Friday, November 14, 2014

Percent Liar

(That's Lying Cat, familiar to fans of the comic Saga.)

I was looking at Arduin for a completely different post idea, and it always brings me back to the funniest statistic in that book: % Liar. Early printings of OD&D put this instead of % Lair. Rather than go along with the errata, Dave Hargrave made this a statistic of its own. The Air Shark, the first monster in the list, has a % Liar of "Too stupid to." And that, my friends, is Retro Stupid.

But % Liar is an interesting idea for a social mechanic (the kind of thing people always say OD&D doesn't have, even though it does). It fits with the aleatoric approach to the game, where random chance is allowed to determine certain key factors. Having a % chance that a monster will lie to you is a perfectly reasonable mechanic given the way that Contact Higher Plane has a veracity percentage to determine whether the result is true or not.

What interests me is that it creates an expectation that some monsters will be more trustworthy than others. Say a goblin has a 50% liar rating while a gnoll has a 30% rating (using the actual % Lair column just for a moment). The trustworthiness of the gnoll is an interesting bit of setting-building. What is it about gnolls that makes them less likely to lie? They are considerably more reliable than a coin flip, which is what the goblins are, but there's still a real chance that listening to them will land you in hot water sooner or later.

There's even a sort of gambling aspect that could emerge from this; once PCs find a particular monster is fairly reliable, they could try to tap the well just enough, pressing their luck that this won't be the time the liar dice come up against them. The lies can also be subtle twisting of the truth, like the rumor about a trapped maiden in Keep on the Borderlands that is designed to trick PCs into "rescuing" the medusa.

This would work well with a two-column rumor table, where one side is true rumors and the other is false, misleading ones. Say a goblin has a 50% chance to be lying; you can roll it organically. If you roll a 1-10 on a d20, you check the corresponding entries on the "lying" table. If you roll 11-20, you check the "truth" table. Gnolls only check "lying" on 1-6. This lets the referee make the lies more varied; for instance, lies 7-10 might be more outrageous, as creatures prone to lie more often will have a tell. So lying rumor 6 might be that the room beyond the statue is empty, when it's actually an ogre's lair, while lying rumor 10 might say it is a dragon's lair.

NPCs could also have a % Liar. We could base it on their alignment; Lawfuls may only have a 5% chance, while Chaotics could be 40% or higher. In general it seems apt to align the percentage to alignment without getting totally out of hand, thus giving a game use to alignment aside from the ever-controversial alignment languages.

Obviously this system places a bit more of a premium on magic and items that help to detect lies versus truth. Finding out about the reputation of various creature types is also a highly useful piece of knowledge, as detailed above. If players know that it is more useful to negotiate with gnolls than goblins, it adds a new strategic dimension when they encounter the more "honest" creature type to parley instead of fighting.

Sure it's a typo, but why not ride it out and add an interesting dimension to D&D?


  1. I took inspiration from your article and went with an alignment-based approach:


    The problem I have with a separate % Liar score is the extra bit of handling as well as tracking another stat, so I folded it into the reaction roll instead.

    1. That's very cool. I like the percentage because it lets monsters get a good/bad reputation.


Comments on posts older than two days will not appear until approved.