D&D 5e: Investigation Encounters


There I was, thumbing through the 5e Player’s Handbook (as one does), when I came a suggestion that non-criminal members of the Rogue class include investigators. Well, I like a good procedural mystery as much as anyone, so pondered how you’d play an investigator through the Rogue class. The baseline Rogue abilities are fine and good – their skill list includes Insight, Investigation, and Perception, while Expertise and Reliable Talent mean they’ll seldom fail a roll. On the other hand, none of the Roguish Archetypes really fit all that well. They feel more like “what you do when you’re not investigating” rather than a reinforcement of your concept, so I think a new Archetype might be called for here.

There isn’t a lot of extant system around Investigation scenes right now, though. (Disclaimer: I do not have the DMG yet, so if there’s useful material on this topic there, I don’t know about it.) It’s not one of the three pillars of D&D – social encounters, exploration, or combat – so all that’s really there is the question of whether a sensing skill (Insight/Investigation/Perception), a lore skill (Arcana, History, Medicine, Nature, Religion), or a divination spell of some kind is appropriate to the situation. The tools are in place, but exactly what they do is a little up in the air.

 

In Which I Talk about Non-D&D Games

The majority of tabletop games aren’t purpose-built to run investigative procedurals, but by golly, I know of one that is: GUMSHOE. Robin D. Laws saw that investigative scenes were the part of any procedural session where things could grind to a halt, because once you’ve failed to find the lead, what do you do next? To solve that, GUMSHOE accomplishes investigation through diceless play – each character has a variety of information-gathering skills, and they spend points out of those skills (that refresh at the end of the adventure) to gain the necessary information.

I played in two sessions of Trail of Cthulhu some years ago, and it worked pretty much fine. We never got jammed up by information failing to flow. I mean, it isn’t compelling gameplay per se; it isn’t trying to be. That part of the system prioritizes functioning smoothly over dice-based tension. Heck, in ToC this is kind of okay – the things you learn build tension just by revealing what cosmic awfulness is in store for you.

When I started thinking about a comparable system for D&D, though, it really bothered me to think of cutting out the die rolls. For one thing, it violates the core game loop of D&D. For another, I’d have to rework all of the relevant skills, practically to the point of having a second skill system. It’s tough, though: how do you allow players to run the risk of failure when actual failure is an intolerable result? They run the risk of death, even TPK, but they usually get to make a lot of tactical choices between initial engagement and untimely demise; rolling skill checks doesn’t offer a lot of inherent compelling choices.

Okay, what do we actually need to have happen? We need the player to take another stab at it, and another, until they succeed. We’re basically okay with failure making things worse for them, as long as the consequences are contained. This brought me to:

  • An Investigation encounter is a new type of encounter. Much like Exploration sequences, Investigation encounters cause skills to function slightly differently, as described below.
  • When rolling any ability check in an Investigation encounter to discover a clue or recognize lore, if the player fails the check, note whether the natural result on the d20 is even or odd.
    • On an even result, the character must spend additional time searching, asking around, or otherwise continuing the legwork of investigation. The DM chooses the unit of time that seems appropriate to the investigation and adventure (minutes, hours, days), and rolls a d6. Once the character spends that amount of time, she may reroll the check with a +5 bonus.
    • On an odd result, the character must spend additional resources: greasing palms, hiring additional aid, purchasing single-use investigative equipment, or paying for the material components of divination spells. The DM rolls a dice value based on the Setback damage of traps appropriate to the party’s level. Once the investigating character spends that many gold pieces in whatever way fits the scene, he may reroll the failed check with a +5 bonus.
    • On a natural 1, the character must spend both time and resources. (Or possibly “double resources.”)
  • After every player takes an action in the investigation (including “Do nothing,” which shouldn’t ever be an objectively right answer), the DM rolls any normal die. On an odd result, the DM takes an action that raises stakes or increases tension.

This system puts some creative pressure on the DM to narrate details into those results, but I am trying to keep it simple enough that a DM could potentially remember the whole rule off the top of his head, or note it in the margins of a DM screen. I may still not hit that mark, I suppose. PCs expecting an investigation will typically carry supplies around with them, for just such emergencies. It’s tricky to come up with what some of these investigative supplies might be in a traditional D&D setting, but I’d suggest that superstition-based hedge magic items put an interesting spin on a game. I do think the story is more interesting with details inserted, rather than a pure handwave response, though I’m also the first to admit that as it gets late, my off-the-cuff creativity dissipates. Also, draining some of the spellcasters’ slots is almost as good as cash out of pocket, if there’s likely to be more encounters before the next long rest, so letting the spellcasters solve this problem sometimes is fine.

If you’re thinking, “Didn’t he talk about this a long while back?” the answer is yes. That implementation had its rocky parts. The main thing I want to reiterate from those earlier posts is that any encounter that doesn’t have a physically-present antagonist needs a trigger for the DM to increase the threat or raise the stakes. Without a clearer cognate of Dungeon World’s 6- results, I’ve added a DM Turn after all players have acted, in which Bad Stuff is 50% likely to kick in. I haven’t playtested this yet, but it’s certainly possible that 50% of the time is too often or not often enough.

The system has a number of usable hooks to make Investigator Rogues better at dealing with it, though I haven’t designed the archetype yet and I don’t know how it might force me to change these rules. We’ll see!

Leave a comment

Your email address will not be published. Required fields are marked *