Old School Ramblings #1 - Play Now, Story Later

NOTE: This series of articles are about what old school means to me and what I like about old school games. As you can see, my understanding of Old School isn't strictly chronological, and might be a bit unorthodox and opinionated. You're welcome to comment, disagree, and add your own opinions in the comments section.

The definition of "old school gaming" is a contested one. There are many interesting opinions out there, including the ones here. I would like to share some of those and some of my own. Eventually, I will discuss the ones described by Matt Finch in his Primer, which carries a great importance in the definition of OS. Although I might differ in some points, it is a very enjoyable read that I would recommend for anyone interested in OS gaming.

One aspect I really enjoy and identify with old school gaming is what I call "Story Later". This is a way to distinguish OS games from more modern RPGs that are focused in creating a narrative with beginning, middle, end, climaxes, cliffhangers, etc.

(Yes, the term is inspired by this article by Ron Edwards; I think my conclusions are quite different, thought. Instead of writing careful counter-arguments for a twelve-year-old article that has been discussed more times than I can count, I think it will be more interesting to talk about my own experiences. Since I don't fully subscribe to the theory, I see no need to adopt the jargon as well).

Some RPGs have mechanics to enforce a story "flow"; Robin Laws, one of my favorite designers, has created great games based on the idea that a session should be managed with "ups and downs" similar to the ones you can see in a play or a movie, for example. Other examples are Fate RPG and Lady Blackbird, games where the management of meta-game currencies (dice pools, fate points) makes you more likely to succeed after you fail many times and vice-versa. D&D 5e has something similar, to a limited extent, in the Inspiration mechanic.

By Robin Laws - source.

Although those mechanics may sound new, the idea of focusing on stories in RPGs isn't a new one. James Maliszewski, who affirms that Dragonlance "ruined everything", puts it this way (emphasis mine):

"No longer were adventures "modules," implying they could be swapped in and out of campaigns with minimal impact. Now, they had to tell a coherent narrative that was dramatically satisfying."

James points similar approaches in the Desert of Desolation and Ravenloft modules, for example. His stance is a critical one, although he recognizes the success of that approach at the time.

There is an important distinction to be made here, so you don't take this the wrong way. This early "story" modules are described as "railroad-y", meaning the players had little input in the story. Railroading has become some kind of offense nowadays, even though that are still people that don't mind it - although even of those people would probably be appalled by the idea of a PC dying by GM fiat. Now, I don't like railroading, but I will not tell others how they should play their games.

The important thing to keep in mind for the folks that eschew "story" is that railroading is not the only path to story creation. The "modern" games I mentioned above are great example on how to encourage story "flow" without resorting to railroads (some would say that they are incompatible, as there is no story creation if the story is already written). I would even say that this methods are better than the ones originally used by Dragonlance, at least for my tastes. The downside is that the "climaxes", resolutions", etc, aren't guaranteed without previous planning, but clever mechanics may enhance the probability that they should happen at the right time.

I think one way to explain the success of the "story" is that "a coherent narrative that is dramatically satisfying" is not a bad thing. In fact, it is often the way our brain works. We naturally expect such things, and I would dare to say that they can make our games a lot more fun - whether old school or new school.

Here is how Nassim Nicholas Taleb defines the narrative fallacy in The Black Swan:

The narrative fallacy addresses our limited ability to look at sequences of facts without weaving an explanation into them, or, equivalently, forcing a logical link, an arrow of relationship upon them. Explanations bind facts together. They make them all the more easily remembered; they help them make more sense. Where this propensity can go wrong is when it increases our impression of understanding.

The thing is, because our brains work that way, you don't actually need to pursue a narrative in order to create one. As far as my own experience goes, there is a natural pull towards "story" even among people that eschew the idea.

This is a point where I disagree with Edwards, as I think that story WILL, in fact, "consistently emerge from play that does not prioritize it". You see, story will even consistently emerge from our lives, because this is how our brain works. Our lives aren't stories, but simple sequences of events; still, when we think about the past, we cannot help but creating stories.

Whats more, we are creating stories during the events. I am having a change in my job next year, and I have thought a millions times about "how it will end". But it will not "end"; some things will happen, some things will be different, but most things will just continue, No true "resolution" other than the one I create in my head.

I believe this happens during old school games as well. Even if you character doesn't have a name, or dies in the first "scene", you will usually be thinking about what happens next, and will inevitably create a narrative to suit the events.

The difference, then, is one of focus. Old school players, the way I see it, aren't focused on creating a story during play, but they subconsciously create a narrative during the story anyway.

The question that arises is: are there any guarantees that the story will be "coherent" and "dramatically satisfying" enough? For me, the answer is NO, not necessarily - specially if viewed from the outside. In fact, games focused on creating stories might, well, create better stories (and that is a whole different thing, one I can enjoy too).

Think of it this way: I'm a regular guy. If I tell you some stories about my life, it might bore you to tears. Still, these stories are interesting to me because they happened to me. I didn't find a soul-mate, got a job, traveled around, etc, to create stories, but, ultimately, to be a part of something. If I were focused on creating a narrative during the events, I think it might detract from the whole experience; but the stories I can tell about such things are usually easier to remember than most stuff I have read from C.A Smith, R.E Howard and others.

Old school gaming, for me, is like that: it creates interesting stories not because they follow a three-act structures or use effective plot devices, but because we are focused on participating as they as they happen.

We have fun now, and tell stories later.

One table to rule them all - using 2d6 [or 3d6] for everything in D&D

Last week, I proposed a new method for D&D skills (you probably should read the original post to understand this one). Although I admitted a few weaknesses of such method in the end of the post, the most common critic I received was this: "looks fine, but you would have to check a table for every roll".

I hear you. To be honest, I dislike most tables too, specially tables with nothing but numbers, and checking a table for every roll isn't something that sounds like fun (I had my share of Rolemaster adventures back in the day).

Now, the system was created mostly for people that are so used to the Basic modifiers that they wouldn't need to check a table at all. And, to be fair, just by comparing you mod to the DC, you can tell success from failure most of the time. If the DC is higher than you bonus, you must roll at least 13 to succeed, and if its lower you can only fail if you roll 8 or less.

If I were to create a new system from scratch instead of using the Rules Cyclopedia, I have to admit I would go for something simpler, but that's not the point here - I specifically want to use the bits of the RC I enjoy, maybe by streamlining them a bit.

So, my the solution to "I have to check a table" is this: you should only have ONE table, so it will become so familiar that you'll never have to check.

Again, I dislike the idea of "unified" mechanics for combat and skills (I think it is a fallacy; this deserves a whole post of its own, and I'll write it soon - for now, you can read some of my ideas here); on the other hand, I think LOTS of tables could be unified to make them easier to deal with. But, for D&D, the unification should be around 2d6, NOT the d20 as "modern" D&D wants to do (read on). You see, there are lots of rules in the RC that orbit around 2d6: morale, turn undead, monster reaction, weather, hiring companions, etc, and I'm certainly not the first one to suggest that some unification is in order.

So, lets take the monster reaction table. I prefer Moldvay's table than the one in the RC. although they are similar. It goes more or less like this:

	Roll	Result
	2	Immediate attack
	3-5	Hostile
	6-8	Uncertain, confused
	9-11	No attack, monster leaves
	12	Enthusiastic friendship

It is quite easy to extrapolate a general resolution table from this. When rolling 2d6, 6-8 happens quite frequently, almost 50% of the time, so we can take this as the "average" or "expected" result, with 2 and 12 being very bad or very good results instead. For example:

	Score	Modifier	Description
	2	-2	Terrible, disaster
	3-5	-1	Poor, worse than expected
	6-8	0	Average, as expected
	9-11	+1	Good, better than expected
	12	+2	Great, unexpected perks

With this in mind, you can use the exact same table for many different situations, The RC already does this, to a point; "unifying" this chart it is just the logical next step.

For example, the retainer reaction table is very similar to this (yes, I replaced "roll again" for "tries to bargain"):

	Roll	Result
	2	Refuses, insulted (-1 to hire others)
	3-5	Refuses
	6-8	Tries to bargain
	9-11	Accepts
	12	Accepts, impressed (+1 to morale)

You don't need a new table to turn undead; you can use the exact same table. Start with the "average" or "expected" result: the undead are turned. From reading the RC, you can find out what a good or bad result looks like (the undead might be destroyed or immune to turning). The result might look like this (roll 2d6 + cleric level - monster HD):

	Roll	Result
	2	This undead can never be turned by you (disaster)
	3-5	This undead cannot be turned right now (poor)
	6-8	Turn 2d6 HD of undead, minimum 1 (average)
	9-11	Destroy 2d6 HD of undead (good)
	12	Destroy 3d6 HD of undead (great)

As you can notice, you can easily make the water movement modification using a similar method:

	Roll	Result
	2	No wind
	3-5	Breeze (halve speed)
	6-8	Normal wind
	9-11	High winds (double movement)
	12	Gale*

(* I used "Gale" to be faithful to the RC, but since i is a negative result, it should happen on a natural 2; since this table isn't testing a characters skill but only the wind, you can use it either way).

And, again, you shouldn't need a different table for gifting Immortals like the on on RC, for example: just consider the "expected" result (Immortal accepts gift) and go from there.

The advantage of using the same table for everything is familiarity; you might check the table, but the idea is that you get so used to it that you don't need to.

And universal tables are fun!

So why I prefer using 2d6 to a d20 or d6?

First of all, of course, because many tables in the RC are already like that. But another thing to keep in mind is that I use the 6-8 range for the "expected" results. These should happen more often than the other "unexpected" results, which is easier to do with 2d6 than with a d20 - a natural 20 happens 5% of the time, while "snake eyes" in 2d6 happen 2,7% of the time.

And why 3d6?

The thing is, you don't need 3d6. In fact, 2d6 tables would be easier to manage, easier to create (they are already in the RC after all), and 5 possible results might make more sense than seven. The cool thing about using 3d6 is that I can add extremely unexpected results, things that would happen less than 1% of the time for ordinary characters. The truly remarkable, sometimes absurd, stuff. The other advantage, of course, is using the same table for ability modifiers and for everything else. Here it goes:

	Score	Modifier	Description
	3	-3	Abysmal, disastrous, everything is lost, beyond repair
	4-5	-2	Terrible, no chance, try a different approach
	6-8	-1	Poor, little worse than expected
	9-12	0	Average, Neutral, as expected
	13-15	+1	Fair, better than usual
	16-17	+2	Good, got more than expected
	18	+3	Great, impressive success, unexpected rewards

And here are some uses for it (very brainstorm-y stuff, untested):

Monster reaction

	Roll	Result
	3	Attacks immediately, fights to the death
	4-5	Monster attacks
	6-8	Monster is aggressive
	9-12	Monster is cautious or suspicious
	13-15	Monster is neutral, willing to talk
	16-17	Monster is friendly
	18	Monster is surprisingly helpful

Turn undead

	Roll	Result
	3	Forsaken by your deity!
	4-5	Disaster, undead can never be turned by you
	6-8	Failure, undead cannot be turned right now
	9-12	Turn 2d6 HD of undead (minimum 1)
	13-15	Destroy 2d6 HD of undead
	16-17	Destroy 3d6 HD of undead
	18	Destroy 4d6 HD of undead

Retainer reaction

	Roll	Result
	3	Someone you hired will betray your trust
	4-5	Refuses, insulted (-1 to hire another)
	6-8	Refuses
	9-12	Demands better offer
	13-15	Accepts
	16-17	Accepts, impressed (+1 to morale)
	18	You hired someone remarkable; you'll see.

Icy mountains weather

	Roll	Result
	3	Avalanche!
	4-5	Snowstorm
	6-8	Heavy snow
	9-12	Cold as usual
	13-15	Nice day, little snow
	16-17	Snow is melting, sun is out
	18	Looks almost like spring

Spell casting (3d6 + MU level/2 - spell level; use with this if you want)

	Roll	Result
	3	Fumble, Corruption, Madness, etc
	4-5	Fumble, cannot cast this spell again today
	6-8	No effect
	9-12	Spell works as expected
	13-15	Spell doubles distance, duration or # of targets
	16-17	Spell quadruples distance, duration or # of targets
	18	Earth-shattering effects

Next encounter

	Roll	Result
	3	Roll initiative!
	4-5	Next encounter happens in 1 day
	6-8	Next encounter happens in 2 days
	9-12	Next encounter happens in 3 days
	13-15	Next encounter happens in 5 days
	16-17	Next encounter happens in 7 days
	18	No encounter this week, roll again in the next

Morale

	Roll	Result
	3	Unconditional Surrender
	4-5	Rout
	6-8	Fall back
	9-12	Hold / Parlay
	13-15	Press
	16-17	Fight on!
	18	To the death!

Wrestling

	Roll	Result
	3	Fumble
	4-5	Prone
	6-8	No effect
	9-12	Grab
	13-15	Take down
	16-17	Pin
	18	Submission

Ok, I got a bit carried away. As you have noticed, this is quite fun to play with, at least for me. Let's see what you can come up with!

Addendum: dice pools and some math stuff.

The system I proposed last week came, in part, from a dissatisfaction about how the math works when you add a modifier to a d20 roll. In short, the modifier is not very relevant. If you use 2d6+modifier, modifiers are a bit more relevant, but still not enough for me. Obviously, if you use 3d6+modifiers, it is even worse than 2d6+modifiers. 1d6+modifiers is good, but you lose the bell curve. Last week's idea is sound for skill systems, but makes the "unexpected" part of the tables above quite expected for those with big modifiers.

An alternate way of making modifiers matter but still keeping within the limits of 3d6 is using a dice pool. If you have a +2 STR modifier, for example, you roll 5d6 (3d6+2d6) and keep the best three dice. Messing with the pool is easy: take a die away if the task is hard, remove another one if untrained, add a die if you have some advantage, add two if you're an expert, etc.

Even with lots of dice, the unexpected results remain somewhat unexpected, and failure is always a possibility (no matter how infinitesimal). And the cool thing about dice pools is that you can move dice around - you can keep an unused 6 to get a bonus in a related task or add a descriptor to the current one, etc.

One last note: whenever you roll twice in the same table, the first roll should affect the second positively or negatively; see the monster reaction table in the RC, for examples. Thus, you can use the modifier from the first roll to the second roll instead of the ability modifier (since it was already taken into account). For example, if you rolled 14 on the monster reaction table, and the GM tells you to roll again for some reason (better offer, situation has changed, etc), you should roll 4d6 (3d6+1d6). This works very well for weather tables, too.

A new method for BD&D skills

I have spent a few posts explaining why I dislike the "modern" D&D method of rolling 1d20 + mods for skills. Basically, I think it doesn't work; take a look at that post if you disagree, and I would be happy to hear you out.

 Some "old school" inspired methods are obviously better in some regards - 1d6+mods, 2d6+mods, 3d6 roll under, 1d20 roll under, etc. The problem is, you must introduce a new mechanic, DCs, roll low, modifiers, etc. Not a big issue - I prefer adding new mechanics than using some alleged "unified" system that doesn't work for me - but, of course, I like multipurpose mechanics and would rather use an existing one if I can.

There are some good candidates already - the 2d6 reaction table, the thief's 1d6 and d100 skills, etc. Using 1d6 with modifiers (1-2 for hard tasks, 1-4 for easy tasks, etc), or Xd6 roll under (3d6 for ordinary tasks, 4d6 for hard tasks, etc), is also a good idea.

But now I want to propose something else: using ability scores modifiers as the DCs/TNs (Difficulty Class, Target Number, etc).

As you know, Basic D&D uses something similar to this for abilities:

	Score	Modifier	Description
	3	-3	Abysmal
	4-5	-2	Terrible
	6-8	-1	Poor
	9-12	0	Average
	13-15	+1	Fair
	16-17	+2	Good
	18	+3	Great

The system I propose uses the modifiers above as DCs.

For example, a "great" feat of strength would have DC 3. A simple, "average" task will be DC 0. If your ability modifier is equal or greater than the DC, you will usually succeed.

To meet the DC, you rely on two things simultaneously:

a) Your ability modifier;

b) A 3d6 roll that generates another modifier (from -3 to +3) in the same way you ability does.

The sum of a) + b) must be equal or greater to the DC for you to succeed.

It's basically Dice + Modifiers versus DC, with a twist: the dice will generate a result between -3 and +3 instead of 3 to 18.

For example:

To succeed in a DC 2 STR test, you roll 3d6, which in turn generate a modifier from -3 to +3, according to the table above. If you roll 14, for example, you get a +1 modifier. Add that to your STR modifier, and if the total is equal or greater than 2, you succeeded. In that example, you would need a STR of 14 or more to succeed.

As a general rule, if the DC is greater than your ability modifier, you will usually fail, because you are relying on the dice to generate a positive modifier (which only happens about 25% of the time).

Another example: the GM says that picking this lock is DC 4 (a very hard task, indeed). You have a DEX of 17, and thus a +2 modifier already. When you roll 3d6, you need to get an extra +2 to reach DC 4; thus, you need to roll rolling 16 or more. If you had no DEX modifier, the task would be impossible, because the maximum bonus you can get from the dice is +3.

I've added some adjectives (inspired by Fate RPG), but the cool thing is that you don't have too. If you have been playing basic D&D for a while, you already have your own understanding of what "STR 17" means.

With that knowledge, you can just say "it would usually take someone with STR 17 to take down this door". Well, since STR 17 has a +2 modifier (as you probably memorized already), you know the DC is 2.

Keep in mind that the dice will give you no bonus/penalty about 50% of the time, and you will succeed in a DC equal to you mod about 75% of the time. So the STR 17 girl will knock down the STR 17 door three times out of four.

That door has no chance.

Add fiddly bits to taste.

- If you have some kind of advantage, skill, tool, etc, just add +1 to the final modifier (not to the roll). This is not too dissimilar to advantage in 5E or NWP in 2E.

- Use a +1/+2+3 bonus for novice/skilled/expert. So an expert in stealth can ordinarily sneak around as if he had DEX 18, even if his DEX is only 12.

Take all this into account when calculating the modifier needed: if you need not only an expert doctor, but the most talented doctor around, you probably need a +5 or +6 to succeed.

Of course, if you roll better than needed, you use the difference to your next roll, get some kind of advantage, etc. It works both ways.

Mathematically, this works way better than most alternatives mentioned above, specially for conflicts. The STR 18 will seldom get defeated by the STR 10 guy, but the STR 16 guy has a fair shot (in direct conflicts, I would give ties to the strongest, by the way). 

It is also quite easy; the total modifier will often be equal to you unmodified modifier (pardon the pun), and you'll never need to add big numbers. In fact, the DCs will seldom be greater than 3 or 4, etc.

Interpreting the results is near automatic: rolling 14 means and above average performance, and rolling a 3 an abysmal one. When writing adventures, abbreviations come naturally: you can describe a door with "great" resistance, or mention an DEX 16 check, for example.

There are many other methods you could use, but this avoids some of the pitfalls of the first that come to mind (requiring fate dice, harder math, further tables, further subsystems, dice interpretation, etc).

The main strength of this system is familiarity (and the bell curve, if you mind this kind of stuff); if you don't think this is important, you could just ditch abilities completely, keep the mods and roll 1d6, for example.

There is some caveats, though. This doesn't take level into account (if you're not using "skill slots" or similar, you might add a fraction of your level to the roll, not the modifier, so that only the highest level thieves could fool someone smarter than themselves, for example), nor the difference between STR 13 and STR 15, for example. Still, better than most solutions I have seem.

So, what do you think? Do you know a better way to do skills in BD&D? Have you heard this idea mentioned years ago? Do you think it doesn't work at all? I'd love to hear your thoughts.

Coming up... (what would you like too read next?)
- An adaptation of this system for 5E.
- Use this exact same table for everything: reaction rolls, retainers, morale, weather, turning undead, etc.
- Could this affect combat somehow? (Hint: the answer is yes)
- Something else entirely.