Thursday, December 9, 2010

Fast Tables & Slow Tables

Since I'm working on a book chock-full of random tables I've been thinking about which ones to include and why and blahhhhhhh...

Alright. Basically there are 3 kinds of tables:

1) Tables designed to use randomness to create dramatic tension (Slow Tables) These operate by purposefully going slower than the speed of thought. They give the roller less information than s/he needs to figure out all of what's going on. Roll initiative. Tension mounts. Roll to hit. Tension mounts. Roll damage. Tension mounts. Roll to see if you save vs, poison. Tension mounts... aaaaaand you took 3 points of damage and are paralyzed on account of the Toadweasel's poison dagger.

It took 4 rolls to figure that out & it was worth every second. (In theory.) You could, in fact, see the whole game as basically rolling on a series of tables that slow the story down and build tension. "Do you, Vrothgrontnar, want to take the Chalice of Especial Destiny across the Plain of Choogle to the Fire Queens of Northern Sucktania?" "Yes, yes I do." And you could just end it there. He did it. But instead, you roll on some tables and do some other things to slow the whole thing down and break it into parts with details.

If, instead of actually playing, you just did all the math on Vrothgrontnar's chances and built a d1000 chart and had him roll on it and then read off a result which included a rich, novel-length description of a whole adventure he just had: "Well, first you woke up and brushed your teeth, then you went and got some henchfolk, pits...gargoyle feast...three wicked sailors...Gates of Skrowwbe..."etc. could possibly provide exactly the same amount of information as the chronicle of the game. But it would have no dramatic tension as a game (and no place for choices, but that's another post).

Anyway, point is, in this case, the table--or, usually, "series of tables"--is used to provide small details which, in themselves, mean little, but which slowly accrete into a story.

2) Tables designed to use randomness to quickly choose between a very large number of plausible options so that the game can continue smoothly (Fast Tables) These operate by purposefully going faster than the speed of thought. They give the roller more information than s/he could have invented spontaneously at that speed or if not that, then at least they allow the roller to have options presented to them with a level of regularity matching reality, genre expectations, etc. that would require a lot of thought s/he couldn't have put in that fast (i.e. I don't roll up "Polar Bear" on the wilderness encounter chart because I couldn't have thought of a polar bear that fast, I do it because I trust the table to be balanced to provide me with polar bears about as often as I or the game designer think polar bears should turn up on the pitiless tundra of Harshlandica.)

These charts should provide the roller with the most information s/he could possibly assimilate at a glance and are mostly useful for things that the game feels should/could be randomized but aren't all that dramatic ("What's the layout of this building?") or for DMs in the middle of a game.

3) Tables designed to handle extremely large numbers of options during DM prep. An example of this would be the Random Magic Item tables from the DMG. The emphasis in these tables is, theoretically, providing a statistical balance of whatever element the DM's trying to roll up. These are the easiest to design since they don't have to weigh statistical accuracy or regularity or verisimilitude or whatever they're going for against anything else. People who use them are usually doing it because they like to and have free time on their hands.

Making them faster is possible, and occasionally desirable for a busy DM about to run something. When trying to design faster tables of this sort you can pretty much follow the same rules as Fast Tables above.


Ok, so a series of Slow Tables should provide a stream of details at a rate that keeps the tension high. A single Fast Table should provide a ton of information all at once so that situations can be nailed down quickly and everybody can move on to the fun stuff.


Designing a good series of Slow Tables usually means that each roll provides some concrete & identifiable detail about where the situation is going which means something to the PCs in terms of good or bad. "Initiative" can go well or poorly, "to hit" can go well or poorly & "damage" can go well or poorly, for instance. A Slow Table which just goes "Ok, roll to see who notices...ok, it's someone lower class...ok, roll again...ok, its someone in the service industry...ok, roll's a charwoman!" is, most of the time, not going to increase tension, it's just slowing shit down. Might as well roll your random innocent bystander on a Fast Table.

Probably the classic example is the hit location table--the problem is not that it's a whole extra die to roll, the problem is working it into the "story" of the combat system tells in such a way that you don't hear "upper arm" and just go "Ok, and...?".

Whatever version of Rolemaster I just played handled that pretty well by folding hit location into a system of high, medium, & low crits, so you hear about what you hit and what that did at the same time.

In other words, don't design a Slow Table to slowly narrow down categories unless those categories mean something to the PCs in terms of what they're trying to accomplish.


Designing a good Fast Table means extracting the maximum information from each die roll. This usually means that you put tons of effort into creating unique results for the table when you're not playing so that all that creativity pops up instantly when you are playing. A d100 chart with 100 results and 1-2 line descriptions of each result is a typical good example, but there are other ways to do it.

For example: Factoring in even one of these other conditions in addition to what's on the die can allow you to parse between far more than 100 options on a single roll:

-what PC rolled the die (race, class, male, female)
-who rolled the die (player position-1 to the left of the DM, 2 to the left, 3 to the left, etc.)
-where the die lands
-color of the die
-size of the die

For example, you could stock a hexmap in a second by assigning different dice (different color, different size) to various monsters likely to show up in a certain biome. D20s would be the weakest monsters and d4s would be the toughest. Then just drop all the dice onto the map (behind the screen). Wherever a die lands, that many monsters of that type are there. Any dice that don't land on the map aren't there. When the PCs approach a hex, you can look it up.

Combat is usually a Slow Table thing. However, there are times a Fast Table could come in handy: mass combat, naturally, and in situations where the DM is rolling several enemies attacking simultaneously & the combat hasn't progressed to the point where PCs are scared that every time a die hits the table they might die.

Character generation is usually in the game-as-written as a Slow Table thing (with options between rolls) but there are lots of times when a player or DM would want it to get done as a Fast Table thing. Games can have both--a slow system for excited newbies & character builders, & a fast system for replacing dead PCs midgame and generating NPCs on the fly.

Here's one for D&D if you have lots of dice: assign each color of die an ability (dex, con, etc.) and roll them. If you're real lazy you can roll 6d20 and just re-roll or rationalize anything outside 3-18.

A common annoying thing in Slow Table character generation is later results which repeatedly void the results of previous rolls. Like I notice in Teenage Mutant Ninja Turtles that half your ability scores don't mean anything when you first roll them since you renovate them constantly at every stage of the process (you're a raccoon? +5 pp, -3 ps. What size are you? Oh, that's a +4 ps) so it's hard to get a sense of the PC taking shape in your mind. (In Paranoia they do this on purpose, but that game's really more like a bunch of Fast Tables stuck together in order to disorient the PCs.) I'd suggest that if you're designing a game where the physical form of the PC is wildly in doubt that you get most the powers and ability-altering weirdness in first so that when you do get the ability scores then they actually feel like you're deciding something.

Treasure is usually a Fast Table, but you could build Slow Tables for treasure (or just about anything) if you want to use the tables to narrate the process of finding the treasure. The trick is the table reveals things from the PC's point of view. Like: "We search the body"(roll, roll)"Ok, the 'What's does it have in its pockets? table' says a cursory pocket search reveals nothing. Do you want to roll on the 'What happens if I take off its clothes?' table?""Ok" (roll) "There's nothing but a dagger and a crude tattoo of a dancing hog under the clothes, but there appears to be an unnatural lump in the creature's throat, do you want to roll on the 'What's in it's throat?' table?'" etc.

Or: "It's...(roll roll)...shiny," "YAY!" "and...(roll roll) about 3 feet tall""YAY!""...and...(roll roll) it's a giant slug and it's alive!""Fuck.".

If you build Slow Tables you might end up spending a lot of time not using them. The best Slow Tables are specific to a certain narrow situation but also re-usable, which is a pretty tough bill to fill. Attaching them to a thing the PCs end up bringing around with them (like a car or a horse or a spell) is usually not a waste of effort.

Aaaaand...that's it for now.


  1. I've been playing around with a kind of fast table called a dice map: roll dice on printed diagram, interpret the result of the dice based on where the dice land. Lately, I've been working with a six-wedge map based on the six attributes, which can be interpreted in different ways. Lots of examples of using that map in my dice map category; I'm working on revising the map to include more hints on alternative interpretations, to avoid looking up results on multiple tables.

  2. @Talysman
    there are some similar ideas in the upcoming Kit, and also here:

  3. What you're calling a "Slow Table" I just call rules ... a "Fast Table" (what I call just a "table") is indeed a way to encapsulate something that would take longer to do the rules way.

    It's also a way to ensure your idea as a DM of what happens next is fair, random, and can surprise you as much as the players.

  4. I don't really like what you're calling 'slow tables'. I understand Roger the Gs's thinking but I also get that in some games, like D&D, attack and saves can be determined on a table.
    Unfortunately, I've never been a big fan of that concept.

    I'm really liking the One-Roll-Engine where a single roll determines hit and damage but still has variables and variations depending upon what came up on the dice. It simply keeps things moving for me and still gives me (as GM) room to embelish.

    I'm finally getting to work on my own self-published game and I'm thinking of a similar mechanic for it, though I haven't decided yet.

  5. Very nice! I'd never considered the view from up here before... :)

  6. Began a minor headache as I read through table design theory 101...skipped to the comments section to add this:

    The bottom line question for you Zak, I think, is what would make me buy a book of your tables? Well, I begin my decision making process with with "What do I like about Zak's games which would present well in tables?" Answer: The random weird quirky creative shit, that I could use in my own games.

    Tables of random mushroom effects. Tables of snakes and what they say if run through that weird snake reader thingy. Bottom line, anything different that has your own creative stamp on it, that makes me say "Yup, that's a Zakian table all right."

    I take that table, insert it into my game, and add a bit of you and your flavor into my campaign. The players get to feel a taste of what it would be to play in your game.

    Tables of just random gems and other boring mundane stuff would not interest me in the slightest. We all have countless resources for that, starting with (and perhaps ending with) the original Gygax DMG.

    My 2 coppers.

  7. @joethelawyer

    remind me to one day write something that that comment might conceivably be a response to.

  8. no prob. like I said, I read this:

    "Since I'm working on a book chock-full of random tables I've been thinking about which ones to include and why and blahhhhhhh..."

    spaced out on the theory section, and decided to add to the part that I digested, the "thinking about which ones to include" part.

    Good luck with the project

  9. Man, i would love a Fast Table for when monsters outnumber PCs... 2:1 fights are okay, but when it gets to 3:1 and 4:1, I want results fast. Maybe it's not the right thing for Vornheim, but if you ever figure out something like that, I'm in.

  10. @Odrook

    Totally got you covered, baby.

  11. I second the interest in the Fast Table for outnumbered combat. I've seen a few attempts at something to streamline the GMs side of combat like that, but in my experience they tend to be excessively deadly to the PCs. I'd love a balanced, functional Fast Table for that.

  12. Excellent analysis Zak. I learned something today because of you, and that makes me very happy. Like trollsmyth, I'd never looked at tables this way. I'll be thinking in these terms when I design and utilise tables (or table-like situations) from now on.

  13. Oh man. I've never thought to consider the different types of tables and exactly what im trying to do with them. A very elegant and accurate solution I think. I have since reviewed a bunch of my own random tables and noticed that they fall into those categories... and now I must adjust a few of them so that they do what I want them to.

    Thanks again for your wonderful creativity and insightful analysis!

    Also: The game system I'm making totally has a slow and fast method of character generation (Story or Easy mode)... Now everyone is going to think I lifted the idea from you. You've ruined everything. Bastard. =]

    ALSO: Toadweasel. So good. Clearly he is a repulsive and horrible cellar-dwelling knave. <3