Dragon Age: Leaving Out the Egg

Back in the day, Betty Crocker rolled onto the market with mixes for making cakes and such. More women were working and there was less time available. The idea was to make it easier to make real home baked food with less time and effort. It was a good idea, and Betty Crocker did a number of really clever things with chemistry – all you needed to do was combine the mix and water then bake.

It failed miserably.

So Betty Crocker sat down and did some serious market research, and they discovered something. Women weren’t using the mixes because it was too easy – it felt like cheating. So Betty Crocker went back to the lab and changed the formula to remove the egg component so the cook needed to add an egg of her own. That was enough to make it feel “home made” and it was a tremendous success.

I mention this because this speaks to a lesson that’s useful for a lot of products: if you “leave out the egg”, which is to say create an opportunity for the user to invest a little bit of effort to make a product their own, they’ll be more invested in it, and more enthusiastic.

In turn, I bring this up because it seems to me that one of the most contentious elements of the Dragon Age RPG is something of an egg left out.

The issue at hand is random character creation. The DARPG creates stats in a decidedly old-school fashion – you roll 3d6 for each of 8 stats, and the sole concession to customization is that you get to swap two stats. The immediate reaction to this is usually a pretty straightforward “What the hell? Is it 1985?” and that shouldn’t be too much of a surprise. Random stat generation is an idea that’s been pretty much set aside in favor of more player choice for a very long time.

The reasons for this are obvious – if stats are important and impactful, you can create a situation where a player with bad luck ends up with a character that’s not much fun to play when compared to his friend who rolled much better. AD&D was a really bad experience for a lot of us who got exposed to the difference between a fighter with a 12 strength and one with an 18/00, and it really soured people on the whole idea. After all, a lot of game design is fixing the problems you had with the last game you played.

There are some real problems with randomly generating stats or other character elements, but it has some real advantages that have been set aside along with the limitations. A random spread of stats has some of the advantages of an oracle – it can suggest ideas and patterns that would not otherwise be obvious. This idea of “what do you do with what you have?” is a tonal one in addition to a mechanical one – less badass but perhaps more heroic depending on perspective. That idea is a potent enough one that a lot of work has been put in over the years to try to capture this part of randomization without risking the flaws.

Random creation is also very quick – spending points requires a number of decisions that depend upon further knowledge of the system to do right. That can be something of a drag, and can end up putting the cart before the horse. It’s often the first decision of the game, so you don’t want to make it a painful one. The randomization also tends to produce more organic spreads – point buys tend to result in all-or-nothing spikes.

Now, this is not an assertion that randomization is the only way to go. There are a lot of other ways to approach it[1]. But I did want to lay out that it’s not as crazy an idea as it might first appear. Most specifically, these benefits sync up with the goals of a game for newbies, notably simplified choices and speed of play.

That’s all well and good, but here’s the thing that struck me during yesterday’s discussion. There are a lot of ways to address the issues of randomness – 4d6 and drop one, roll 12 and keep the best 8, roll then sort; the list is endless and has been kicked around for decades. It would take maybe a sentence or two to mention these options, so the choice not to do so is an interesting one.[2]

And this is where I come back to eggs. To leave out the egg from an RPG, it needs to be something that is obvious and trivial to address. Certainly, every RPG has a certain amount of egglessness – house rules are our bread and butter – but it is a little bit trickier to put in something that is (for lack of a better term) blatantly trivial. If you can do so, especially for someone with very little experience with games, then it can be a real win because it makes the first step much less scary, Once they’ve made the obvious house rule, they’ve crossed an invisible threshold into a sense of ownership of the game.[3]

The rules for generating stats feel like an egg left out. There are so many possible ways to address it if you feel it’s a problem that it seems like a gimmee. It’s easy to see and easy to make he change without disrupting the rest of the game in any way.

The thing I’m left wondering is whether or not it was intentional. If it was accidental, then it’s a lucky thing, but if it was intentional, then it’s freaking brilliant. And if it was intentional, then man, I am going to find a way to buy Chris Pramas a drink, because that is some badass ninja stuff.[4]

I am, by the way, entirely aware that I’m taking a very positive (and somewhat quirky) perspective on the Dragon Age RPG, and some of it absolutely hinges on a certain amount of hope regarding what’s still coming. My predictions and expectations could be totally wrong, and even if they’re right, the whole game could crash and burn for unrelated reasons. I’m pretty comfortable with the idea that others aren’t going to share that perspective, so objections and counterpoints are welcome, but I’m likely to stick with my optimism for the time being.

1 – One alternate example is equitable randomness, where the randomness determines which good thing you get, rather than whether or not you get a good thing. REIGN chargen is based entirely on this model, and the DARPG uses it for the bonuses you get from your backgrounds.

2 – Now, here I make a brief aside. This is an obvious omission, and it’s one of many obvious omissions in the game. You can tell they’re obvious omissions because the reader’s first instinct is to think “Why didn’t they include THIS?”. With that in mind, take a look at the credits page for the game – this is a pretty good list of folks with some serious stuff under their belt, and it’s safe to say that they thought of most of these things, but they made the conscious (and ballsy) decision not to do so. Paring things down to 64 pages required resisting the completist urge of game design, and that’s not an easy thing. It would have been easy to do this all in a standard 256 page full color hardcover, and that probably would have been a very good game with moderate commercial success, but it would have been just like any other game out there. The risks involved in the design are the risks necessary for this game to maybe make the jump to broader adoption.

3 – This flies in the face of the school of thought that says rules should be complete and that if they require house ruling, then they’re bad rules. That’s all well and good for pure design, but house ruling is engaging, and the power of that should not be underestimated.

4 – And, hey, on the off chance that I do get an answer from Chris, I have one more question: is it a real box? Please please please tell me it’s a real box.

EDIT – One last bit of credit where it’s due. The Betty Crocker story is from a fantastic book called “Something from the Oven: Reinventing Dinner in 1950’s America” by Laura Shapiro. It’s one of those books like Pollan’s Botany of Desire which is about one thing, but is really about a number of other very interesting things. Well worth a read.

21 thoughts on “Dragon Age: Leaving Out the Egg

  1. Daniel M. Perez (Slow Bike Miami)

    Chris has been pinged.

    The simplicity of DA is what made me buy it. I certainly don’t need another fantasy RPG, but the idea that it was made to be a brother to the red box pushes my buttons in a good way. Your two posts confirm precisely why I decided to plunk down my cash.

  2. Rob Donoghue


    And yeah, if it we just one more fantasy game, I doubt this would have moved my needle much, no matter how much I like the CRPG. I mean, I picked up SIFRPG, and I _like_ it, but I like a lot of things, so it sort of fades into the noise. This could just as easily have been one more of those.

    -Rob D.

  3. gamefiend

    I feel the need to say in spite of all my ramblings, I’m going to pick it up. I definitely want to see what’s going on with it, and I’m always up for more games.

  4. Mike Mearls

    Basic D&D still has the best approach to random character generation – drop your lower stats to increase your highest one. It compacted the choice down into one important nexus. Highest stat = class.

    Even better, only Dex and Con really mattered at all beyond your prime requisite (that’s “stat my class cares about” in basic D&D speech). You could have a terrible Charisma or Intelligence and it never had any mechanical effect on the fighter. You never used those stats for things like skill checks.

    Crap vs. good stats really only became a problem once D&D expanded beyond its roots. Once you add in skill checks, sliding damage and attacks that demand good Dex and Con to survive, and save or die effects in a world where stats modify saves, random char gen falls to pieces.

    If you have that stuff, and your players are rolling stats, you probably really screwed up.

    If you really only care about a highest stat AND restrict the range between average and maximum to a reasonable number (BD&D’s +3 seems about right) then random works really well.

    I think complex systems stack with interrelated mechanics and zero sum stat assignment are what have taught us to hate random character generation, and I think that’s a bad thing.

    (Edited for clarity.)

  5. Justin D. Jacobson

    Tying this in with some of the various DARPG convos here and elsewhere, I’m skeptical of intentional egglessness based on the design of the rest of the game.

    By way of example: One of the oft-mentioned bugaboos about the game as released is the failure to account for Gray Warden PCs. I actually don’t have a problem with their absence mechanically from Set 1. My problem is with the lack of attention to it. The only mention is an incredibly brief sidebar on page 7 of the player’s guide. No mention at all in the GM’s guide. There’s no discussion about building toward it, planning for it, thinking about it from a PC or GM standpoint. I cant’ imagine a person coming to the TTRPG not expecting more spotlight on the Gray Wardens from the point of play.

    It just feels a little rushed to me. I’ll be very interested to see what happens with Set 2. I’m much more optimistic about that.

  6. Rob Donoghue

    Man, I have noting to add to Mike’s comment except to nod like a yokel and say “Yep”. Seriously awesome breakdown.

    @justin A actually worry a little bit about that too. Ideally, the time when you need to start making decisions that lean towards that will be level 6, but there’s no guarantee of that. If this were an indie game I would point to this as a great opportunity for supplemental support, but despite the fact that GR is very aggressive about providing web support, I don’t know if that’s within their gameplan.

    That said, I have also been finding myself thinking about the release schedule, and what will look and feel different after all four boxes are out, for better or for worse.

    -Rob D.

  7. Adaram

    Ummm … I am a little confused. Which part of the Dragon Age character creation is random? Please review the following page …


    Go to the Starting Attributes section, and review. I don’t see anything Random at all.

    It’s probably me being dense, though, so please correct me where I am probably going wrong.


  8. Rob Donoghue

    @adaram Ah, sorry about the confusion. While stats are not random for Dragon Age:Origins, the CRPG, they are for the Pen & Paper Dragon Age RPG based off the DA:O license. (Terribly retro-nerdy, I know).

    My fault for not being entirely clear and just shorthanding it to Dragon Age when there was the potential for confusion.

    -Rob D.

  9. Adaram

    Ahhhh ok thaat makes much more sense now!! I haven’t looked at the PnP version yet — I had planned to in the hopes of learning even more Lore, but some of my friend’s initial impressions were that the PnP version did not add a lot in the way of new Lore.

  10. Chuck

    This is stupidly awesome. I had no idea about that Betty Crocker story.

    It speaks very much toward ideas I have regarding fiction — more and more, it occurs to me that the reader must own the story in some way, and one way to do that is to leave blank spaces. They’re the unseen equivalent to the empty spaces in a game of Mad Libs — let the reader fill in certain details and draw certain conclusions, and they’ll be more invested in it.

    Really, even the old “show don’t tell” advice plays to this, in a way.



    Thanks for being awesome, Rob.

    — c.

  11. Rob Donoghue

    @Adaram Yah, there’s a bit more lore (and a large map of Thedas which is nice to get a sense of where Orlais is and such) but not enough to merit getting the PnP game as a lore resource.

    -Rob D.

  12. Rob Donoghue

    @chuck I think it can backfire sometimes in fiction – some writers make the lead/perspective character as bland as possible to maximize reader identification. It’s a good way to add an egg when it works, but when it doesn’t it’s SO BORING!

    -Rob D.

  13. Chuck

    Oooh, yeah, definitely no-go on the “boring” route —


    This is worth a blog.

    To the blogmachine!

    *dons a cape, runs into a wall*

    — c.

  14. Reverance Pavane

    I’m actually a fan of random characteristic generation for fantasy games. But that’s probably because I start such games with a Campbellian “Call to Adventure” (the characters are knocked out of normal life by some event and then have to resolve the event). I don’t think life is fair and balanced and I detest the sameness of optimised characters.

    Although I do admit that I think most RPGs have it backwards, in that they generate characteristics and then choose classes. Instead you should either generate characteristics according to the characters desired class (my preferred method), or give a bonus to characteristics based on the character’s chosen class and skills (such as a STR bonus for Blacksmith, or CHA bonus for Etiquette for example).

    It worked for pre-Greyhawk D&D because the characteristics didn’t really have much effect on the game (so I don’t get a 10% bonus to XP…).

    Whereas the natural inclination of a player is to want to play a certain class of character, character generation systems should follow that ideology and let the class determine the characteristics.

  15. Mike Mearls

    The interesting thing to me about class first, then stats, is that I think it makes the race-as-class thing more logical. If you decide to play an elf, then roll stats, it feels more like an origin story and less like a career.

  16. Alan Kohler

    Fantasy Craft is my new gaming beau of the moment, and the “egg” I’m busy putting back in is random generation.

    The thing about the attitude that “best stat = class” is that it forces a certain monotony. Even the best players can get a little shoehorned if there is a “best way” to make a certain character type. A little randomization helps mix things up a bit.

    But, to be fair, DARPG sounds like it has a bit too much randomness and too little player control.

  17. Stephen

    Here’s my 2 cents.

    1) Random stats in a game that encourages min-maxing? Just mean and not fair. There’s a bit of a chance of roleplaying and being inspired by the numbers, though.

    2) Point Arrays in a game that encourages min-maxing? Bye bye, role-playing.

    3) Random stats in a game that’s system light and encourages RP? Potentially good, especially if everyone has to do it. You roll before you come up with a character concept, though, otherwise there’ll be some disappointment.

    4) Point Arrays in a game that’s system light and encourages RP? Probably just as good as the random array, plus it lets you come up with a character concept ahead of time.

    For me, I guess it comes down to pre-existing character concepts versus discovering one as you see the numbers appear on the dice.

    Random thoughts, poorly stated, sorry.


Leave a Reply

Your email address will not be published. Required fields are marked *