Bored of the Tables
Random tables of D100 zany-silly things that might happen in your adventure need to cease, because they've stepped out of bounds.
I get the motivation:
- emergent stories are great, it's what the game's all about.
- dice are great at 'emerging' those elements which become the story - the random challenge, the unexpected situation.
The theory's still sound, the practice remains fun, but so many of the OSR books I see on youtube show-cases and itch.io download step into the wrong arena, and try to pass off underdeveloped ideas by making them into lists and adding a die-roll.
OMG: New Twist!
Wouldn't it be amazing if...?
- The mayor is actually three kobolds in a coat!
- One of the NPCs is a saboteur in disguise!
- The PC finds a book with a dark prophecy written in it!
- ...and other wacky things!
Roll a D8 and find out what crazy shit happens! It's so random!
"Yea, mad", I think for half a moment, while hurriedly pressing the space-bar, as I leaf through another pdf. But when I slow down, double-back, and think through actually using any of this, it's unworkable.
Why does this book assume that my setup has:
- a mayor, who
- lives in a civilization which recognizes the position of being 'a mayor', and
- is relevant, and also
- may plausibly have been three kobolds in a coat all along.
This kind of reveal needs foreshadowing, or it feels cheap. The 'random saboteur' NPC simply does not work with every NPC.
Players should throw the GM a curve-ball, not the book. But if I'm rolling to find out there's a 'dark prophecy', then need to make up what that is, then ensure that it becomes relevant later, and figure out who would write that kind of thing (because clearly, the players will ask) then the list of random encounters has done more harm than good.
These random tables have something in common: they don't suffer repetition well, which is exactly what a random table needs to do. A random encounter table might go like this:
| Roll | Weather | Encounter |
|---|---|---|
| 1 | Lightning storm | 2D6 Goblins |
| 2 | Light rain | Manticore |
| 3 | Clear skies | 2D6 Dwarven Bandits |
| 4 | Overcast | Lost soul |
| 5 | Windy | 1D6 adventurers |
| 6 | Misty | 1D6 griffins |
You can roll 2D6 on this table a hundred times, because it specifies very little. This puts some heavy expectations on the GM to come up with something quick, as the manticore encountered may be hostile or chilled, it may have a plan or just feel so hungry it attacks the PCs on sight. The barren table demands a lot, but the demands are honest: 'you must create an encounter with a manticore on the spot, which makes sense for the scene'. The weather, like the random antagonists, bear repeating. A hundred encounters can occur with 'light rain'.
But so many of the modern tables say a little more, but still not enough:
- An antagonistic woman (really a wight in disguise)
- 2D6 kobolds jump out from a nearby river, riding giant frogs.
- 5D6 giant bats (the swarm has 1 HP per bat, and +2 to hit).
- A vampire from a nearby village, with a fetish for old men.
- A powerful mage, hunting for a magical item.
- A band of 1D6 paladins on a noble quest.
It scans well enough, until you consider rolling a real die. It's tempting to think of 'random numbers' as even spreads, which let you roll '1, 4, 5, 2, 6, 1' ('so random'). But in reality, random numbers (and real dice) will roll '2, 2, 5, 2, 5, 5' (just to spite you). The first group of random kobolds feels 'random'. The second already feels like repetition. The third breaks the spell: all travel involves a (random) river, which has kobolds, all of which travel on frogs.
On first reading, it seems like the table helps create something, while the standard random encounters leave the GM to fend for themselves. But in reality, the zany-wacky-fun-tables stipulate enough to restrict the encounter, but not enough to run it without heavy improvisation. Every time 'paladins on a quest' come up, the GM has to think of a quest, and if three 'paladins on a quest' come up, the players will (rightfully) conclude that something fishy is happening, and start to ask questions which the table provides no answer for.
The tables provide a middle-ground for a short adventure module, and an encounter table, but do badly at both exercises.