Talent guides of the future: Is all choice good choice?

Okay, so it’s no secret that I spend a LOT of my time coming up with talent guides. I’ve done talent guides for leveling. Talent guides for end-game raiding. Analysis of what talents I think are good. Analysis for what talents I think are bad and should be changed. I also apologize in advance for just rambling theoretically about guide writing and talent choices. None of this reflects actual things we know are coming, but I want to talk about the implications & problems with a possible design philosophy that I don’t necessarily agree with.

While I originally passed over one of GC’s posts last week, there was something that caught my eye today that I wanted to nitpick a little bit. The original post you should have all read is here. Context matters, so you should go back and read the full blue post. I’m not going to repost the whole thing. Instead, the quote I’m interested in particular says:

Because you earn passive bonuses just for spending points in the tree, those fun, niche or utility talents won’t seem as expensive as they do today. We want to create a lot more choices where you are choosing utility vs. utility. We want to see far more “cookie cutter” build guides that say “Spend the last 5-10 points where you want.”

First, I like the idea of the passive bonuses. It takes away the need to spend the majority of our points just on things that essentially say “increases damage/healing by X”. That’s a great goal to work on.

On the surface, more choices seems like a good idea. The idea of more utility talents and more options seems like a neat goal.

However, where I have problems is his last sentence about writing guides that tell people to just spend 5 or 10 points wherever they want. When I do that in one of my guides, it usually means that none of the options are good anymore and you can just dump some points into basically “garbage” talents, or that the choices just aren’t meaningful so I’m basically giving up on trying to make a decision.

Talent choices should still be meaningful, and there SHOULD still be choices. However, there is a right and a wrong way to allow for choice. All the choices have to theoretically not just be throw-away talents that people think are bad, but put left-over points into anyway. I don’t have a problem with presenting 3 or 4 different talent specs, like I do for my healing guide right now. I don’t mind 5-man healing specs looking different than 25-man raiding specs looking different from PvP healing specs. I don’t even mind if the talent trees allow enough flexibility that people can disagree about the right spec to have, or can choose not to follow the cookie-cutter build if they want.

However, I do mind the possibility of either having talents being so desirable OR undesirable that we can’t present a good “cookie cutter” spec in our guides. The goal should be, instead, for people to not WANT to follow the cookie-cutter builds, and for people to find neat paths of their own through the talent trees in places where it better matches their particular play-style.

At the point where we say “spend the last 10 wherever you want”, I feel like I’m not doing my job as a guide writer. A “best” spec should always emerge out of guide writing, because that’s the entire point of what people like me do. In testing phases, I’ve been known to spend hours jumping off cliffs trying to reduce my health enough to test HOT healing in a controlled environment. Having HOT overheals show up in the combat log was a significant improvement to my personal testing quality of life, because I’d do practical testing rather than using spreadsheets and simulations, so having the #’s show up with less effort meant I could waste less time running back up hills.  Guide writers go to great lengths to be confident in the talents that they recommend to people, and being 10 points short isn’t going to be something that the guide writers will ever do. That’s just part of guide writing: fill up the talent tree & present it for other people to decide whether they follow you or not.

If we are investing in utility, it should be useful utility. Giving us bonuses that seem neat but have no real impact on our playstyle are just throw-away talents, and it’s usually at the point of throw away talents where we suggest just to dump points wherever. For moonkin/cat druids in a PvE raiding environment, our job is to be doing damage to the boss. We want to be investing points in talents that are going to help us do our job to the point where we aren’t just going to get laughed out of guilds like we were in the original days of the game. It doesn’t matter if we can daze things unless daze is an important part of the boss mechanic. In the days of redundant buffs, giving people more options to spec into redundant buff talents (that may be wasted in some circumstances) isn’t really an improvement to the trees.

The goal shouldn’t be to have guide writers suggest that people just dump points wherever. At that point, why bother writing the guides at all? The goal (instead) should be to get guide writers to have to write about decision points in the trees and how to decide between options based on their playstyle, which I do in my resto druid healing guide right now. I offer multiple decision points & help my readers be informed about their choice between options for different PvE healing build choices. For example, you may have to choose between having Celestial focus or not in a healing build (which is an interesting choice). Another interesting choice ends up being whether or not to pick up several other talents (empowered touch, living seed, revitalze), – depending entirely on whether or not they are good for your particular playstyle. I think it would be interesting for the talent choices to differ with slight differences in playstyle, for example – for there to be clear PvP talents versus clear PvE talents for DPS specs, OR multiple ways to come up with good DPS specs that alter how you do your DPS rotation (ie. mangle-focused versus shred-focused builds for cats, or something like that).

The good talent trees of the future are ones that allow for choices to be made based on playstyle, where there will still be clear guidelines for guide writers to follow about what are good choices for a certain playstyle. I like the idea of “choose your own adventure” talent trees, but there is always going to be one right path down talent trees for people who are trying to maximize their DPS, healing, or tanking potential – and if the talent points aren’t helping the player accomplish their goals, then people shouldn’t be investing points in those talents in the first place. I would rather feel like the talent trees are too full than feel like they are too empty, and the only time I’m going to throw my hands up and force my readers to choose totally on their own is when the trees are too empty and the choices aren’t meaningful.

More often than not, the talents we skip are ones that aren’t meaningful to our roles. So, improved barkskin is a really interesting PvP talent, but I would never pick it up for PvE. If the choice for me was between imp barkskin and imp tranquility, I wouldn’t care because both of them I consider to not be useful talents for my healing role. For my moonkin, if I had to decide between genesis & dreamstate, it wouldn’t be a meaningful choice, since neither of them would actually increase my DPS potential by enough to care.

So, the choices of the future need to be meaningful, without swinging the other way and feeling like the talent trees are so full of “bloat” that we can’t make a decent enough spec out of it. There has to be a clear pathway through the tree for a particular playstyle and role because that’s how we make sense and meaningfulness out of our choices (IE. I choose a tank healing build because I want to tank heal – OR, I choose a Panzerkin tanking moonkin/feral build because I’m a crazy tanking moonkin, OR I want to be a restokin that picks up both moonkin & tree form). We want the talent choices to meaningfully reflect a play-style, and for our investments to feel worthwhile based on the spell combination and role that we have chosen.

The choices should still be meaningful, and the guide writers should still be able to make enough sense out of it to come up with the carefully crafted “cookie cutter” builds. Even if we have to come up with multiple builds, and points where we ask people to make choices, these trees will still be complete. Figuring out the “best” specs to have for a given role & playstyle is part of the game that we play – and it’s something I enjoy doing.

The goal should be to have interesting choices, not to take away the role of cookie-cutter specs.

Posted in Cataclysm, Druid - General

24 comments on “Talent guides of the future: Is all choice good choice?
  1. aramis says:

    LOL…I have Imp. Barskskin in my PVE build. Cause let me tell you, it always seems to be that when a mercenary in Heroic Halls of Reflection breaks free from the aggro of my tank to pop behind me and try to gank me, or when Marrowgar decides to chain flame-spike-whirlwind me barkskin is a lifesaver, literally (It’s always me! I swear, Marrowgar hates trees!).

    I agree with you though. “10 Points wherever you want”. Seems a bit off if there’s nothing you can do with 10 points that you can just sprinkle them wherever.

    The guide writers will definitely have a challenge ahead of them. Right now as it were, there’s at least five different builds for Restos with just the points we have that we can shoot off the top of our heads that go with different playing styles. Can you imagine when they are retooled that we have space to work with up to 10 extra points? Especially if they are “utility” talent points. Because as you’ve stated, you aren’t simply going to stop with 10 points left and say “yeap, kk, that’s it, put the rest where you thnx bai”. No, someone like you will FIND a place for every single point and then say “this is the decision you have to make…do you want this or this and if you want this, you’ll probably go for that and this is the recommended build that results,” and end up with like 12445 builds. (An exaggeration, but you get the point, lol…yay for hyperboles!).

    Things are about to get very interesting. Ain’t it neat? 😀

  2. Spinks says:

    I don’t know. I can easily imagine writing a talent guide that says, “Look, these are the core talents. You really want these,” and then a brief description of the strengths and weaknesses of the other ones to let people make a reasoned choice.

    “The goal should be, instead, for people to not WANT to follow the cookie-cutter guides,”

    The only way that can be true is if there are no true cookie cutter guides. The cookie cutter choice becomes popular because it’s flat out better for some aspect of the game.

  3. Ardol says:

    I think the feral tree in its current incarnation is a pretty good example of what GC wants for all talent trees. I can invest in all of the necessary tank talents and still have points left over to put in talents that will help me while solo-ing (like KotJ and Predatory Instincts). I also recall GC saying that the warrior’s protection tree and the paladin’ retribution tree were good examples of what they were going for with the talent trees. In those trees, you can take all of the necessary talents and still have points left over for fun talents. What I am saying is, GC’s vision is already a reality, and it works.

  4. Lissanna says:

    @spinks – Yeah. I think that wasn’t the right phrasing, either.

    @Ardol – The risk with “utility” talents is that you can end up with some talents that just aren’t actually all that fun, but instead end up being filler.

  5. LXj says:

    I think you’re taking GC’s words too literally. Like when they said “Bring the player, not the class” — everyone started shouting “AMAGAD, what a horrible design, now you can clear any raid with 25 paladins!”. No, it wasn’t the idea of BTPNTC, and good luck on Lady Deathwhisper without good variance of melee/ranged dps and buffs. I think there definitely going to be interesting choices in talents in Cata, and their intention is not to fill our trees with meaningless crappy talents

  6. Lissanna says:

    LXj – crappy talents in the trees are always a risk. So, it’s okay for me to point out that risk, even if their goal is to not just have filler junk in the trees. 😉

    We want GOOD options and hard decisions without having so many great “must have” talents that the trees feel too bloated. This isn’t actually easy to do at all.

    I also maybe rambled on a little more in my post than I meant to… However, I was also trying to stir up controversy & conversation. Just posting something everyone agrees with can be boring. 😉

  7. Kaelik says:

    The thing is, I think WoW was designed to have talent trees be full of choices and tradeoffs. Except making a number of options that are equally good is difficult in practice, so the theory crafting folks discovered which options were best, and that’s what people who want to be on top of their game use. It’s not that there is a lack of options, it’s just that the other options aren’t as good.

    The trick is to have more talents that help in ways that don’t directly compete with each other. Two talents that both offer some kind of DPS increase will probably result in one increasing DPS more than the other, so one will be a better choice. On the other hand, if you have two talents, one that increases personal DPS and one that buffs GROUP DPS (or even better, group survivability), you offer a real choice to players.

    A perfect example of this is the resto druid talent revitalize. The choice to put points into that talent is a balance between increasing personal healing ability (which this talent hardly helps at all) and giving a little extra boost to group DPS and endurance. Taking it is a choice that resto druids have to make for themselves, and while we can all understand the trade-offs, a cookie cutter build can’t say absolutely take it or not.

  8. Lissanna says:

    The problem with helping group versus self is that for specs like moonkin & cat druids, being a walking stat-stick for other people is actually a really not fun playstyle if it means sacrificing personal DPS to just be invited for your buffs.

    The utility of bringing neat & interesting tools to your group works better for healing specs than for DPS specs. For a DPS spec, the answer is almost always to go with what is going to increase your ability to contribute DPS to the raid, ESPECIALLY now that no one ever brings unique buffs that aren’t redundant with other people.

  9. scaresome says:

    This is a great conversation, I’ll come back to read more comments. Plus, I follow the guides on this blog.

    My imagination must’ve gone awry.
    I thought by making fewer passive talents that we’d get “core cookie cutter” builds depending on play choice like tank vs raid builds. And then, at the end … the last three would be something totally fun. Like maybe you could have a “tree on fire” talent that would give everyone a fire dps buff every five minutes. Or, “roots from below” that prevents all movement impairment. Or, “fruits and nuts” that drop a sudden speed buff or makes it slippery like the hunter’s trap.

    So, you’d have a great, solid healing build; for example. Then a quirky but useful talent to choose at the end. Necessary? No. Fun, yes!

  10. Robert says:

    I think what they’re after is more like what the Marksman Hunter tree is like, where there’s “technically” a best spec, but it really makes little difference where you put your last few talents–the top 5 specs are all within 2% dps of each other.

  11. Hamlet says:

    I think some of this is a bit of a pipe dream–specifically, the idea that we’re going to get away from there being one identifiable optimal spec (modulo a few weak filler points) for a given situation.

    I think what you say in the first half is right. We (guide writers and the strategy community in general) will always try to figure out what’s best, and it’s not a good thing for the game if some talents don’t emerge stronger than others. This doesn’t have to mean there’s only one reasonable spec, since different talents are better in different situations. My guides both in fact take exactly the approach you describe–list the talents will be used in any reasonable spec, and then describe the specific uses of the remaining options:

    So that would be nothing new–talent variation based on situation. But then you go on to talk about “playstyle.” Let me tell you that I have a negative association with that word: 9 times out of 10 it’s used by a first-time poster on EJ in discussing the special-flower spec they cooked up, or as a justification for using bad talents. It’s a pretty solid predictor for a quick ban, actually. But while I’m warning you in advance about my mentality here, I think it’s all for good reason.

    DPS classes accept that there’s a better and worse way of doing things. Well, maybe there are “playstyle” champions on the WoW general forums, but not anywhere where anything actually gets done. I won’t explain why that is–it’s obvious. Now, how different is healing really? I’d submit that it’s more complex to figure out precisely what’s ideal in a given situation (since the metric is not as clear-cut as “do the maximum points of damage and don’t die), and there’s more variation between situations, but the goal is same. Find the most effective way to get the job done. And there will be pros and cons and disagreements, fine. But this is all the case currently, and it will all be the case no matter what talent changes go in the future.

    Now I know this all depends on your perspective. My theorycraft work and guides have a clear perspective: be as effective as possible in a raid environment (which is why they’re different from yours). “Playstyle” is only a distraction from this. People are always free to pick different talents based on their own idiosyncratic preferences. Again, that’s fine–it’s the case currently and it will always be the case in the future. People can do what they want, but guides and strategy discussion will let them know, as well as we can manage, what’s most effective.

    So this got a bit long, but question I’m trying to get at it: what precisely do you expect to change? There will always be an ideal spec for a given situation. We’ll always try to find it. People will always do things differently if they feel like it. People other than me will always be interested in doing alternative theorycraft work to accomodate the aforementioned idiosyncratic preferences that people have about their “playstyle.” Nothing Blizzard does will change this underlying pattern. It’s inherent in the ways that gamers behave.

  12. Lavata says:

    Hamlet you make the point of saying playstyle does not matter and is a distraction from the pure optimal choice. This assumes that everyone has equal skill at a class. So already the “perfect/optimal” model gets broken right there.

    Playstyle can be related to skill but not directly. There are drastically different play styles between PvE and PvP so already you have to split the optimal choices between 2 playstyles. I use to really enjoy playing fire on my mage. Then I got into arcane initially because it was somehow “proven” to be the better spec, this was during BC. But even then on the alter of the optimum spec raiders/players should be respecing between every fight if they want to be the best. What guides for the average player and not the min maxer from hell are about finding generalities that work for the most playstyles as possible.

    But really Fun > Performance unless you are doing hard modes. Because if you are not having fun playing the game, what the hell are you playing the game for.

  13. Siobhann says:

    This is the most self-centered blog post I have ever run across. It reads as if you think Blizzard should cater to GUIDE WRITERS and not the player base.

    Right now talent trees in WoW are far too restricted. The prevalence of cookie-cutter builds is one of the worst aspects of the game because if you choose fun, interesting talents it costs you DPS or mitigation and you’re branded a failure.

    There is a solo questing game, we keep doing 5-mans, and some players like to both raid and PvP. Why should you have to trade between all the fun solo/survivability talents to max DPS? Think outside the box, Lissana. The more points that can be freed up to take “fun” talents like imp. barkskin, owlkin frenzy, or typhoon, the better.

  14. Chipster says:

    Honestly, I think you are worrying far too much about what these hypothetical 5-10 talent points will be.

    If the goal of this entire exercise is to reduce bloat in the talent trees, wouldn’t having 20 useless talents to choose from after creating your “cookie-cutter” spec defeat that entire idea?

  15. Hamlet says:

    Ok, got a bit ranty before because class was ending and I was typing fast. To clean up a little:

    The OP is long, but I don’t get a coherent thesis from it about what exactly you expect is going to change. I don’t see at the moment what even could change. In the next expansion, Blizzard will put in a bunch of talents. Some people will eventually figure out what’s most effective for a given raid situation. Players will either
    –use that approach worked out by the community,
    –innovate on it and add to or modify the body of strategic knowledge for everyone, or
    –ignore it and do something else which they find more fun.
    My underlying point it, all of this is true now, and all of it will be true in Cataclysm. Anyone who thinks they have some idea for changing this pattern is making themselves out to be quite a wizard of game design.

    Now, I do think some things will change. Talent trees will hopefully be cleaner and better designed. But this post was about choices, and highlights two kinds of choices: talents choices based on situation, and talent choices based on playstyle.

    Talent choices based on situation we have now, and will continue to have. One avenue for improving the game would be to have more interesting talent choices based on different situations (but then you have to deal with the respec sytem, etc.–different topic). Talent choices based on playstyle, however, is simply a term for “doing what I find fun rather than what’s most effective.” Some people do that now, and some people will do that in Cataclysm. And I’m not trying to look down on it all–it’s just not me nor the people I play with.

    But as far as I know, guide writers are concerned about telling people how to play [i]well[/i]. Each player has to figure out on his own what approach to the thinks is fun. And when we’re trying to figure out how to play well, any choice in talent points will eventually look the same as what we have now–a variety of options out of which good players will figure out what’s most effective.

    I think I’ve run out of different ways to say it. Right now, you can use whatever set of talents you like. Some are better than others. They’ll give us new talents in Cataclysm and we’ll distill the good from the bad once again. The alternative playstyles of the present, once players have had enough time to explore them, will always become the superior and inferior playstyles of future.

  16. Lissanna says:

    Siobhan – The problem for talent trees like the druid balance tree is that there are a set of clear “good” talents and a set of clear “bad” talents (ie. Genesis is a horrible DPS talent for raiding moonkin, because it doesn’t give you a good return for your point investment). So, just having 5 extra talent points to throw into the existing balance tree (or trimming down some of the DPS increase talents) leaves us with basically throw-away points without much really to invest in. I would actually like to see more “fun” points in the balance tree that actually still help us fill our roles (ie. improved bandages could be a “fun” point, but it would be wasted). If you are running in a Damage Dealing raid role, there should be a clear pathway with talents that have an effect on your damage dealing ability, and then there could also be more pvp-oriented moonkin points that PvE’ers skip, but PvP’ers think are the greatest thing. I don’t even mind having 5-man utility talents available for people who run 5-mans. However, there would be (very clearly) pvp versus raid versus 5-man talent specs, and it would be pretty clear on how to draft specs for those different roles. They could even offer things like wrath-spam spec versus starfire-spam spec or something silly to have different ways of going through the talent trees to have about equal DPS. However, if the about equal DPS talents come from the fact that you have extra 10 points you spend on things that don’t actually impact how you play, that you just dump points into for your passive mastery bonuses, then you risk having talents actually not be that fun.

    For the druid’s restoration trees, there is more flexibility in terms of different healing roles that change the way that resto druids play (ie. how much you use nourish versus HOTs impacts what talents are better for you). It’s easier to make utility talents interesting for healing specs, but harder to make utility talents actually contribute to a failing DPS spec like balance druids.

    @Hamlet – I didn’t really have a point. I was just rambling in the first place, which I’m allowed to do from time to time. 😉

  17. Hamlet says:

    That makes me feel a little better since I was basically doing the same 🙂 . But it’s easy for me to get going on how, while “playstyles” have their place in letting people enjoy the game, the whole idea tends to be at odds with learning how to play better (which is what guides and commonly accepted talent builds are about).

    On that note, where should I comment on a few things in your guide that I think cross the line from “style” to “inaccurate”?

  18. Lissanna says:

    I also use the term “playstyles” different than the average noob. I use it to refer to the differences between PvP and PvE, or raiding vs 5-mans, and a much broader level than you are defining it. I also think that by having “flex-fun” points, Blizzard is trying to make multiple VIABLE playstyles into something that is meaningful (ie. replenishment WG/rejuv-rollers vs. nourish w/supporting HOTs healing; OR CF vs. non-CF builds).

    At that level, playstyles can be something that varies between people without it having a negative connotation. I’m not sure what term you would use instead of “playstyles” in that case.

  19. Lissanna says:

    If you have things that you think I should update in my guide, you can forward suggestions to my blog’s e-mail: [email protected]

    I had a hard time writing this version because my guide is used by such a wide variety of new druid players that it’s hard to come up with something that works for all the different possible roles a resto druid can fill in the content.

  20. Hamlet says:

    Mmm, that helps a little, but then all you’re really describing is different setups for different situations.

    But this goes to what I’m saying. There are already different approaches and different associated builds (Nourish, HoT’s, etc.). And some people use each of them. But if you’re trying to deal with heavy raid damage, one is best (heavy WG/Rejuv spam). We’ve just internalized that fact so well that it’s easy to forget that Glyphed HT was once merely a “different playstyle,” but it’s there in the game for anyone who wants it.

  21. Rhabella says:

    I think you are very eloquently missing the point and defending the idea GC is trying to convey here. When he says, at least IMHO, “spend the last 5-10 points where you want” he and the other developers are encouraging debate and discussion. They are promoting discussion where instead of the builds being EJ cookie cutters, you will find very talented and analytical writers like you who will be able to justify talent W because of situation X but others will say “no, you have to take talent Y because it was a life saver in situations similar to Z.”

    The utility talents will change the game for the better, and though your guides will still easily be filled in because of your opinion and dedication to the art, your recommendations will then be more about playstyle instead of raw output, and five years into WoW and the constant debates about homogenization of the classes, isn’t some variety in 5 resto druids what we really want to extend the life of WoW?

  22. Lissanna says:

    Yes, but some playstyles will always be better than others. 😉

    I think what GC really wants is multiple equivalent talent specs that don’t really outshine eachother. I’m not sure how realistic that ends up being.

  23. scaresome says:

    I just read the Ghostcrawler comments.
    He said that some of the talent points that give a passive bonus will be gone; imagine if you didn’t have to spend five points in Gift of Nature. That’d be five more points to play with.
    Plus, he confirms we’ll give five new points for leveling up to 85.
    Finally, he says there will be new talents yet the talent trees won’t go deeper.

    I think a true Restokin build be possible and common.

  24. Dal says:

    To be honest, I hope GC succeeds.

    I’d love to see ideas like scaresome’s implemented.

    Have talents that can’t just be dismissed because they don’t add up to the right numbers in a spreadsheet.
    Talents like the priest “Body and Soul” are another example. It doesn’t add to HPS/HPM, and was dismissed early on in EJ, for that very reason. Now it’s considered a valid alternative option. Put 10 talents of that nature in a tree, and make me pick the 5 I like.

    Make “playstyle” mean something, other being an object of the type of derision demonstrated by Hamlet’s post.

    I don’t want to be in a game world where every Resto druid is one of 1-3 builds, all wearing the same equipment at a given progression point.
    It’d be nice to be in a world where a second or third person of a class brings something useful and different to a raid.

    Give me meaningful / interesting options for talents. Give me several sets of gear at each tier level that interact with those talents.

    I know this isn’t an easy goal for Bliz’s designers to reach, and I know they won’t necessarily succeed in keeping a balance over time. Patching will happen. Classes will have times being OP or weak. The wheel turns. I’ll put up with that.

    But please, please take me away from an Azeroth where one can say “my calculator proves that this is the right and only way to design and build your toon for raids or you fail, noob.”


Featured Blogs