plausible. It's also worth mentioning that in order to get accurate numbers, you sort of have to play the profession/build in question in the first place, which kind of defeats the purpose of this graph.
    My main intention was to provide a new approach for profession evaluation that is fun and not to come up with a complex, scientific accurate model of the professions but what you are saying is very important to understand. This is also why I reccommend using your own factors. You may not be able to calculate how mobile a profession is. But you can always rate your feelings. Let us say, you are a big fan of the Charr. I can say that because of your avatar and because Charr are just pretty damn great. Now let us also assume, that you love role playing and you want to choose a profession that, among other factors, fits the lore of this race.

    This is precisely where this method shines without making it a lifetime project. You can not calculate by what percentage a profession suits to the lore of a race, but you can tell what feels right for you and rate it. For me it might be the engineer that suits the most and the Mesmer that suits the least. For you this could be totaly different, but that does not make a factor flawed or weak. It makes it subjective and tailormade and strong.

    Quote Originally Posted by Shortcut
    My advice: if you want to use the BCA, give it a try, but think things through.
    There's more to those factors than just their name.
    And if you have trouble defining your factors or converting data to results, ask for help.
    There are always people around who are willing enough to help you.
    Hey mate thank you very much for this. I really did not expect that amount of attention and did not plan to push it that far as GW2 is coming pretty soon and we will al be busy with more epic things than this spreadsheet. Everybody who is willing to add something to this is very much invited to do so and every post, pm or what so ever with feedback or your own proposals for factors is much appreciated!

  18. #18
    It's important to know what you like for this to make an accurate analysis right While picking up a profession (yet again) I noticed there is stuff I like "on paper", but when backtracking what I had most fun playing very different patterns emerge. It's helpful to go game by game and analyse what you liked most playing without being influenced by what you believe you like playing. It can make you score class priorities more accurately.

    My part in this story has been decided. And I will play it well.

  19. #19
    The Patient
    Join Date
    Jul 2012
    Location
    Zurich, Switzerland
    Posts
    201
    Quote Originally Posted by Repefe View Post
    It's important to know what you like for this to make an accurate analysis right While picking up a profession (yet again) I noticed there is stuff I like "on paper", but when backtracking what I had most fun playing very different patterns emerge. It's helpful to go game by game and analyse what you liked most playing without being influenced by what you believe you like playing. It can make you score class priorities more accurately.
    Absolutely true. One of my art teacher thought me to keep my horizon as wide as possible. This is usally achieved by not following your personal patterns.
    Last edited by Migrin; 2012-10-26 at 07:36 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •