Article Header Image
A Matter of Priorities
Mike Mearls

A s we move forward with the design of D&D Next, it’s worth taking a step back to look at some of the general principles of RPG design that the team has employed over the past year.

Of course, we’re busy right now finalizing design and pulling things together. The results of the last round of playtesting were overwhelmingly positive. Although we still have things to polish up, we hit our highest levels of approval. Believe me, you’ve been tough critics in these surveys. It feels good to look back at where we started with the very first round of material and see our progression from there. We’ve made huge strides forward in keeping our playtesters happy while maintaining a consistently high level of participation.

So, without much new design to show you, it would be useful to address some of the techniques and approaches we used in building D&D Next.

Priority Ranking and Balance

Here’s some insight into our design process, specifically regarding how we worked to balance spellcasters against their nonspellcasting allies. Our weapon of choice was a simple priority ranking of options.

A priority ranking for D&D is a list of character option categories, such as classes and proficiencies. The priority ranking places the most important elements at the top of the list and the least important ones at the bottom. In terms of D&D’s content, classes sit at the top of the list, followed by races, spells, backgrounds, and feats.

The ranking answers the following question: What wins in a contest between option A and option B?

If my class gives me the ability to become an awesome archer, I should be a better archer than other characters whose classes give them no archery ability, but who took feats or can cast spells on themselves to gain that ability. Class is ranked higher on the list than spells or feats. Therefore, class wins.

This kind of ranking helps make sense of the interactions between options as a game is designed. It’s a tool to help answer questions, resolve conflicts, and guide design decisions. It helps shape all the elements in the game by clarifying their relationships.

More importantly, using a priority ranking meant that we could sort out options and abilities at the class level, knowing that assigning a feature to a class made that class the best at that particular feature (or tied it with another class that needed that option, too).

When two types of character options start to crowd each other, we know which one is more important. In general, a more important piece of the game should trump a less important one. There are exceptions, but as a general guideline, this rule gives you a place to start in design.

Take spellcasting as an example. If you want to become a powerful caster, you need to take levels in a class such as wizard or cleric. You can take feats to augment your casting ability, or you can pick a race that offers some minor innate magical ability. However, climbing to the top of the spellcasting heap requires you to invest in a character class.

That example might sound obvious or intuitive, but it becomes much fuzzier when you delve into more nuanced class features. If rangers can track but all characters can take a proficiency and feats that allow them to track and improve their ability, who should be better at it? Is the ranger equal in tracking to the character with the proficiency? What about the character with proficiencies and some sort of tracking-related feat?

The line is rarely obvious and easy to spot, but the ranking of options helps us understand how we should weight things. In this case, we’d expect that a ranger could be overshadowed at tracking only by a character investing several choices in background options and feats. A nonranger needs to make a real commitment to stepping into the ranger’s niche.

On the other hand, something like stealth is less clear-cut. In prior editions, a wizard might use invisibility while a rogue makes a Stealth check to hide. In D&D Next, we decided that stealth and other checks are of utmost importance to the rogue—the elements that help define it as a character class. Thus, a rogue who takes stealth options within the class shouldn’t be overshadowed by the invisibility spell. A class trumps a spell.

As with everything relating to game design, much of the process is more of an art than a science. The ranking of options is a useful tool, but not a straitjacket or a checklist. As a team, our understanding of the ranking was fairly explicit about a year ago, but became more intuitive and obvious as our work progressed.

The biggest benefit of this approach was that it made it clear when a spell was threatening to overshadow the abilities of a nonspellcasting character. In almost every instance, we opted to improve the abilities of a class such as the rogue or the fighter to exceed the temporary boost afforded by a spell.

The other big benefit was that it allowed us to create a flexible system for customization. We knew that we could give options for stealth, casting, and so on in feats, backgrounds, and races, as long as the options available at the class level served as an upper limit that those lower-ranked options couldn’t overshadow.

Mike Mearls
Mike Mearls is the senior manager for the D&D research and design team. He led the design for 5th Edition D&D. His other credits include the Castle Ravenloft board game, Monster Manual 3 for 4th Edition, and Player’s Handbook 2 for 3rd Edition.
Sort Items By: Newest First Oldest First Top Rated
I like it!
Posted By: Pyrate_Jib (11/21/2013 12:19:40 PM)


I don't have a problem with any of this. But I'll still have to see it in action first.
Posted By: Prom (11/20/2013 2:35:29 AM)


I like this framework as it provides a consistent rubric for measuring balance, which is not easy to define and varies from player to player by taste.

However, if class is going to reign supreme then it REALLY makes it critical that classes not step on one another. Early on, there was talk of having a Basic, Standard, and Expanded version of 5E. I would really hope to see:

Basic - Core Four (Ftr, Clr, Rog, Wiz) only

Standard - Core Four plus a few optional classes that definitely stand on their own.

Expanded - All the weird stuff, whether it can stand on its own as a class (ex: Monk) or cannot but some people want it anyway (ex: Warlock).
Posted By: TheWilds (11/19/2013 10:10:30 PM)


In principle this sounds great. But the real test comes when the rules get endlessly modified as expansions and Player's Handbooks 2, 3, 4... come out and add feats, backgrounds, and other options. That is certainly one of the fundamental problems that occurred with 4E: WotC put together a great and consistent system, and then they proceeded to crush it under the weight of seemingly endless rules modifications (feats and other options that unbalanced the game). We'll have to wait and see if they do it again with 5E.
Posted By: recktenbenwald (11/19/2013 12:46:30 PM)


One of the things that make character interesting and different is the options in feats. Feats are where the fun is and I think they have been marginalized too much. Indeed, they are OPTIONAL! I think this is designing for the 20% that don't want to mess with them instead of the 80% that love them. Create a couple of standard and simple feats that are enabled if players don't want to mess with them, but let the majority of us that love to tweak and play, have the ability to have fun with all of the feat combinations.
Posted By: Spykes (11/19/2013 11:26:26 AM)


Feats are really fun for making your ranger different from other rangers, but I would have to agree with the article over your opinion. I'd rather a Ranger be definitively different from a fighter with a few bow and spell casting feats.

Feats shouldn't be enough to 'remake' a class, in other words.
Posted By: korndawg64 (11/20/2013 12:16:41 PM)


Sounds like a solid frame work!
Posted By: moes1980 (11/18/2013 10:22:31 PM)


If the problem is the power of a spell, RogerWilco has already answered: give the counter-power at the same level.
The same for magic item.
Posted By: Eilistraecomeback (11/18/2013 7:42:22 PM)


My 4e epic rogue gets neutered because blindsight, tremor sense, truesight, etc were too easy to come by. Handing out too many counter powers can weaken class choices as well as spells. Very tricky balance.
Posted By: TurkeyBreh (11/19/2013 12:19:46 AM)


Also is sad, at very high levels, when a powerfull enemy go down only by a rogue with improved invisibility.
But most of his minions and normal people don't live with truesight permanently on them.
Is not an easy task DMing at very high levels, when PC's have a lot of resources and so their enemies.
The rogue of my group, even when detected use deceive, feint and flanking to hit with sneak attack.
Posted By: Eilistraecomeback (11/19/2013 6:14:22 AM)


If the feats are set at the bottom of the priority rankings, why yours made most of them so powerfull?
If a spell can't overcome the class speciality, even for few rounds, what is the sense to cast high level spells?
If yours want a class to be better at certain task, why give to all class the same precision to hit?
If the race is so important, why this have a lesser impact than background or feats in character creation?

Mr. Mearls, I understand the need to have priority, placing class and race at the top, but don't categorize so rigidly the others elements.
Is simply obviously that the feats need to be a real option instead of an ability improvement (with lesser slot than 3rd edition to obtain) they need to be more strong.
The most important thing to balancing capacity, is if a character w... (see all)
Posted By: Eilistraecomeback (11/18/2013 6:50:34 PM)


It takes too freaking long to roll up a character already, don't encourage them to differentiate any more stuff. You don't think feats are strong enough? LOL and I don't think the low to-hits are here to stay any more than the spellcasting bonus is gone for good, that was a rush job. At least it means the new game will be out sooner! It wasn't necessary to paraphrase what Mearls already said about competing with a Ranger as to tracking, but that's a very good point about high-level spells. I'd like to see Tenser's Transformation not be nerfed for once.
Posted By: RadperT (11/19/2013 11:12:04 AM)


Sounds like a solid design framework to start from and I hope you enjoy the artistic aspects of what your doing as the design process unfolds.

Had a great session with my group using the final playtest materials. I'd love to see minor updates to the playtest to preview some of minor balancing changes if possible. Might be useful to have a lot of people testing out your numbers, I dunno.
Posted By: JDizzleton (11/18/2013 3:25:32 PM)


Very cool to see the inter workings of design, and I really like the system of priority, as class should trump others options. Well thought out, two thumbs up!
Posted By: tirwin (11/18/2013 1:55:36 PM)


I really think feats should trump both spells and race. Spells are ubiquitous, feats are rare...ergo, they should have more impact than a spell the wizard can switch out tomorrow.

And race should largely be thematic and cosmatic. No one should ever feel like they need to be a certain race to play a certain class.

That second point is a book throwing problem. If it feels like my character must be a dwarf to be the best fighter, or a tiefling to be the best warlock, I will (metaphorically speaking) throw the books across the room.
Posted By: Salamandyr2000 (11/18/2013 1:21:03 PM)


Mike Mearls doesn't suggest that you must be a certain race to play a class, just that some races are better at some things. Thus, a halfling makes a better rogue, dwarf as a fighter/cleric, orc as a barbarian, elf as a mage and humans as whatever their heart desires.
Posted By: Darkwon (11/18/2013 1:30:38 PM)


I don't think he's even saying that certain races will be the best at certain classes. He's saying that no spell or feat should let a human be a better elf than an elf. If you're going to have dwarves be defined by their ability to stand in one place and be immovable by enemy pushes etc., then don't give the halfling a feat that lets him be just as immovable. If you do, the dwarf becomes less special.
Posted By: dreamwolf (11/19/2013 2:17:12 AM)


Do not forget about "items" in the priority list.
Posted By: WahSword (11/18/2013 1:19:22 PM)


I like this design process, but I think the stealth vs invisibility spell is a poor example. Stealth applies to two sensing, hearing and seeing which can both be overcome by a PC or NPC with training in perception. However, invisibility applies to one sense, seeing. To be invisible is "not visible; not perceptible by the eye" as the first reference from Thus, invisibility can not be overcome by the a perception check, only by guess work from hearing presuming the caster is making a noise, such as walking or casting a spell with a verbal component.
Posted By: Darkwon (11/18/2013 1:11:43 PM)


Why does it matter that Invisibility is only tied to vision and Stealth is applied to Sight and Sound. The net effect is mostly equivalent.You can use a Perception check to detect both of them.
Posted By: Ramzour (11/18/2013 2:49:40 PM)


At first, it seemed to me that feats being at the bottom did not make sense, especially with the new bigger feats. But after more I thought, I realized that feats are equivalent to stat boosts, and ability scores are the basic resolution mechanic of the game. That means ability scores/modifiers (and the linked feats) are base line and everything else surpasses or trumps that.

I assume that backgrounds encompasses skills/proficiencies, which again makes sense because this is the next step up from ability scores, as skills/proficiencies are just a bonus to an ability check.

The only question I have at the moment is where do at-will spells fit into this?
Posted By: GilbertMDH (11/18/2013 12:51:40 PM)


I like this insight into the design process, it's starting to make me understand why Next is as solid a game as it is, already.

One question I'd pose, though - how do you decide how to differentiate classes that occupy the exact same niche, such as an Archer Fighter and an Archer Ranger? How do you ensure that neither one is "better" at the thing they're the best at, and how do you ensure that they do it differently?
Posted By: wetsail (11/18/2013 11:31:41 AM)


As an AD&D guy, I'm like, they both pull back a bow and hit the same way, what's the problem? Having played within the niche protection paradigm since Drizz't days, I understand your concern. Rangers have gone from being those who aren't, to the crack shots, to the super-fast shooters who do more damage while fighter archers do more damage, and effects, and more damage. Dialing it back a little, Fifth Edition seems to give the fighter more rapid advancement in shots per round, while rangers have greater potential to damage at a distance, particularly when you take their spellcasting capability into account. Inasmuch as I am not a game designer, I promise you will be able to take a feat to make your archer a crack shot. Insofar as either class can take the Archer specialty, I admit that there is plenty of overlap, but myself am not concerned, as along with that diversification comes opportunity for customization.
Posted By: RadperT (11/19/2013 10:53:32 AM)


Class being top priority presumably means that single class characters should be better than multiclass characters about specialties. However it would seem a multiclass Rogue/Mage might be better than a rogue in some cases.
Posted By: KoboldAvenger (11/18/2013 10:26:43 AM)


I think it's important to not just look at the raw numbers, but also at where the choices are made. Even if a class ability is more powerful than a spell, the spell can win out, as the class ability can only be chosen when the character is created or the level is gained, while a spell offers a choice during play.

A second point would be to look at abilities that counter other abilities. The power of Invisibility depends on how hard it is to get See Invisibility.

The third point is related to the first two: How are you planning to keep the balance you set out with after releasing more books? In my experience the relative power of an ability can change a lot over the life of an edition.
Posted By: RogerWilco (11/18/2013 8:02:47 AM)


I kinda want to comment, but this reply system is so atrocious this is all I could manage.
Posted By: Ashtoret (11/18/2013 6:35:50 AM)


It still seems to be escaping some people that if you want to make an ampersand, all you have to type is & (Period.)
Posted By: RadperT (11/19/2013 9:53:38 AM)


I like the Priority Ranking concept! Just goes to show that they DO think very critically about different aspects of the game and how they interact with each other, contrary to what many forum-dwellers always complain about.

I'm curious to know how similar effects are compared. Take the example of Rogue Stealth and the Invisibility spell. Is it the duration? the frequency it can be used? Does the Hidden condition trump the Invisible condition somehow. For instance, Hidden probably refers to Sight AND Sound, whereas Invisibility only covers Sight. What do they value effect that lasts a long time or an effect that can be used with greater frequency?

I doubt we'll be privileged to the ranking criteria, but it would be see nonetheless. It would make homebrew classes, feats, and spell a LOT easier to design and balance! I hope we at least see a rough version of this printed in the DMG.
Posted By: Ramzour (11/18/2013 4:58:39 AM)


"Thus, a rogue who takes stealth options within the class shouldn’t be overshadowed by the invisibility spell. A class trumps a spell."

Good to hear. One of my fears of 5e DnD is the drop back into castor superiority.

Now, if you can just write in adding damage dice to martial characters; like how spells get extra damage dice as casters level up, I will be somewhat pleased.

Oh, and game design is 'art' and 'science'. Your math has to work. Period. The language used and reasons given do describe that math have to be plastic, yet comprehensible to even the most ardent rules lawyer/edition warrior. So they all shut up and play. :) Make the game math clear, not hidden. Straight up tell us all the game math. i.e.: don't publish any chart or whatever with out breaking down and describing the math in the DMG.
Posted By: seti (11/18/2013 3:18:47 AM)


If your math is 'proprietary', well then your system cannot be effectively DMed. Share it. Period. All the old editions suffered from that. 4e suffered the least. 4e angered people with its errata. But, I think a lot of those people mad about errata didn't realize that the internet allowed errata to be released, for free, to anyone who cared.
Posted By: seti (11/18/2013 3:25:09 AM)


MMearls was talking about 'art' vs 'science' in terms of categorizing abilities by priority, which was the point of the whole article. He wasn't talking about the game math.

Also, I don't think every chart needs a mathematical formula attached. Mostly because there's not always a simple function to govern a particular set of numbers. A chart and/or progression by level is more than enough to understand the under-the-hood mechanics.
Posted By: Ramzour (11/18/2013 4:51:12 AM)


Now, if they can deliver on the promise that class will be more important than spell.... I'm not holding my breath.

Of course if they manage it that's going to infuriate the people who are convinced magic must be better than mundane abilities or it won't feel magical.
Posted By: Bluenose (11/18/2013 5:15:24 AM)


"Good to hear. One of my fears of 5e DnD is the drop back into castor superiority. "

Casters should always have a niche over other classes because their spells can manipulate effects from other class at the cost of low hit points and lack of armor. This is what makes the Wizard/Mage an advanced class playable by advanced players. Spells are the wizards livelyhood just like Martial Path Benefits are to fighters, so their spells should be equivalent.

My question is, why must a feature of the rogue trump a feature of the wizard ? Spells are a feature of the wizard as they are to the cleric. As per the stealth vs invisibility spell example, stealth can be used any number of times per day while the wizard can only cast it a number of times which then last for a short duration.

This same train of thought would suggest that the wizards primary cantrip attack of Burning hands or Chill Touch should be less powerful then a weapon attack, because it is... (see all)
Posted By: Darkwon (11/18/2013 1:26:43 PM)


Casters will be very limited in 5e, except on a 5MWD. It will be mostly spamming cantrips otherwise. Overreaction the the LFQW meme is baked in the cake since Monte left.

Mr. Mearls believes your 1 spell per day that lasts a minute (if saves, dispel, disruption, SR, ER, etc. don't make it fizzle outright) should be overshadowed by the Ftr/Rog who can do it all day (plus don't need to cower on the back line). Oh boy.
Posted By: mcintma (11/18/2013 4:37:30 PM)


Earlier packets were so much fun for exactly the opposite reason. You could create a character that was an Archer, and a Cleric who had started out as a Soldier before being called by their deity. That character could hold his own as an Archer, alongside a Ranger or Rogue. That was an excellent feature, because in doing so the character sacrificed other options, but was still relevant.

What you've got now is a the exact opposite of customization in the character creation process. What you end up with is boring cookie cutter classes who's only real difference is the Background Story (not to be confused with 5Es background choice) a player may, or may not, put any time or effort into creating.

Let me just say, that I truly and absolutely do not believe you concerning the Playtest Survey results. I've never given such a negative review across the board as I did on the last survey. Given the high levels of contention on WotC's forums themselves as well as many o... (see all)
Posted By: LupusRegalis (11/18/2013 2:30:27 AM)


Oh no! Your opinion isn't in the majority so the only explanation is that Wizards is lying to everyone! Conspiracy!
Posted By: Ramzour (11/18/2013 4:46:28 AM)


An excellent argument for making Backgrounds more like Specialties, with the potential for a benefit like Fighting Style, and advancement at higher levels. I'm not a super-imaginative guy, so when I'm playing I appreciate something like a prefabricated background (simple to start, please, e.g. specific skill sets) to jumpstart my image of a character.

I noticed, reading this article the next day, that no one had rated any of the comments above (yes, those are all my opinions; one person apparently rated the 3 below early on the day of publication). The fact is, we 12 are trolls, and the vast majority of the playtesters ran the game (my experience leaves me certain that most players don't bother to fill out the surveys) and dutifully gave their feedback without coming here to virtually jump up and down. It is safe to assume that Mike Mearls is telling the truth; on the other hand, I have spoken to plenty of people who have read the playtest materials and didn't participate... (see all)
Posted By: RadperT (11/19/2013 1:39:38 PM)


I like it.
Posted By: Azzy1974 (11/18/2013 2:21:26 AM)


This makes a lot of sense.
Posted By: Germytech (11/18/2013 1:52:00 AM)


I like what I hear. The hierarchy and how it is applied does solve a lot of problems for class and feat balance.
Posted By: Alter_Boy (11/18/2013 12:30:57 AM)



Create Comment
Follow Us
Find a place to get together with friends or gear up for adventure at a store near you
Please enter a city or zip code