Robin LawsTalking at Yourself

by Robin D. Laws


Experienced GMs know to avoid situations in which multiple NPCs carry on a conversation. These conversations with yourself are hard for GMs to sustain and for players to follow.

Most of the time you can engineer events so that this doesn’t happen. If one or more PCs are present in the scene, you can have one of the NPCs draw them into the discussion. This slices up a group scene into mini-scenes, mostly between an NPC and at least one PC.

Let’s say the king and his chancellor, both NPCs, are conferring to decide what to do about the bandits up north. The party stands by as they hash the issue out. The king prefers caution; the chancellor, decisive action. Express this not with long snippets of dialogue from each, but with leading questions that draw the PCs into the scene.

The chancellor might throw the question to a hotheaded player character: “Sir Eobald, I see you champ at the bit to put down that impudent rabble!”

The king might address the most cautious of the player characters: “Cedric, I see the rashness of Eobald’s proposal disturbs you.”

This allows you to establish the king’s reluctance and the chancellor’s hard line, without having them speak much at all to one another.

If you find yourself having to avoid these scenes often, take it as a sign of a larger problem. It suggests that you’re letting your own supporting characters upgrade themselves to protagonist status and drive the plot. It implies that the PCs have become spectators. The process of putting  them back into dialogue is just part of the broader objective: to return them to a central place in the storyline. Give them the power to move events. Have powerful NPCs work through them, giving them the leeway to make the key decisions that propel the narrative. Think of the NPCs as supporting characters who bring out aspects of the main cast, whether as foils or antagonists.

Still, sometimes you’ll find events pushing you toward inter-NPC dialogue.

In many cases you can simply keep the conversation brief, boiling it down to a couple of sentences per NPC. Dialogue sequences in roleplaying games go on for much longer than their equivalents in fiction. As we improvise them, we repeat ourselves, include placeholder pleasantries as we buy ourselves time to think, and fumble around in pursuit of the main point. In other words, we speak like we do in real life. Screenwriters in particular compress dialogue to the essential core, leaving in only as much of this extemporization as needed to make the words feel real. Do the same when forced to stage a self-conversation. Unlike a scene between you and a PC, or two PCs, you know what both characters want and what the upshot of the conversation will be. You’re improvising the words but know the outcome already. So compress like a screenwriter and get to it in as few words as possible.
But let’s say the premise of the scene calls for an extended conversation. A PC might be eavesdropping on two NPCs without their knowledge, preventing you from having the supporting characters acknowledge them and draw them into the dialogue. Player characters might listen to an already recorded conversation. Or the story might take you to a situation where it strains credibility for the NPCs to care about what the PCs have to say—for example, when a conversation occurs between two jailer NPCs keeping them prisoner.

A simple trick allows you to turn even these seemingly closed conversations into exchanges—not between characters, but between you as the GM and the players.

Elide these conversations the same way you would the details of a long journey, with summary narration instead of dialogue:

  • “The king and chancellor argue for a long time about your usefulness to the court, and whether he should risk your lives sending you up against the bandits.”
  • “On the tape, Chu and Big Head discuss a laundry list of triad business, most notably the pressure from the mainland cops to replace the rotating leadership with a single gang leader handpicked by them.”
  • “Your captors, clearly not suspecting that any of you speak their tongue, mostly make bored small talk about this crappy assignment and the pleasurebots back on Araatis Station. But they do let slip some guesses about the coming succession war.”
  • “Each of the cultists rises to toast Ephraim on the occasion of his one hundredth birthday. Bloch speaks wittily, Howard with inebriated gusto, while Derleth gives a fussy genealogical discourse. Smith finishes with a poem of cosmic insanity that sends you reeling, blood dripping from your left ear.”

These recaps invite the players to then ask for more details, which you can answer on a Q&A basis, in GM-to-player mode:

Ken (a player): Did they mention the name of the mainland cop?

You: It didn’t seem like they knew.

Ken: Any idea from the tape who might know?

You: Big Head was passing on gossip from his boss.

Ken: (consulting his notes) That’s Uncle Bell, right?

You: Yep.

Ken: Any idea where this mainland cop operates from?

You: Big Head says that Uncle Bell just got back from Guangzhou.

Carrie (another player): Do they say anything about who hit Uncle Gong?

You: Both agree that it was an inside job, but Chu thinks it was a spontaneous mutiny, while Big Head says they must have been paid off by Wai.

Daniela (another player): Do they say anything else interesting?

You: They mostly talk about the leadership situation and the pressure from the mainland. Anything else you’re looking for?

 Sometimes conversations between NPCs just provide flavor, as in the above case of the toasting cultists. But if a player does come up with an interesting question, the answer to which would move the scenario onward, you can improvise an answer to it.

When you have preplanned information to impart, make sure to imply a clear line of questioning the players can pursue to get it out of you. Otherwise your solution to the talking to yourself problem mutates into another classic dilemma, the pixel-hunt.

This month’s topic comes courtesy of “Professor” Ken Thronberry, as a perk of his Badlands Overlord reward tier from the Hillfolk Kickstarter. Thanks for the great question, Ken!



by Kevin Kulp

[Ed Kevin Kulp is working on a GUMSHOE setting involving time travel – this is from his introduction to the system.] General Abilities are how you get stuff done.

Sneaking, fighting, running… all these are done with General Abilities. If you have a General Ability rating of 8 or higher, you’re incredibly talented at that activity (and may get access to cool bonus stuff when using it, depending on which GUMSHOE game you’re playing). If you don’t have any rating at all in a General Ability, you stink at it and won’t generally succeed at non-trivial tasks. A 0 in Driving, for instance, lets you drive to the store and back but you’d fail at any driving maneuvers difficult enough to require a die roll. In comparison, an 8 in Driving makes you an expert wheelman. Similarly, a 0 in Shooting means you’re no good whatsoever at using firearms, while an 8 or higher in Shooting makes you an expert marksman. You get the idea.

It’s traumatic for your dicebag, but in GUMSHOE you’ll only need one die: a d6. Roll it. Your Target Number is usually 4; remember that. If you roll a 4 or higher with a General Ability like Athletics, you probably succeed.

Obviously, that would mean you only succeed half the time. You raise these odds by spending points from your General Ability pools and adding them to your d6 roll. Want to shoot someone? Spend 2 points from your shooting pool, add it to your d6, and you usually only fail if you roll a 1. Spend 3 points and you’re guaranteed to hit even on a d6 roll of 1 (as 1 + 3 = the target number of 4). When your pool drops to 0, you’re stuck just rolling a d6 until you get a rest and the GM says your pool refreshes.

Don’t be shy about spending these points. Dropping enemies quickly is a great idea, and you’ll have chances for your pools to refresh.

Investigative Abilities are how you learn stuff.

They’re what makes GUMSHOE games unique. Ignore your General Abilities for a second and look over at your Investigative Abilities. These are broken into three sections to make things easier to find – academic, interpersonal and technical knowledge – but they all work pretty much the same way. If you have 1 or more points in any of these, you’re an expert at it. This matters because during the game, all you need to do is tell the GM that you’re using an appropriate ability and you’ll automatically get the clue if there is one. Yes, automatically, no roll required. The fun here is in what you do with that information, not how you get it.

So let’s say you’re searching a private library for vital information. The GM may ask, “Do you have any points in Research?” Say yes and she’ll tell you everything you can find out. No roll is ever required. Same thing with Interpersonal Abilities; if you have Flattery, tell the GM you’re flattering someone (or better yet, roleplay it) and it will pay off.

Spend these points to get cool in-game advantages. Take the interpersonal ability Flirting, for example. You meet the evil mastermind’s stunningly attractive protégé. Tell the GM you’re Flirting with the NPC, and he or she will let slip important clues during the banter. Tell the GM you’re spending 1 or more Flirting points to get cool stuff, though, and you’ll get a special benefit; in this case, the protégé may become infatuated with you and double-cross his or her boss at the best possible time.

Just remember, spending a point from an Investigative Ability doesn’t stop you from knowing that topic. It just limits how many times in a game you can ask for special cool stuff.

And really? That’s it. Your GM can tell you anything else you need to know.

13th Age logoby Wade Rockett

As you probably know, Robin Laws is hard at work on the second edition of  The Esoterrorists, his legendary horror game that introduced the GUMSHOE investigative system. I hadn’t planned on running it any time in the near future, because I’m focused on my 13th Age campaign; but I very much wanted to see where Robin was taking the system. I placed my pre-order and downloaded the latest draft of the game in PDF.

I’m glad I did, because Esoterrorists 2e includes a wealth of system-neutral advice on running a campaign full of secrets, conspiracies and treachery…which just happens to be where my Blackmarch campaign is going.

In particular, the Antagonist Reaction section in the Station Duty campaign frame reminded me that my bad guys are not taking  a break while the PCs hunt them down. They’re becoming aware that someone’s poking around in the forest and asking nosy questions down at the docks. They’re gathering their own intelligence, and laying out their options — ranging from “watch and wait” to “kill them now.”

(But, um…don’t tell my players.)

Here are just a few of the Esoterrorist antagonist responses that Robin presents in the game, divided into the three categories of Defensive, Offensive, and Supernatural, arranged roughly in order of intensity. Whenever the player characters make significant progress, either as part of an investigation or in the campaign as a whole, the GM can run through the list of responses and pick the most appealing one. (Cross a response off the list after you’ve used it, to keep the campaign fresh.)

Defensive Responses

Destroy Evidence: The Esoterrorists take steps to erase the evidence of their activities. They might destroy old newspaper articles and Town records or cover their tracks by removing forensic evidence — especially evidence that the player characters are about to find. The investigators arrive at the Town archives just in time to see a fire break out; they find the Esoterrorist ritual site, but, during the night, someone sprays the field with caustic chemicals to destroy any trace of evidence.

13th Age note: Whether there are records to destroy depends on how close to civilization your campaign takes place. If the Archmage, Emperor, Priestess, Dwarf King or Elf Queen lack an institutional presence, evidence of your conspiracy might take the form of eyewitnesses who have to be eliminated in ways that look like an accident. 

Misdirection: The Esoterrorists attempt to deflect the investigation, by providing either an alternate explanation for the horrific events or a scapegoat. Everything wraps up neatly. It’s more than enough to satisfy any mundane investigators, like the police or FBI, so this tactic can remove such assets from the player characters.

The Specialist: The cell calls in support from the wider Esoterror network, bringing in a specialist. This specialist, who might be an assassin, sorcerer, technical expert, government official, or even an ODE, arrives in Town under an assumed name and makes covert contact with the cell. Effectively, you bring in a “special guest villain” for the episode.

Offensive Responses

Buy-Off: The enemy targets one of the Local Residents, offering something the character wants. Money’s too easy; the Esoterrorists are more likely to play on the character’s ambitions or personal needs with some Faustian bargain. That girl you’ve always wanted? Medical treatment for your dying parent? The respect of the whole Town? Anything you want… as long as you promise not to ask any more questions.

13th Age note: Want to buy off a 13th Age character? Look at their One Unique Thing, backgrounds, and icon relationships. Rob Heinsoo once ran a demo where he offered a character who loved the Diabolist the chance to turn her into a good guy — all he had to do was sacrifice one of the other characters. The game ended with one less gnome adventurer in the world.

Break-In: The cell sends a specialist from out of Town, a disposable asset, or just some hired criminal to break into the Station and steal the characters’ records, weapons, or other tools. If the characters have acquired any solid evidence of the Esoterrorists’ activities, like ritual items or the remains of alien creatures, then such items are stolen in the break-in, leaving the investigators empty-handed. This response can be combined with Bugging or Arson to add injury to insult.

Hit Squad: At this stage in the response ladder, the Esoterrorists have lost all pretence of subtlety and dispatch a few armed goons to eliminate the investigators. They’ll cover up the killings later — right now, their priority is to kill the Ordo agents and their local allies.

Supernatural Responses

13th Age note: Depending on the tone of your campaign, you may or may not want the bad guys to save the magical fireworks as a last resort.  Crazed sorcerer-priests of the Serpent God tend to shoot right past “buy them off” and proceed directly to “summon giant snake to eat them.”

Subtle ODE: The Esoterrorists summon an entity from the Outer Dark to spy on or harass the investigators. Such creatures do not present a physical challenge but may still disrupt investigations or cause added problems for the Ordo agents and their allies. Bleeders, death tappers, dream tearers, a Man in the Bar, and scourgers (all from The Book of Unremitting Horror) are good candidates. The Esoterrorists might even loose a sleep hag if they want to eliminate the player characters in an untraceable manner.

Possessor ODE: One of the player characters’ close contacts (maybe even a Source of Stability) is taken over or replaced by a summoned horror. Dementia larvae, sisterites, soliloquies, the Glistening, nesters or skin crabs can all cause behavioral changes or replace human victims. Alternatively, the Esoterrorists can arrange for a creature to possess a human host or call an alternate version of their target out of the shifting planes beyond the Membrane. This possessor does not reveal itself immediately but attempts to subvert the player characters’ investigation or insinuate itself into their trusted circle. If possible, it eliminates the player characters one at a time, perhaps even jumping from host to host as it works its way through the group.

13th Age note: Ahh, possession. Here’s a great opportunity for your players to take the ritual rules from 13th Age out for a spin as they try to exorcise the spirit or demon. Once they become aware that their friend or ally is possessed, you might want to drop hints that exorcism is an option. Otherwise they might just say, “Welp, guess Bob’s a monster now,” and do what adventurers do to monsters.

Overt ODE: Many creatures of Unremitting Horror are savage, destructive monsters. The Esoterrorist cell is unlikely to escalate the conflict to this level until late in the game, when their plans are advanced enough that they can weather the attention that monster attacks will inevitably attract. Potential “hired guns” from The Book of Unremitting Horror include packs of blood corpses, a clootie or even a feral drowner in Towns near water, an organ grinder if you want to hurt your players, ovvashi, residue daemons, or torture dogs. Scaulers also make good assassins.

By Robin D Laws

Though most players find GUMSHOE simple and straightforward, we’ve heard from a few folks who’ve had trouble assimilating it. Usually this happens when they see that it’s a little different from the roleplaying rules they’re used to, and then assume that it’s even more different than it really is. Here are the questions we tend to get from players as they grapple with GUMSHOE, along with the answers that helped them make the adjustment. Use these to guide any of your players who haven’t yet had the cartoon lightbulb of recognition appear above their heads.

Doesn’t the game railroad the players down a single path?

No more so than any other investigative game in which the players attempt to unravel a mystery whose answer the GM has determined in advance. (Nor do you have to determine it in advance) If the only source of narrative branches in a scenario is the possibility that the PCs will fail to understand what’s going on, it’s already a railroad. For this reason, GUMSHOE actually allows you to see the clue path more clearly and construct it to avoid single-track plotting. You do this by ensuring that there are multiple paths to the eventual solution.

In many instances, the feeling that players enjoy freedom of decision-making matters more than the actuality of your plot diagram. A story replete with chances to fork the narrative in unexpected directions may feel like a railroad if the players feel pressured or constrained. Conversely, a single-track plot might feel free and open if they feel that they’re forging ahead and you’re scrambling to keep up with them. When players feel hemmed in or see only one undesirable way forward, the GM may need to point to their various options, showing them that they’re not being railroaded.

Won’t the players just rattle off all of the abilities on their character sheets every time they enter a scene?

No more so than in a game where you have to roll against your abilities to get information. Players who imagine this happen are assuming a much greater difference between the traditional style and the GUMSHOE approach than actually exists. In each case, players always have to describe a logical course of action that might lead to their getting information, directly or indirectly suggesting the ability they use to get it. In the traditional model, there’s a roll; the GM supplies the information on a success. In GUMSHOE, this step is skipped—but it’s the only step skipped.

Traditional style:

Player: I scan the area for unusual energy signatures.

GM: Roll Energy Signatures.

Player: I succeed.

GM: You detect a harmonic anomaly on the quantum level—a sure sign that Xzar technology has been used here, and recently.

GUMSHOE style:

Player: I scan the area for unusual energy signatures.

GM: [Checks worksheet to see if the player’s character has Energy Signatures, which she does.] You detect a harmonic anomaly on the quantum level—a sure sign that Xzar technology has been used here, and recently.


In neither style do you see players grabbing their character sheets as soon as they enter a new scene and shouting out “Anthropology! Archaeology! Botany! Cybe Culture! Evidence Collection!” They don’t do this because it would be weird, boring, and stupid—and because in neither case does it fill all the requirements necessary to get information from a scene.

The only difference is the lack of a die roll. It has a big effect on play, but that doesn’t mean you’re suddenly taking the express train straight to Crazytown.

What if the players come up with a different way to get the information than the scenario specifies?

Give it to them. GUMSHOE always provides at least one way to get clues into player hands. Reward player creativity when they find others. Disallow this only where it:

  • pushes aside another player who ought to be able to use an ability he’s heavily invested in to get the info, and who would be upset to see his spotlight stolen
  • makes no fricking sense whatsoever

In the latter case, work with the player to suggest a more plausible means of using the proposed ability to acquire the clue in question.

In some cases, an unorthodox ability use might require a spend or trigger some negative consequence in the story. In general, though, GUMSHOE is about allowing access to information, not disallowing it. The default GMing style handed down by oral tradition from the hobby’s early days trains us to be on the lookout for actions to disallow. GUMSHOE works best when you always look for ways to say yes.

How hard should I hint if the players are floundering?

As much as you have to, and (ideally) no more.

The barrier we traditionally erect between player autonomy and GM intervention is like any other roleplaying technique—it’s useful only insofar as it makes our games more enjoyable. In GUMSHOE or any other system, frustrated players are generally happy to be nudged back on track, even if you use techniques that would otherwise seem intrusive. Hint as unobtrusively as you can, but hint all the same. When possible, disguise your hinting by using the mechanisms the game provides you. In this case, use your Investigative Ability Worksheet to find an ability that would logically provide the insight needed to see past the current roadblock. Then narrate it as if the character who has the ability has had a hunch or breakthrough:

“Suddenly you remember the phrase your Forensic Accounting professor kept hammering into you: follow the money!”

[Industrial Design] “Maybe it’s the engineer in you, but you can’t help thinking there’s something about that schematic you missed the first time around.”

[Interrogation] “It occurs to you that maybe it’s time to take somebody into custody and ask a few tough questions.”

The extent to which you subtly usher the players along is also a pacing issue. What seems intrusive and railroady in the middle of a session may feel satisfyingly efficient as the clock ticks down toward the end.

What are the common causes of player floundering?

Stopping to ask why players are stuck is the first step to hinting them out of their conceptual paralysis. A few common syndromes lie behind most floundering incidents:

Problem: Someone already came up with the right, simple answer, but it was dismissed or forgotten.

Solution: Tell the group that they’ve already considered and dismissed the right answer.

Problem: The group is stuck in endless speculative mode.

Solution: They need more information. Remind them of this basic investigative principle.

Problem: The group knows what to do, but is too risk-averse to proceed.

Solution: Tell them to nut up. That’s why they get paid the bigcreds.

What if players over-investigate every little detail?

Expect players to surprise you by applying their investigative abilities to tangential descriptive details. For example, as the players explore a palace on a Tudor synthculture world, you might mention that a medieval-inspired tapestry hangs over a wooden throne. The core clue is a residue of alien protoplasm on the bottom of the throne. You mentioned the tapestry simply to add another evocative detail. Now your players are asking you what’s on the tapestry, whether it’s antique or modern, and whether the star pattern shown in its sky tells them anything.

A useful clue that dovetails with the episode’s central mystery might occur to you here. If not, though, you can still treat this as more than a null moment to be quickly dismissed. Instead, treat tangential queries as opportunities to underline the characters’ competence, while at the same time signaling that they have no great relevance to the case at hand. You can do this simply with a “no big deal” tone of voice or body language, or you can spell it out explicitly.

[Astronomy] “You can recall a thousand star systems from memory, and can say right away that the pattern of stars is just an arbitrary pattern chosen by the artist.”

[History, Human] “The images depict an idealized image of Henry VIII—exactly what you’d expect from someone who didn’t bother to delve into the actual history.”

[Chemistry] “What’s it made of? The usual synthetic fibers, exactly as you’d expect.”

What if the player actions suggest a clue that isn’t in the written scenario?

This will happen all the time. No scenario, no matter how tightly written, can provide every answer to the questions players will use their abilities to ask. When this comes up:

Using your knowledge of the scenario’s backstory, think up the most logical answer to the question.

  • Pause to make sure that your answer doesn’t contradict either the facts needed to supply the solution to the ultimate mystery, or any of the core clues along the way. If it does, modify it to fit the rest of the mystery.
  • Supply the info. This might lead to new scenes and alternate ways of gathering the core clues. Improvise as needed to keep up with player actions.

Doesn’t the clue structure make the game hard to prepare for, or to run on the fly?

It’s true that good mysteries are hard to plot, in roleplaying or in other media. You have to be able to plot in two directions, creating both a logical backstory that makes sense when reconstructed, and (as a bare minimum) at least one logical path for the investigators to follow when unraveling it. However, if you keep the backstory reasonably simple, you can rely on the players to provide all the complications and red herrings you need. With this in mind, preparation for a game session can be as easy as jotting down a few point form notes sketching out the backstory and scene structure. Provided you keep the basic details and story logic straight in your head, this very basic structure makes plotting easier, not harder.

In this case, you’re in luck: space opera conventions mean that Ashen Stars mysteries can be simpler than those in police procedurals or horror games.

In my group, we never see the game ground to a halt on a missed information roll, so why play GUMSHOE?

Play it because it focuses and streamlines play, eliminating the elaborate workarounds your GM has to use to make the missed information rolls invisible to you. It replaces these moments of circular plotting with more interesting scenes that move the story forward.


Optional Rule: No-Spend Investigative Spends

Although most groups enjoy the investigative spend rules, a few have reported problems with them. Some players find that the need to ask for investigative spends intrudes too much on the illusion of fictional reality, or makes it too clear that there are certain actions they ought to take during particular scenes.

Here’s another method of providing the flavor clues available through investigative spends, for groups that prefer it. This optional rule is equally applicable to all GUMSHOE games. Be aware that, like most optional rules, this imposes a trade-off you should be aware of before implementation. In this case, the GM takes on a greater bookkeeping burden in exchange for making the system more transparent to her players.

Before play begins, the GM checks all character sheets for investigative abilities with a rating higher than 1. She complies a master list, arranged per ability, ranking the characters in order of their ratings.

Graz Prister has Downside at 4. Clementine Heidegger has it at 3, and Arno Black at 2. The entry in the GM’s master list looks like this:


Graz 4

Clementine 3

Arno 2

Players alert the GM whenever they add to their investigative abilities, so they can keep the master list up to date.

Whenever the PCs enter an investigative scene in which a spend is available, the GM checks the master list to see if any of them could afford to make the spend. The first time this happens, the GM chooses the topmost character, and puts a number of ticks next to the name equal to the size of the spend. During subsequent scenes in which a spend can be made in the same ability, the GM chooses, from among the PCs whose ratings equal or exceed the spend, the one with the fewest tick marks. The tick marks do not represent expenditures; under this system it is possible for a player with 2 points in a particular ability to get two or more 2-point clues, if no one else in the group qualifies to earn them.

This approach doles out the flavor clues in a way that favors players who’ve invested the most points in any given ability, but hides the mechanism from them, so they can’t see the plot gears in motion. It also tends to result in the revelation of more flavor clues.

The PCs are interviewing a witness, a hollow-eyed spaceport hanger-on named Lou. The scenario notes say that on a 1-point spend, a character with Downside will know the meaning of the decorative glowing sub-dermal implant that Lou wears on his left wrist. You, the GM, check your master list for Downside, and see that no spends have been made against it this scenario. So the highest-ranked character with the least tick marks is Graz. You describe the implant and tell his player: “The bracelet indicates that he’s a timestooge—a dupe of a bogus nufaith run by con artists pretending to be temporal travelers.”

Two scenes later, another opportunity for a Downside spend comes up. This is for a 2-point spend, to know that the radiation scars on the arm of a witness were probably put there by the notorious smuggler who loves to brand enemies with a jury-rigged weapon. You check the list, which now looks like this:


Graz 4 ♦

Clementine 3

Arno 2

Graz already has a tick next to her name, so Clementine gets this clue. You then put two tick marks next to her name:


Graz 4

Clementine 3 ♦♦

Arno 2

The GM can either start fresh with no tick marks at the beginning of each scenario, or continue the existing list from one case to the next.

An extract from the forthcoming Botts RPG Miscellany, Adrian Bott’s big book of GM’ing wisdom.

By Adrian Bott


Some Games Masters fear failure – not so much their own, but the players’. For a none too confident GM, the thought that the players might experience failure can be daunting. Failure is generally understood to be a bad thing, so what if the players interpret their ‘bad’ experience as the Games Master’s fault? Will they have an unsatisfying evening, or even want to drop the game altogether? Will they assume the GM is being sadistic, or punishing them for some reason? If games are meant to be fun, then how can failure ever be a good thing?

If you’ve ever had such worries, then please set them aside. Failure is not something to be dreaded. Failure is like a basket of tasty, juicy summer fruits just waiting to be devoured. To understand why, we have to look at what failure in a gaming context means, how it can enrich the game, and how trying to avoid it can make for stagnation and tedium.

Old School Death

Player death is one of those issues that really sets old and new school games apart. There are exceptions, of course, but I’d argue that the general trend has been towards softening the blow of player death and the adoption of a more ‘video gamey’ approach. In the days of early D&D, there were many ghastly ways to lose: you could fail a saving throw and die, you could run out of hit points and die, and you could even have your experience levels drained to the point at which dying probably seemed like a mercy. In other games, such as Pendragon, characters would perish in a blood-drenched variety of mediaeval ways. It’s not really any surprise that the game system Paranoia arose out of old-school gaming, as a satire of deservedly maligned ‘trash the party’ attitudes.

There were a splendid number of different ways for a character to end its life abruptly in the gaming sessions of old. Death was common, though by no means the only option. Permanent insanity was ridiculously easy for a Call of Cthulhu character; and it’s worth noting that CoC occasionally required the sacrifice of permanent Power, a very important stat, to keep the horrors of the Mythos at bay. We got our limbs hacked off in early Runequest, climbed into motionless Spheres of Annihilation and were destroyed forever, fell into boiling mud, broke Staves of the Magi… we played for keeps back then.

In more modern incarnations of D&D, death seems more like a temporary inconvenience. Player loss is more of a setback than a disaster. There has definitely been some borrowing from the way video games tell stories; as we will be finding out later on, a one-track narrative absolutely requires a ‘save point’ defeat mechanic, because no part of the overall story can be allowed to be too challenging to experience. The players simply have to stay on course until they pass through the difficult part, and they can’t do that unless they can bounce back from defeat as many times as necessary.

I believe that the change in how player defeat is typically handled in RPGs is directly related to the shift from open-ended location-based play to linear narrative-based play. Back in the glory days of hex grids and graph paper, players chose their challenges based as much on their own assessment of their readiness for them as on any narrative-derived reason. There was no sense that the environment had to be suited to the players’ strength levels. If you found out that there was a dragon in a cave, you couldn’t rely on it being tailored to give you a balanced fight; in all probability, if you went and attacked it, it would probably devour you in a few rounds. The consequences of defeat were uncompromising, because without the overbearing pressure of an externally composed ‘quest’, ‘plot’ or ‘story’ bearing down on them, the players became uniquely responsible for their decisions and so were not entitled to any sort of insurance in the event of failure. Even if the cause of the defeat was (as it often was) sheer bad luck, that was just the way the dice fell, and there was no point in taking it personally.

In the post-Dragonlance era, things changed. Story being paramount means that player decisions are made in a context of unfolding narrative, in which certain actions are expected of the protagonists. When players take on challenges, it is mutually understood that they are supposed to; this is what the story demands. There is thus an implication that the challenge will be balanced and fair. If they weren’t meant to fight, the thinking runs, then the enemy wouldn’t be in their path. Stories require persistent forward progress, and a player engaged in a story-based game (as the vast majority of games are these days) cannot help but be aware of that.

I won’t pretend there wasn’t a lot to dislike about the old school approach. Emotional investment in a character is more of a risk if that character can be wholly trashed because of a few bad die rolls. Some GMs really were juvenile or sadistic, and took pleasure in ruining select players’ fun by wiping their characters out. The ‘killer dungeon’ is perhaps best left gathering dust in the past. And yet, most players who remember those years can chuckle at the excesses of slaughter and obliteration. By making defeat less bitter and death less severe, do we perhaps run the risk of taking some of the savour out of the game?

Failure is a Great Teacher

If everything that players attempted were to succeed, they would never get to experience the pleasure of adapting and varying their tactics. When confronted by a challenge whose solution isn’t immediately obvious and which could potentially be dangerous, players tend to go into a sort of huddle, discussing what the safest way to proceed might be. At this point, the GM can safely go and put the kettle on, or have a cigarette, or whatever the GM likes to do in those rare moments during which his/her presence isn’t required.

These planning sessions are, however, much less intense than the discussions that inevitably follow a near-total defeat or similar colossal setback. Few things are so motivating as coming within a hair’s breadth of outright disaster. Shocked out of complacency, players will often come up with clever, disciplined battle plans or bizarrely inventive subterfuges.

Although failure can cause ‘blamethrowing’ sessions, in which the players turn on each other and argue about who should have done what, it can also motivate the group to work together more efficiently and achieve better synergy. If the GM is fair and even-handed, then it should be obvious to the players when disaster was caused by the group’s own failure to work as an efficient team.

Failure Can Be Hilarious

There’s a whole media industry based around the comedy of failure. Watch any of the shows where audiences send in clips of things going wrong, or visit sites like Since long before Leeroy Jenkins precipitated the most famous team wipe of our times, humans have found comedy (often black comedy) in disaster.

There are few gaming groups, I imagine, who don’t have favourite stories recounting some disaster or other. (My own D&D group invented a new species, the Flatufrog, to explain an elven ranger’s disastrous failure to move silently through woodland.) If you’d like to read a classic, go and Google for the Head of Vecna; you can thank me later.