I’ve been writing an expanded version of my post at Finarvyn’s OD&D boards, which sparked some interesting discussion. The long version will hopefully be in Fight On! magazine #8 – while you’re waiting, go read the other seven!
One thing that doing this made me realize is how much I like using hyperlinks to support digressions and scholasticisms. Not having been able to do that in the essay, here is a repost of the original thoughts about random events, followed by a collection of related links.
I flew back from Gen Con on the same plane as nerdNYC’s jenskot, who turned me on to an e-book by Graham Walmsley called Play Unsafe about using improv techniques to become a better player. I was particularly primed for this, having just played in a game run by my old Ars Magica homie Bob Karcher, whose Second City long-form improv chops let the scenario seem both beautifully planned in advance and responsive to our choices, when in fact he admitted later that he had nothing more than three broad ideas to start with!
Three of the things Walmsley talks about strike me as particularly relevant to old-school play, which uses dice to achieve / reinforce the goals he identifies:
– Always say yes. Accepting that the dice have spoken is very useful training in this. Instead of rolling a random encounter/treasure/etc. and rejecting it as nonsensical, find a way to make it fit. Instead of fudging to make the players succeed, say yes to the possibility of failure and see what the next step in the story is.
– Don’t plan in advance. Having random event tables goes a long way to making me feel comfortable with this. Using raw dice rolls to determine things like morale checks and NPC reactions that can have a huge impact on how a situation plays out forces me to be open to the unexpected. The first time I started using these old-school techniques it was extremely liberating to have a dice roll say to the players and myself “look, I’m not invested in any particular way this encounter might turn out, there’s no wrong answer, let’s all see together where this is going to go.”
– Hold ideas lightly. It’s OK to do some pre-planning if you accept that your plan might never happen. For me, making my own table of random events is one of the easiest and most potent kinds of old-school world-building, and it’s a great exercise in coming up with a bunch of ideas that might or might not be used in any given session.
Lots of the other things in the book, like saying the obvious and using reincorporation to make earlier random events into meaningful closure, also strike me as very useful techniques for participants in any kind of RPG.
Chgowiz’s accounts of games played with these ideas in mind are here and here at his kick-ass blog.
Some evidence that you can trace just about any possible thread of human intellectual history through Major David Wesely appears in the “Fantasy Vietnam” post in The 20′ By 20′ Room blog. A good place to get started on the history of Braunstein and its relationship to Blackmoor and D&D is Ben Robbins’ ars ludi post.
Soon I’ll find a good link to Arneson talking about dungeons as a way to constrain player choice.
Kellri’s CDD #4: Encounters is mentioned in the comments below, and it is awesome, so I linked it here – also the random name generator I used to get Philomena’s name.
Also Tables for Fables, Jeff Rients’ Miscellaneum of Cinder, and the OD&D boards Resources for Randomness thread all deserve links.
Charlatan
Eric Minton
maldoor
Greengoat
David Wellington
James Nostack
tavisallison
The Mule Abides 
It may be a case of self-fulfilling prophecy, but I also think this sort of dice-induced randomness is a huge part of plotless sandbox play. Using dice keeps the Old School GM appropriately “mindless” and reactive, and establishes the game as more of an environment to be explored than a fixed story to be played through.
Two questions arise from this:
1. Eric, what’s the role of random rolls in your game? Tavis seems to really love them.
2. What’s the line you draw on when to roll and what to assume? Although I really liked Geoffrey McKinney’s “Carcosa” supplement for 0e D&D, I thought the guy was a little TOO roll-happy. And I noticed in Tavis’s game on Saturday, the Nameless Patriarch of the Dark One’s attitude toward us was fixed: he was super pissed. The roll was simply to determine what form his vengeance would take. What’s the interaction between conscious choice and randomness–particularly when you’re creating the random table yourself?
Yes, I guess I never explicitly make the connection to sandbox play! I wonder if it’s possible/desirable to run a non-sandbox game with a lot of dice-driven improv. The example of the Nameless Patriarch might suggest why it would still be helpful even if you had pre-conceptions about how events would unfold. In this case, I thought there was basically no chance the Patriarch wouldn’t be super pissed that you’d looted his chest. But when I thought about planning how he would respond, I found that I was getting personally involved – it started to feel like me taking revenge on the players via the proxy of the Patriarch, which I recoiled from. Making a random table did a few things:
– It opened up the possibility (2 in 6) that he would be super pissed but not act on it right away, which I liked because it kept the idea that the PC’s actions have consequences without them being predictable.
– Basing his response on a random roll forced me to be flexible how it might unfold. As it happened, the number you roll was the response I’d first envisioned – but I felt perfectly justified in unleashing it on y’all because it wasn’t me being vengeful, it was the dice! And I didn’t have to worry about making it work out as I’d envisioned; if things had gone differently, I had five other options for working the theme of the Dark One’s anger into the game (even if not all of them would have been visible to you).
– Making the table caused me to do some useful prep – for example, I used Kellri’s Old School Reference #4 to generate the adventuring party the Patriarch would enlist if he went that route. I knew I wanted some more adventurers in the campaign anyway, so having those guys ready would help me be flexible; and the process of generating them made them turn out more interesting than I’d imagined. For example, rolling random genders gave me Philomena, and rolling their spell-books gave me a much more interesting and open-ended approach (her suggestion and her fellow-mage’s invisible spider climb) than if I’d designed them myself to be a strike team.
I think Carcosa is a great example of a supplement written with randomness in mind. Here’s its author, Geoffrey, in response to the original post on the OD&D boards thread: “As a referee, I strongly prefer having the dice decide things rather than myself. It makes it feel more like a fun game and less like mind games that way. I really like this part of your post: “Instead of rolling a random encounter/treasure/etc. and rejecting it as nonsensical, find a way to make it fit.” Accepting the results of random rolls enables me to do things as a referee that I never would have thought of on my own. It expands the horizons of the game. Without random rolls, a campaign starts to feel like the same sort of thing all the time.”
I’m particularly interested in the variable hit dice system – it seems to me that increasing the swinginess of combat reinforces the theme of the supplement. If the outcome of encounters were predictable, going up against shoggoths and the like would be foolish and hanging around the village cellar killing rats would be wise. Making it harder to know how things will turn out encourages you to engage with the setting. You might as well try to get lucky and fight a shoggoth with d4s for hit dice, because it’s too ignominious to be wiped out by rats with d12s.
What were the aspects of Carcosa that you thought were too roll-happy?
For me, the extra-variable Hit Dice, the extra-variable Hit Points, and the extra-variable Weapon Damage were sort of randomness for the sake of randomness. I frankly don’t see much harm in just flatly assigning monsters hit points = HD x 4.5, or making all weapons do 1d6 damage. But, I guess some people have to roll for things: so rolling for Hit Dice meets the need for unpredictability and variation.
So if the goal is just to make things unpredictable, rolling dice the usual way already achieves that end. With Carcosa, I think the goal is to make the setting unpredictably unpredictable, which I feel is an uncertainty too far.
Characters in Gygaxian D&D are very fragile at low-levels, and even at high levels Fighters are pretty wimpy. (All characters in Carcosa are Fighters more-or-less.) This means that the only way to survive is through strategy and tactics. But you can only plan if certain things are known in advance. As you say: the rats could turn out to be hardier than the shoggoths. And Mr. Green’s low-Strength character could end up doing more damage with a dagger than Mr. Red’s awesomely high-Dexterity character with a magical ray-gun.
Not only do these results really screw with a player’s suspension of disbelief, they also render planning almost impossible. As a player, this would just give me a feeling of “learned helplessness.” Why should I try to be clever and daring, when the dice will just undercut whatever I do? I don’t necessarily object to learned helplessness, especially in a game inspired by Lovecraft and Derleth. But I think that could probably be achieved by less “zany” methods, where a particularly valorous house cat eats Cthulhu’s brains.
There’s a lot to like about Carcosa, but that stuff just drives me wild.
Your concern about not being able to plan due to high variability is valid, at least potentially. I’m playing in a Carcosa PbP game at the OD&D boards and so far the variability hasn’t become a major issue or altered my player behavior, but playing online makes the dice less palpably real & we’ve only fought two combats so this is hardly conclusive.
Grognardia linked to Greg Costikyan’s slide-show about randomness, which made the point that the more random rolls you make, and the less importance is placed on each one, the more that regression to the mean makes the outcome predictable. For me, the harm in reducing variability – saying that creatures have fixed numbers of hit points per die, for example – is that feeling that the outcome is already more or less known, which haunts me as a player in the 4E D&D campaign I’m part of.
Yeah: just as there’s a delicate line between absolute gibbering chaos and the good keep-you-on-your-toes uncertainty, there’s also a line between enough-consistency-to-develop-expectations and stultifying invariability.
But overall, yes, I agree with your post.
Bine , eu înțeleg că este posibil să aveți telefon cu lucrurile tale ,
pe într-un buzunar și asta e bine ! Vreau să
spun că sunt printre femeile cele mai încântătoare din întreaga lume .
Care poate fi destul de nefericit , pentru că ea nu este .
Am avut relatii importante , dar cele mai multe din viața mea de adult am fost o singura
femeie . Fiecare persoană are diverse motive pe care doresc să apară minunat .
Te pui la pericol teribil crescute de a face acest lucru .
Choose whether any of your constraints are holding you back or whether you could employ the aid of others to pack in the spaces.
As with any online marketing activity, keep the message short, sweet and direct and be sure
to include the necessary contact information needed
to send in a referral. Learn how to market your home based sewing business online
and in your local community.