Formalizing the illusion of choice

100 Tauren Paladin
17020
Some are just not equal however. Silencing shot vs Wyrven sting really that is not a choice. A choice is Pepsi or Coke both are cola.
Reply Quote
90 Worgen Warlock
4980
Anyone else miss the old talent trees because I really do.
Reply Quote
90 Pandaren Shaman
10620
Tell me, if you don't want talents that are better in one situation, but not another, what exactly do you want?

Seriously, give an example of 3 balanced talents that are equally useful in all fights but meaningfully different to provide a playstyle difference.

You might be able to get that with damage dealing abilities (ie have a short cooldown low damage ability vs long cooldown higher damage ability), but even there you typically run into issues where one will be better in PVE, the other will be better in PVP; or one fight might have high uptime and steady damage, while another boss favors quick burst damage less often, so you still want to respec per fight.

Honestly, what you want is impossible to deliver. Providing talents that are more or less beneficial depending on the fight is about the best balance you could ask for. If the problem with this is respeccing all talents for every fight, then maybe they should make talent swapping something you can only do once per day or once a week, rather than something that can be done at will.


The level 15 Shaman talents are a perfect example of Blizzard's talent philosophy. Natures Guardian is a passive damage reduction that procs when you go below 35% health. Stone Bulwark Totem is a shield that absorbs damage. Astral Shift is a straight damage reduction button. You can take whatever you want for whatever you need. AS is better for taking a straight up nuke, but that's about it. You pick something passive, something with a short duration short CD, or a big DR low CD. The only thing that will really dictate which you pick is fight length. You might have a boss that does something every minute, so SBT will be better than AS because you get more uses.
Reply Quote
1 Human Mage
0
06/20/2013 05:14 PMPosted by Hiroran
I am not required to do anything in regards to the talents i can pick whatever I choose to. I choose to pick whats optimal for me, but I am in no way required to optimize it.

(n.b. we will assume a 'rational' player for the purpose of this thread. i.e. one whose ideal is to, in fact, play optimally. Hence the, "I do not care about optimality and so will pick no spec and tunnel FFBs into the boss" is not a valid argument - for the sake of this thread).

Indeed, there is no slight on your person, or anything 'personal' of that sort within this thread.

For context's sake, however, we are assuming we all are 'those types of players who have chosen to play at that level'. If nothing else, but for arguments sake.

Do you think that the Priest Level 90 talents are an example of a talent tree that, while it has its optimizations, can be chosen according to personal preference without a significant loss of throughput?

On the aforementioned predefined spectrum? Then yes.
Or at least, the priest 90s are much further off from the point that the mage bombs are so securely occupying right now and with the PTR change.

I actually quite like the priest tier 90 talents, for both my shadow and my healers. I think divine star is the most 'interesting' one of the lot, since it adds a cool dimension to my disc healer, that being, positioning. I play the more 'mobile' disc healer, whereas my holy is somewhat more static, with cascade/halo fitting better in her playstyle.

That being said, I will admit the choices are far more interesting from a healing perspective. They could be more interesting for spriests, but that, imho, is a separate story.

06/20/2013 06:48 PMPosted by Cuddlebolt
Natures Guardian is a passive damage reduction that procs when you go below 35% health. Stone Bulwark Totem is a shield that absorbs damage. Astral Shift is a straight damage reduction button. You can take whatever you want for whatever you need.

I also like the hunter 15s. Like a few other 'good' matchups, they are also difficult to directly quantify.

I feel that is an important trait for these talents to have, since it forces each individual player to come up with their own metric by which to assess the talents, which in turn, promotes diversity.

Anyone else miss the old talent trees because I really do.

Aye, I do miss 'true' hybrid builds. But nostalgia is something that must be kept in check during such analysis.

That being said, I think I lost more friends to Rift these past few weeks then I can remember (it just went full free to play).
For most of them, their number one reason was, in fact, the talent trees.

Rift's talent systems are much more akin to original WoW then most other talent systems out there.

Directly quoting a really good rogue friend of mine who just left, he commented "In Rift I can be a rogue who is not like every other rogue out there. There is more diversity!", which was something he attributed to their talent system.

So maybe there is something to be said about the old tree design? Perhaps. It certainly felt more diverse back then *shrugs*
Reply Quote
100 Troll Mage
16155
Indeed, there is no slight on your person, or anything 'personal' of that sort within this thread.

For context's sake, however, we are assuming we all are 'those types of players who have chosen to play at that level'. If nothing else, but for arguments sake.


First off I dont really agree that that is a good assumption to be made.

and also direct you to part of my last comment

That being said sure its preferred if they are all equal and its more of a stylistic choice. But I see that as such an idealist look to the point I see it as pointless worrying about it.


Maybe this isn't a discussion i should join in on from my apathy idk. Just seems rather idealistic and pointless.
Reply Quote
1 Human Mage
0
Just seems rather idealistic and pointless.

On the contrary, the concept of the 'rational player' is actually very well understood and has shown to be of great help in game theoretic analysis (decision making), which is what talent selection essentially is.

So while it may seem idealistic, it is in fact a very helpful tool to use in this analysis.
Reply Quote
90 Blood Elf Mage
11510
I don't have a problem with changing talents on a per-fight basis. What I have a problem with are three virtually identical spells that are arbitrarily designated in the tooltip as being best for certain types of encounters. That's way dumber than, say, picking between Divine Star or Cascade as Disc.
Reply Quote
1 Human Mage
0
What I have a problem with are three virtually identical spells that are arbitrarily designated in the tooltip as being best for certain types of encounters. That's way dumber than, say, picking between Divine Star or Cascade as Disc.

Agreed.

The thread is indeed focused on the bombs specifically. The thrust being that if we, for the sake of structure, isolate all talent match-ups on a spectrum of "pure illusion of choice" all the way up to "real choice", then the bombs sit at a point of that scale that can be deemed "far too close to the negative limit".
This is why their choice matrix is looked upon as an "illusion".

As a secondary point, I am using the bombs to elucidate the greater issue of this illusion phenomena. Of whether the concept of being "forced to change talents per fight" is something we should snuff out of design, primarily since it is this concept itself is allowing a talent tier like the bombs to exist in the first place.
As an example, if the "respec per fight" issue was (let us pretend) completely abolished and not allowed to be, then, through extension, talent tiers such as the bomb wouldn't even be allowed to exist in the first place!
Hence, we can actually ensure that the issue that exists with the bombs today, can never happen again. And we can do this by removing the core of what is allowing them to exist right now, that being, that "respec per fight" is not only being allowed by design, but being encouraged as well, especially now, in the rather blatant manner that the new tooltips describe.

However, as stated before, I completely agree that the choice to change should always be there. But I affirm that it must be a choice, and not a need.
Reply Quote
90 Blood Elf Mage
0
Honestly, what you want is impossible to deliver. Providing talents that are more or less beneficial depending on the fight is about the best balance you could ask for. If the problem with this is respeccing all talents for every fight, then maybe they should make talent swapping something you can only do once per day or once a week, rather than something that can be done at will.


Then the entire new talent system is a failure.
Reply Quote
Then the entire new talent system is a failure.


Not entirely.

It has, to some extent, made the playing field more about singular ability choices versus a couple of actual choices buffered by a large amount of "Increases <insert random stat or damage type> by X%" talents. It has also succeeded in a minority of instances such as the Hunter Level 15 and Priest Level 90 talents given players a choice between talents that don't have a readily perceptible superior choice and are of relative parity in terms of throughput so as to enable the average player to choose whichever spell they feel most comfortable using, looks the coolest, etc. Both of these show that the capacity exists to make a relatively good talent system.

However, aside from the problems that ZomgDPS has illustrated, I would like to put forward another. The fact that Specialization and Talents are independent of each other is part of the problem. There are two examples of this I would like to draw attention to:

1) The Mage Level 90 Talents: The myriad of problems with these aside, the most basic is that they all share a common mechanic of being tied to mana regeneration, despite the fact that only one of the three specs care about mana. And even then, Arcane's mastery and play style favor Rune of Power for the steady, consistent mana regeneration hinging on being stationary. I believe that part of the reasons for the botching of this talent tier is because they had to create talents that seemed "useful" for all three specs.

2) The Hunter Talents: Why do I bring this up? Well, it best highlights the other problem with having three different specializations share a talent tree. The fact that while the talents have their own mechanical merits and drawbacks, they contribute to a general dilution of flavor and theme between the specs. Barrage, for instance, is thematically beautiful for the Marksmanship spec. Beast Mastery, or even Survival? Not so much.

While these are the two I bring up, they exist across the board, even in relatively good talent tiers. As ZomgDPS mentioned, while choosing between Cascade, Divine Star, and Halo yields a very interesting choice in playstyle and utility for the healing specs, for Shadow it's less compelling.

Blizzard's stated reasons for the shift to the common talent tree were trying to distribute utility and survivability more evenly across the specs. In a lot of cases this succeeded to an extent, such as Ice Barrier being available to all Mage specs, Repentance being available to all Paladins, etc. But at this point I feel that this could have been best accomplished by simply identifying what the truly critical or advantageous survivability tools were and making them baseline spells.

My suggestion for Blizzard for the next expansion is; continue the layout of the talents, but triple their number and make them on a per-spec basis; this helps to create visual and thematic distinction between specializations and more importantly, means the talents can be adjusted on a per-spec basis. This avoids the problem of having a talent having overpowered synergy with a particular specialization yet being difficult to adjust due to the impact on the other two specs.
Reply Quote
100 Goblin Mage
16535
bump for the post. I agree.

Slightly off topic:

On a side note I still believe our level 90 talents are in the exact same situation. each of them in the end are truly bound by what spec you are playing. RoP for Arcane, Invocation for PvE Frost/Fire and Incanter's Absorption for any fight that has a constant pulse of AOE for PvE. IA is the UNANIMOUS talent choice for PvP for all specs.
Reply Quote
90 Human Death Knight
12620
If there are 5 or more targets, you should take Frost Bomb, or you are playing sub-optimally.
If there is just 1 or 2 targets, you must take Living Bomb, or you are playing sub-optimally.
If there is 3 or 4 targets, you have to take Nether Tempest, or you are playing sub-optimally.


Let's say you're fighting heroic horridon. Do you want the 5 target orb so you can hopefully hit as many at once as possible during doors 1/2/4? Do you want the 3-4 target one so you can get the most damage on the primary adds while still doing some to the smaller ones? Or do you want the single target one so you can give up some dps early for a faster rage phase?

Same with council. Do you want to be spreading as much damage equally as you can? Do you need better single target for pushing phases/adds?

Again with tortos, you'll want different depending on if you're doing bats or turtles.

I'm sure I could come up with more examples, that's just off the top of my head.

Sure you'll want the same one for all the single target fights this way, but your options on aoe fights are actually pretty flexible.
Edited by Lailala on 6/23/2013 8:17 AM PDT
Reply Quote
1 Human Mage
0
Sure you'll want the same one for all the single target fights this way, but your options on aoe fights are actually pretty flexible.

Actually, this is not accurate.

In pretty much every situation you describe, there exists a highest damage done option and a rational mage will pick that option. While you can say "I can pick the single target bomb for AoE fights and get away with it", that will actually not be the optimal choice.

Again, we are assuming rational players, i.e. those who will pick the optimal damage choice per fight. I am certain you can pick a non-optimal choice and still muddle through, but then you aren't a 'rational' player in the strict definition of the word.

Also, please understand, I am not saying that the non-optimal bombs somehow cause the mage's DPS to 'break' or cross some threshold of 'non-viability' or cause mage DPS to be unacceptable.
I am just saying that when a clear choice of 'highest DPS spell' exists (i.e. a dominant strategy), the 'rational' player will pick it without thought, hence making choice an illusion.

With that said, I feel you are missing an important point in all this, that being, player preference.

The idea being, that my preference for a talent should take precedence over the relative power of that talent over the other talents in that tier. That is the core point of even having a talent system in the first place, to allow for player diversity through preference.

In some ways, the situations you describe are actually what is breaking the system itself, and propagating the "respec per fight" issue. Now, if you are 'ok' with repseccing per fight, that is a different issue.

My assertion is that 'respec per fight' is not a valid design philosophy, since if we officially condone that behavior, then what is the point of player preference in the first place?

Talents are not meant to separate noobs from good players. This was a major driving factor of redoing the old talent system in the first place, since there existed many "traps" that noob players would fall into when selecting talents.
Why redo the system to get rid of that issue only to officially condone it in the new one? Therein lies the issue.
Reply Quote
90 Human Paladin
8390
Most 5.4 changes are bandaids for/till 6.0. Riposte for Blood and the bomb tier are obvious examples of this.
Reply Quote
90 Tauren Priest
0
That being said, I will admit the choices are far more interesting from a healing perspective. They could be more interesting for spriests, but that, imho, is a separate story.


What choice is there though? You take Divine Star if your group is stacked for most of the fight, you take Halo if your group is spread. Cascade isn't even a choice 90% of the time at least in 10m content. There is no choice when it comes to a raiding Priest as to what spell they will use in that tier, regardless of what they may prefer. The throughput lost by making the wrong choice is just as drastic as the damage a Mage would lose by choosing the "wrong" bomb.
Reply Quote
90 Draenei Shaman
16690
In pretty much every situation you describe, there exists a highest damage done option and a rational mage will pick that option. While you can say "I can pick the single target bomb for AoE fights and get away with it", that will actually not be the optimal choice.


The optimal choice for a rational player will be what they are best at using rather than what is actually optimal depending on the math. If a player is mechanically better at noticing and refreshing multiple dots than he is at paying attention to an extra low-cooldown ability than at 5 targets Nether Tempest may remain superior to Frost Bomb for that player, and on another side of that if the player is better at managing that cooldown than he is at multiple dots Frost Bomb may become superior to Nether Tempest solely due to the skill of the player using it.

The rational player will choose what he is best with, because that will in most situations give better dps than what is mathematically optimal. The perfect player is equally good with all 3, and as such will pick what is mathematically optimal.

Would also like clarification on your promoting the idea of having different talents be strong for different types of situations (respeccing by fight) and whether or not you think it's a bad idea because you start off stating it's a bad thing and then hold up talent tiers that are the best at pulling off that concept (except perhaps HoT v Shadowfury) as ideal.

I like and agree with most of your points, it's just the concept of the "rational" player that bugs me.
Reply Quote
90 Human Warrior
10675
I'll keep all this in mind when deciding on whether to use tclap or whirlwind. Welcome to about 6 years ago.
Reply Quote
1 Human Mage
0
I like and agree with most of your points, it's just the concept of the "rational" player that bugs me.

I am actually using the standard definition of the 'rational' player. Standard from a game theoretic and decision theory standpoint.

I am trying to find some easy to read resources that go into this in depth, though for now, you can probably just hit up wikipedia for some reading on the subject if you are interested and/or curious:

http://en.wikipedia.org/wiki/Decision_theory
http://en.wikipedia.org/wiki/Rational_choice

Needless to say, the definition of the 'rational' player in these contexts is actually quite standard. It is that definition that I am using in the context of this discussion.

06/23/2013 10:25 AMPosted by Gistwiki
Would also like clarification on your promoting the idea of having different talents be strong for different types of situations (respeccing by fight) and whether or not you think it's a bad idea because you start off stating it's a bad thing

My issue with 'respec per fight' is two-fold, with one issue being one that is independent of inference, and the second being based more in deduction.

The first is that the forced 'respec per fight' idea goes contrary to a core design goal of the new talent system. That goal being, that we should pick talents based on personal preference. If you are being forced to change your talents in any given situation, then that act is in direct contradiction of the stated design goal. Now, you can say that this design goal for the new talent system has changed, but that is another topic.
For this topic, we are working under the assumption that what Blizz stated as their design goals for the new talent system are indeed still in effect. If that is the case, then situations like the mage bombs and a few others, are in direct opposition to that goal.

And secondly, my more inferred issue with the 'respec per fight' is that if such a behavior is intended 'by design', then what is preventing it from propagating across the entire system? As we can see from the case of the mage bombs, if 'respec per fight' is accepted to be the norm of the talent system, then each talent matchup will just end up being an illusion of choice, since the factor that separates the talents in a particular matchup have little to do with player preference, but instead, have to do with which boss you are fighting, which, as we all can agree, is something completely independent of player choice.

A deeper issue forms when we extend ^ this idea.
Say now that we do live in a WoW where 'respec per fight' is considered 'perfectly fine'. If this is the case, then we can say that rational players will in fact respec per fight, but more importantly, we can say that rational players will all respec per fight to the very same optimal talent, hence, breaking the core principle that is justifying the existence of the talents in teh first place, i.e. if all players just end up respeccing to the same talent per fight, then the core reason talents even exist, that being to promote player diversity, is lost, since everyone will just respec to the same talent, making everyone the same.

It is due to these issues, some more basic and some more deduced, that leads me to conclude that on a core level, 'respeccing per fight' is just a bad concept to have in the system in the first place, since it will lead to homogeneity within talent choice selection, and hence, making the entire point of having talents in the first place, obsolete.
Reply Quote
90 Draenei Shaman
16690
am actually using the standard definition of the 'rational' player. Standard from a game theoretic and decision theory standpoint.

I am trying to find some easy to read resources that go into this in depth, though for now, you can probably just hit up wikipedia for some reading on the subject if you are interested and/or curious:

http://en.wikipedia.org/wiki/Decision_theory
http://en.wikipedia.org/wiki/Rational_choice

Needless to say, the definition of the 'rational' player in these contexts is actually quite standard. It is that definition that I am using in the context of this discussion.


I'm still trying to figure out how the rational player would come to the decision that he should use something that provides a slight increase in dps in theory over what he is better with, and as a result does more dps with, in practice.

The mathematical differences aren't so pronounced that they will overcome those differences in skill in most practical situations unless the player is approaching perfection.

I'll read the links and see if I can find an answer.

Edit: Let me see if I understand, the "rational" player will always choose based upon the theory regardless of how they'll perform with that choice compared to the others in practice correct?

It is due to these issues, some more basic and some more deduced, that leads me to conclude that on a core level, 'respeccing per fight' is just a bad concept to have in the system in the first place, since it will lead to homogeneity within talent choice selection, and hence, making the entire point of having talents in the first place, obsolete.


I feel like the only viable solution to this is to remove most, if not all, quantifiable value from talents. I'm not really sure how I feel about that. It seems like a model like that has the potential to be absolutely amazing, but can also fall flat on its face really quickly.

Thank you for clarifying though, I had misunderstood you initially.
Edited by Gistwiki on 6/23/2013 12:01 PM PDT
Reply Quote
1 Human Mage
0
06/23/2013 11:56 AMPosted by Gistwiki
I feel like the only viable solution to this is to remove most, if not all, quantifiable value from talents.

Perhaps. Though I do not think we need to get that extreme, not just yet at least ;)

As eluded to by me earlier in the thread, it is certainly the case that talent match-ups that 'work better' as far as "do they provide real choice?", are the talents that are much harder to quantify, e.g. movement talents. And in this way, perhaps making talents hard to quantify is something to aim for when designing talents.

However, there is another aspect that we deduce from the 'working' talents that I feel also bears worth mentioning, since there are 'mathy' talents out there which work (e.g. priest 90s), however, one stark difference between those mathy talents and say, the mage bombs, is that the talents aren't only separated by the math.

I.e. there is more than just numerical difference between the talents.

Say, for example, NT vs LB. They are both single target dots that are 10-15 seconds long, but apart from the numerical difference, there really isn't anything separating the talents, i.e. there is no other axis or dimension of comparison.

Now take, for example, Divine Star vs Halo. Both talents have 'mathy' parts, but they have this entirely 'other' element, e.g. positioning, which are much harder to quantify. Making the talents more interesting choices for a healer.

This simple difference is, imho, also another factor that can be kept in mind when looking at talents and seeing why some work and others don't.

Concisely put, I find talent match-ups work better when the talents involved are separated by more than one dimension along more than a singular axis, with these added axis of separation being things that are harder to quantify.

Let me see if I understand, the "rational" player will always choose based upon the theory regardless of how they'll perform with that choice compared to the others in practice correct?

To put it more simply, we do not need to worry about the rational player actually executing the action(s) itself.
We can safely assume the rational player has enough 'skill' needed to use all options optimally if he chooses to (i.e. he is just as good at using A vs B).

To some extent, this is actually true even in non theoretic situations. The better players of WoW can use almost all talents optimally where it pertains to actually executing on the use.
It is even simpler in terms of the mage bombs since 'using' them is just the expenditure of a global, and they are all used in an almost identical way (i.e. after 10-15 seconds, press button X).

In short, I do not think 'usability' is a real factor in the analysis.
Edited by Zomgdps on 6/23/2013 12:10 PM PDT
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)
Submit Cancel

Reported!

[Close]