A sign like this, however removed culturally, can really make players paranoid when they encounter it in a dungeon
…and if you look really closely….
Sign Image from WikimediaImages via Pixabay,
Beartrap image by Clker-Free-Vector-Images from Pixabay,
Cave image by kobitriki from Pixabay,
color modified by Mike
 
 

I’ve always had problems with traps in D&D.

Some of these problems have their origins all the way back in AD&D, others are more recent in origin.

Rules changes with the different editions have solved or mitigated some of my concerns while creating whole new headaches to take their place.

I have solutions to these problems. Some have been in place in every D&D campaign I’ve run this century – these are all on the GM’s side of the screen and have no direct impact on the players (other than yielding better adventures, of course). Others are new and have yet to be tested in a live-fire environment.

But, let’s start by reviewing the problems that these solutions are intended to address.

Problem #1: Trap Placement Makes No Sense

Internal logic used to fall a long way down the list of desirable attributes that a dungeon should posses. Challenge, Loot-vs-Danger ratios, a Narrative through-line, Responsiveness to Player Stimulus, Environmental Rationalism – these were frequently considered more important. Every village had a dungeon, sometimes several. And every dungeon had traps; that’s just the way it was.

• Challenge: Dungeons should not be completable by bazorting everything that moved and looting anything that stayed put long enough. There needed to be some input from a stray brain cell or two, as well. At the same time, it was very bad form to make completion of the dungeon dependent on a player thinking the right thing, solving the right puzzle, or making the right blind choice – and even worse to have it come down to a PC die roll. These competing interests created a lot of debate and discussion back in the day when all was shiny and new.
• Loot-vs-Danger ratios: These were also the subject of intense debate back in the late 70s and early 80s.
  • There were those who advocated a strict policy of risk-vs-reward, assigning a linear relationship between dungeon level or character-level-required and these variables.
  • Others advocated a fuzzier relationship (and this is the faction that eventually came to dominate thinking, because the Monster Manual and it’s treasure classes supported it).
  • A few people applied mathematics – every monster HP was worth so much in treasure, every monster Special Ability was worth a magic item, and an exponential or geometric relationship existed between stacking of such things and upgrades to these in number or value.
  • A few people thought it was fine to be niggardly on the upper levels of a dungeon if you were more generous to an equal extent in the lower levels.
  • Some people observed that this made the lower levels much more fun for their players and simply deleted the niggardly upper levels from their designs – and thus the term “Monty Haul” entered the gaming lexicon.
• Narrative Through-Lines: I subscribed to this priority from word one. It means that there was a story to the dungeon, a history that the players were becoming part of. To understand that history was to acquire clues as to the solution to problems that might otherwise be insoluble, because the history was used to place everything within the dungeon.
• Responsiveness to Player Stimulus: This simply meant that the patterns of behavior within the dungeon changed in response to the actions of the PCs. You couldn’t take out an Orc Patrol without triggering changes in the behavior and status of certain Orc encounters later in the Dungeon, for example. This is often harder than it looks to do well – how does killing an Owlbear on level 3 impact a cult of Troglodytes on level 6, for example? Who gets additional resources, and what do they do about them? This involved perceiving the dungeon as a dynamic environment with its own social and political structures, which would react and respond in response to every action by the PCs. I once met a GM who had been working on the one dungeon for more than 20 years; his notes on the internal realpolitik ran for more than 600 pages. It was a work of art – and totally impractical to play. But less complex interweavings added massively to every aspect of a dungeon.
• Environmental Rationalism: Ever found a 10′ by 10′ room containing an ancient Red Dragon and his hoard? I have. Ever wondered how you could breathe at lower levels? How much air does a Beholder need, anyway? What do all the residents eat? Solving these problems creates a Dungeon ecology, and is a way of imposing rationalism and rationality upon the irrational. There was a time, stimulated by a series of articles in The Dragon, when this became the dominant design goal of GMs all over the place, because the dungeons immediately became more believable, more interesting, and immersion immediately increased.

I’m sure you’ll agree that all of these are worthy of consideration, if not deep reflection. But by the time you get to the end of that impractically-long list, if you even got that far, Trap Placement seemed pretty irrelevant. Stick ’em anywhere, and Don’t – at all costs – let them become a trail of breadcrumbs to the major loot!

Problem #2: Spotlight on Thieves/Rogues

Thieves, in the AD&D paradigm, had to be good at everything that might be considered the involuntary transfer of ownership. Sure, you would be better at some than others, but general competence in all was part of the package. And, if you’re already skilled at picking locks, surely finding and disarming traps wasn’t such a difficult leap?

Besides, this was a time when Thieves were a LOT less effective in combat than Fighters. Many GMs (myself included) subscribed to the notion that you couldn’t Hide In Shadows if you were under direct enemy observation – you needed a meat-shield to create a distraction (Spellcasters were more effective at creating a distraction but too vulnerable to last in that capacity).

[Related question: How many thieves can Hide in a single Shadow? Answer: An infinite number – according to the rules, at least by implication, because no limit is specified!]

Placing traps meant ensuring that the Fighters and Mages didn’t get to hog the spotlight. I’ve even seen it suggested on some bulletin boards that there should be a fixed ratio: so many traps at such-and-such a difficulty modifier for every Hit Die of monsters present. Others suggest that the Dungeon be allocated a budget – every HD of monsters comes out of that budget, and whatever’s left gets split up and placed as Traps.

Somewhere along the way, it became accepted that Thieves could Hide in one Shadow, and then move from that shadow to another without breaking cover if it was within their normal movement range. Dungeon environments, hardly models of excellent illumination, encouraged this thinking.

With that, Thieves – now called Rogues – became scary-good at combat, when used properly. They might be slow-firing, having to Hide and then Move before striking, but they could inflict massive damage every two or three rounds; all the fighter-types had to do was pin an enemy down in the meantime and keep that enemy’s attention on them. Paladins, by virtue of their always-immaculate armor, and Barbarians, because they were all shouty and in-your-face, made excellent choices as rogue meat-shields. And the major justification for this treatment of Traps went away – but design practices didn’t change.

Problem #3: Trap Resolution runs counter to good roleplay

When a fighter encounters a strange creature, there follows an intense ballet of inspection and debate even as a flurry of blows are unleashed. What special tricks has the GM loaded the creature with? What trickery (pardon: “clever design technique”) is he going to use, this time?

Are you better off drawing your broadsword or holding onto your torch and waving it in the creature’s face? Is it vulnerable to cold, or heat, or lightning? What are it’s capabilities? Every combat round presents new evidence, yielding new insights.

In a word, there is engagement.

Throw in tactics, and multiple opponents, perhaps mixed groups of opponents, and there was plenty to keep Fighters, Mages and Clerics occupied and in the spotlight.

Every turn, that spotlight would swing around to the Rogue. And what was his involvement? Most of the time, it was at best “I’m still hiding in shadows and maneuvering to get behind [target X].” – or, worse yet, squealing like a stuck pig when forced into the front lines, “I’m a frikking thief, not a fighter, damn it!”

The Thief was most useful in between combat encounters – exploring, scouting, mapping, searching, climbing – and most of that simply got hand-waved. Quite frankly, there were times when the thief could have been completely forgotten were it not for the occasional trap to put them into the spotlight.

On those “rare” occasion that the searches turned up something – a trap, let us say – what was the usual response? “I attempt to disarm the trap,” followed by a single die roll. A little exposition from the GM and the “encounter” was over. Compared with the multiple rounds of a combat, this is anticlimactic at best.

There’s no engagement. There’s that word again.

Some GMs tried to counter this by demanding the thief describe what he’s doing. The first few times, the character might do so off the top of the player’s head, but before long a standard list would materialize, to be extended and annotated when necessary. “Push, pull, slide left, right, up, down, turn it to the left then the right – just a little, just enough to know which way it operates….”

The third or fourth time you hear this list recited verbatim, the temptation to hand-wave it becomes overwhelming because you know that this litany will happen every single time thereafter. And you’re suddenly back at the single die roll again.

What’s more, extensive descriptions of the traps had to be prepared by the GM to support these detailed action requests. Every time. That’s easy at first, but by the fifteenth trap, it’s a lot harder to be creative and original – or to sound interested. Going back to the single die roll relieves the GM of that burden.

Problem #4: Save Or Die

Okay, so there’s a trap. The rogue is about to attempt to disarm it, but isn’t sure of success. If he’s got it right, the game continues; if not, one or more of the PCs could be killed. It all comes down to this die roll….

How is that different from a save-or-die spell?

As I’ve made clear in the past – see Beyond Unreasonable: Challenge Failure Modes and Exceeding the Extraordinary: The Meaning Of Feats for discussion (“rants” is impolite, but possibly more accurate) on the subject.

The more narrative-oriented your campaign is, the greater the risk of it going off the rails because of save-or-die circumstances. The only obvious solution (there are inobvious solutions but they take me a long way off-point and I don’t have the time) is to eliminate save-or-die situations. Well, that’s fine in terms of Mages and Clerics, but where does it leave the Rogue?

Answer: Making a single die roll in which the outcome doesn’t really matter.

My, what a long way we’ve come.

So, let’s talk solutions to these problems.

GMing Solution: Moderation is key

First, let’s talk about Trap Placement. All those considerations of Dungeon Design are worthy of consideration, and I never design one or emplace one without giving each some thought – but only as means to an end.

I start with the History – but this is not a fully detailed, specific item, it’s a paragraph per level of broad outlines that I produce in synopsized form. This gives clues to the architecture, original content, and style of the dungeon. In particular, I contemplate the ergonomics of the original architects:

• Ergonomics and the Non-human,
• By Popular Demand: The Ergonomics Of Dwarves, and
• Studs, Buttons, and Static Cling: Creating consistent non-human tech.

I then move on to the Ecology & Environment – refer to my specific articles on the subject,

• Creating ecology-based random encounters: The Philosophy of meanderings,
• Creating ecology-based random encounters: This Eats That,
• Creating ecology-based random encounters: Encounters with meaning, and
• A Population Of Dinosaurs and the impact on RPG ecologies.

This identifies current “occupants” and their habits.

I then employ these as context in determining trap placement – every trap should have a purpose, and should not interfere excessively with a more important function of the structure as it was at the time.

An especially nasty trick: traps that can only be disabled if a specific trap or two early in the dungeon have NOT been disabled. Forces the party to go back and reset it, exposing them to incidental dangers. What’s more, since they can’t assume this is the only one, they have to reset them ALL.

Each new occupant of an area will change it to fit their needs. This may include modifying or improving or breaking/removing traps (The first time a rogue identifies a space for a trap only to find that they can’t access the mechanism, watch the paranoia ramp up! YOU know that the mechanism has been removed, but THEY don’t…)

The history and ecology will give you some idea of who inhabited the space after the original designer, and how they lived, and hence, provides a guideline to the changes to the Traps. In particular, watch out for poison changes, the addition of diseases as vicious complicating factors to traps, disabling and resetting of traps, the installation of new traps, and misdirections.

More intelligent designers may well have had a deliberate “trap strategy”, EG a series of easily-beaten traps to lull intruders into a false sense of security, traps that are intended merely to delay or inconvenience intruders, traps to channel intruders into position for ambushing, etc.

The more successions of occupancy, the more this original strategy will have been diluted, though intelligent occupiers may well have preserved it – and learned from it. Sure, it may be more convenient to off intruders before they get their hands on the loot, but getting them on their way out is better than nothing!

The result: sensible placement of traps.

GM/Player Solution: Break Down That Die Roll

Next, let’s deal with that “one die roll” problem. The answer to this comes from TORG: Divide the ‘traps process’ into four parts (five, if you count detection in the first place).

1. Inspection
2. Analysis
3. Theory
4. Application

1. Inspection

The Inspection action is all about identifying the key features and components of the trap. Fail, and you will miss something – a safety mechanism, a secondary trigger, a secondary effect, or simply some piece of information that you need in order to disarm the trap. Failure by more than 5 should mean you’ve missed 2-4 items on the list.

The “expert” may be aided by other characters searching under his direction. Unless these have their own “Disable Traps” skill,

2. Analysis

The Analysis action compares the identified features with theoretical knowledge of a variety of traps, the goal being a complete understanding of what the trap was designed to do and how (in general) it does it.

Any sort of mechanical engineering skill or equivalent should add +2 to the die roll for this action. A success may be able to overcome a failure in the Inspection phase by identifying a feature that should be present but hasn’t been found. How the GM handles this is up to them, but I recommend that attempts to re-roll the Inspection phase on this basis should incur a -5 per attempt after the first (or +5 to the DC per attempt after the first, or +20% to the target or die roll), depending on how your game system describes such things.

The above was originally without the “after the first” qualifier, but I realized that it could be argued that this was a roll to fill in missing details, not to start over from scratch, so it should be easier. But, if you fail again, knowing what you are looking for, the difficulty ramps up very quickly. Once you reach the point where only a critical success will achieve the task, further attempts may not be made.

Every suggestion by another character as to how the trap might work adds 1d6 plus the INT mod of the character making the suggestion and subtracts 1d6 to the characters skill at achieving this step. Some GMs may permit the substitution of WIS mod for INT mod if the character is attempting to apply “common sense” to the discussion. The negative reflects the possibility of sending the character’s thoughts off in entirely the wrong direction, i.e. hindering the process. These modifiers may be applied after the initial roll by the rogue but before the GM details the results – which might then become “You thought you had it figured out, but after considering Magnu’s suggestion, you aren’t so sure any more”.

3. Theory

In the theory phase, the Rogue attempts to deduce the point of greatest vulnerability of the trap, given the resources available to him. No-one can help him do this. However, the rogue can ask specific questions of a character with appropriate knowledge that may earn him a +1 to his attempt, at the GM’s discretion.

For example, knowledge of the culture that built the trap originally, or modified the trap most recently (if those are known) may contain a hint as to the way they do things.

The character asked rolls against his knowledge and the answer either adds or subtracts 1 depending on whether or not the roll was successful. If the GM considers the question irrelevant, the result is +0.

Each such question after the first incurs a +5 difficulty penalty to the “expert’s” die roll – so if you ask too many questions, you will definitely begin impairing your ability to succeed.

4. Application

Once you think you know how to disarm the trap, you enter the Application phase, in which you apply your theory. If the GM deems your activities appropriate to solving the problem, a successful roll in this phase disarms the trap; if not, you may think it’s been disarmed when it has not.

Note that the solution may draw upon more than one party member; while the rogue has his training and intuitive sense of such things to draw upon, he may require more raw force to be applied than he can muster (cue a fighter with an appropriate tool), or heat or cold from the mage may be necessary.

Another nasty idea: A trap that is disarmed by moving a hidden control in a certain direction, but which triggers a second trap if moved too far. Most rogues stop searching after the first success. Though the fact that some of the elements identified in the Inspection phase don’t seem to have any purpose after Analysis may offer a clue.

What this is all about is (1) bringing about engagement in the process of disarming the trap, making it more than just a die roll; and (2) involving more party members than just the rogue in that process to further encourage that engagement.

An example

I had a very detailed example thought out, but time is getting away from me, so the illustrative diagram I intended to provide has had to go by the wayside. Here’s a simple top-down map instead:



The results of a successful Inspection roll. A is the top step of the staircase into the room, which contains a pressure plate. B is a thick slab of stone with steel bands reinforcing it; the slab weights 32 tons and rests on two steel rails (C). D is a relief sculpture on the wall with a number of suspicious-looking holes suitable for the firing of poisoned darts. E is a pair of pedestals recessed into the wall which also contain pressure plates. F is a series of brass urns of different sizes full of sand; G is a matching set of iron urns full of honey. Finally, H is a large lever that is set into the floor.

There’s a lot here to Analyze, but geometry offers some clues. The lever, H, is locked unless someone is standing at A, fully exposed to the darts from D. Throwing the lever releases the darts unless the trap’s safety mechanism is activated. This obviously involves the pressure plates (E) and the urns (F and G) – put the right urns on the right plates, and stand at A, and the lever retracts the stone slab (B) along the rails (C) and into the recess instead of firing the levers. That makes this a combination lock with 256 combinations and a booby trap for those who pick the wrong combination.

So, to the Theory of the trap. There is only one combination of urns that holds exactly the same weight – the 6th and the 4th urns from F and G respectively. The pressure plates have no way of distinguishing brass from iron or honey from wheat, so it doesn’t matter which of these two urns goes on which pedestal. On a very good roll, the rogue will realize that the wheat has long since dried out significantly, and he needs to add a half-filled waterskin on the ‘wheat’ pedestal, if not then he will go one size urn too large, matching the 4th G urn with the 7th wheat urn. Or he could empty the wheat out and replace it with fresh – if he has any. He also assumes (correctly) that the urns have to be placed simultaneously, making this a three-person job. Four, if someone is going to shield the rogue from the darts should he get it wrong; those placing the urns can move toward the rear corners of the room and be out of the line of fire.

Finally, the Application phase, putting the theory to the test: the urns have to be lifted into place and positioned simultaneously, which is easy enough if the rogue gives a countdown. If the party get everything right, B will slide open and reveal another staircase going down into a part of the dungeon they have not explored. If not, poisoned darts will fire from D at A as expected. What won’t be expected is that a pair of trap doors positioned at those back corners will open up under the feet of anyone taking refuge there; after a thirty foot drop and a forty-five foot drop, respectively, this will deposit those falling on different levels of the dungeon. These counter-weighted trap doors will close automatically and re-lock until another attempt is made to throw the lever (H). As things stand, then, the rogue can get this 90% right and still split the party, even if he survives what he considers the primary threat.

I refer to this as a Compound Trap in my notes. It’s a clear and simple term which covers a great number of possibilities – all it really means is that two or more traps are designed to interact in some way.

If the rogue gets a really good application roll, he might get the sense from the resistance of the lever that it’s doing something he wasn’t expecting, but until he figures out what, and finds the trapdoor elements that he missed, he has no hope of disarming the trap.

Of course, this might not be the only entrance; it might simply bypass a High-risk Low-reward sector of the dungeon. That makes it a short-cut that those who belong there know, and intruders don’t – with the intent of making an entire adventuring party vulnerable if they try and get through it.

Difficulties/Targets

You’ll note that I avoided any mention of actual numbers in that example. That’s because I haven’t discussed what those numbers should be as yet – so why muddy the waters?

There are six philosophic approaches to how the difficulty / target numbers for a “disarm traps” that has been broken down into four separate actions (I’m tempted to say, “four separate interactions”) in this way, grouped into two strands of 4 and 2 respectively:

• • Philosophy 1: All DCs at the original
  • Philosophy 2a: All DCs at 5 less than the original
  • Philosophy 2b: All DCs at 5 more than the original
  • Philosophy 3: All DCs at 1/2 the original
  • Philosophy 4: Calculate the Odds
     
  • Philosophy A: All at the same difficulty
  • Philosophy B: Trade-offs, capped

I’m not going to go into these in great detail, but want to make a few notes. The first is that either A or B can combine with any of the 1-4 results – so 1A is a valid combination, and so is 2aB and so on.

Philosophy 1: All DCs at the original

This makes it a lot harder to succeed – but, since DCs are notoriously easy for higher-level characters, this is not necessarily a bad thing.

Refer “How Hard Can It Be?” – Skill Checks under the microscope for more on the phenomenon in question.

If you have to make 4 rolls at 12/- to succeed, the net chance of success is 12/20×12/20×12×20×12/20, or (12/20)^4, or 12^4 / 20^4, all of which equal 12.96%.

If you have to make 4 rolls at 17/- to succeed, the net chance of success is, similarly, 17/20×17/20×17×20×17/20, or (17/20)^4, or 17^4 / 20^4, all of which equal 52.2%.

If you have to make 4 rolls at 12+ to succeed, your chances of success are a little harder to assess. You have a 20-12=8/- chance of succeeding on d20; so the chance of success of all four rolls is a mere (8/20)×(8/20)×(8×20)×(8/20), or (8/20)^4, or 8^4 / 20^4, or a mere 2.56%.

Even if you succeed on anything but a 1, your chances of success on all four rolls is only 81.45% – meaning that almost one time in five, you’ll blow it.

Philosophy 2a: All DCs at 5 less than the original

This represents an attempt to compensate for the diminution of chances at success. It also means that the chances of success are effectively capped in D&D, unable to rise any higher than 14/-, because a 20 automatically succeeds. Any improvement beyond that simply vanishes.

It means that a 12/- chance overall becomes a 17/- on each of the 4 rolls, or a 52.2% chance of overall success. That’s not quite as good as the single-roll 60% but it’s not far off.

How about a low roll, like 2/-? Without the +5, you would have 0.01% chance of success, but with it, the chance of making all four rises to a whole 1.5%.

Philosophy 2b: All DCs at 5 more than the original

Some GMs might note that there are opportunities within the system of getting bonuses through assistance from other characters, and want to “make room” for them. At low character levels, this is probably too extreme, but if you’re dealing with a mid-to-high level campaign, getting +5 or more to different die rolls (especially the first two) is relatively easy. Or you might decide to encourage the cooperation between characters while not letting it get out of hand, which suggests a smaller negative, something like 2 less.

Philosophy 3: All DCs at 1/2 the original

Assessing the impact once again requires a little math.

Let’s start with a character who has a skill of 4, going up against a trap of DC 10. His normal roll required is 6 or better, which means that he has a 15/- chance of success using the standard one-roll.

Under this ruling, the DC for each of the checks becomes 5. The die roll required becomes 1 or better, succeeding 19 times out of 20 – which we already know gives roughly an 80% chance of success.

Same character, DC 15. Which means we have to decide how to handle halves – you can round in the character’s favor, or round against the character, or split the difference, giving two rolls rounded up and two down. I tend to use the first of these, but it’s your choice.

Be that as it may, the character needs an 11 or better to succeed, so he has a 10/- chance of success or 50% – on the one-roll-takes-all original system. If all four rolls were at 10/-, his net chance on four rolls would be just 6.25%, which argues really strongly in favor of SOME sort of adjustment.

Halving the DC means that it becomes 7 or 8. Subtracting the skill of 4 means that the character requires rolls of 3 or 4 respectively. Rounding in the character’s favor chooses the lower of these. Which means that the character succeeds on 17/- for each roll, which we already know means that he has a 52.2% chance of success, overall. Slightly better than the 50%, but close enough.

Same character, DC 20. The character needs a 16 or better to succeed, which is the equivalent of 5/-, or 25% chance under the one-roll method.

Splitting that means halving the DC to 10. Taking off the 4 skill, the character now needs 6 or better, or a 15/- chance of success on any one roll. Result: 31.64% chance of success.

So, for a relatively low skill total, this adjustment works well. Let’s look at the other end of the scale.

Character has a skill of 22. If he’s up against a trap of DC 25, he needs 3 or better on the dice, an 18/- chance, which is the equivalent of 90%.

Halving the DC and rounding in the character’s favor gives DC 12. A roll of -10 succeeds – but you can’t roll less than 1, and 1 is always a failure. So his chance of success is 19/-, and the chance of getting all four rolls is about 80%.

Let’s up the ante to DC 40. Now the character needs 18 or better on a single roll – 15% chance of success. Four rolls at DC 20: the character needs -2. But the lowest result that doesn’t fail is a 2. So that’s a 95% chance of succeeding on any one roll, or 81.45% chance of success.

So the system fails at high skill levels. Plan accordingly.

Philosophy 4: Calculate the Odds

We’ve been doing a lot of odds calculations in these examples – and they aren’t all that hard. If you can work out what the chance of success would be on the one-roll, it’s not all that difficult to tailor a roll to match.

The calculation is the square root of the square root of the chance of overall success as a decimal.

Let’s look at the example that failed the last proposal: Skill 22, DC 40. Roll required is 18 or better on a d20 – meaning a 3 or better chance of success or 15%, which is a decimal value of 0.15. That gives 62.23% chance for each individual roll – call it 60%, which is a 12/-, for three of the rolls. So, 0.6×0.6×0.6×FourthRoll = 0.15; that gives a 4th roll of 0.69444444, or close enough to 70%.

So that’s 12/-, 12/-, 12/-, 14/- – which translates back into 9+, 9+, 9+, 7+.

An alternative might be 9+, 9+, 8+, 8+; that gives 15.21%.

Another way of calculating the correct result is 10^[log(target chance)/4], with both chance and result in the decimal format.

Philosophy A: All at the same difficulty

Some people like to keep things simple. This does that.

Philosophy B: Trade-offs, capped

Others like the notion of being able to reflect the specifics of the challenge by making some parts of the four-step effort more difficult than others.

Let’s take the same Skill 22 DC 40 check that I used above, and which yielded rolls of 9+, 9+, 9+, 7+.

What if the first number was a 4+, meaning that the trap’s elements are quite obvious, but the difficulty in understanding them is rather trickier. 0.15 = 0.85×X×Y×0.60 so X×Y = 0.294 – and the square root of 0.294 is 0.54, which is close enough to 55%, or 10+. So we get 4+, 10+, 10+, 12+.

A simpler answer

Here’s a much simpler answer: instead of a single “Disarm Traps” skill, why not have four of them, one for each stage of the process? When the game system gives an improved chance of disarming traps, you simply have to allocate 4x the increase amongst your four sub-skills.

This permits characters to specialize in different stages of the process, which can make a group of four rogues system-breaking to some extent. But you can avoid that by specifying that the worst of the four skills has to be within 4 out of 20 of the best – so there’s a small band for customization, but the scores are close enough that such trickery has a minimal effect.

This then opens the door to magic (and mundane) items that affect just one of these sub-skills – a manual of traps that gives +1 to the Analysis phase, for example. Once again, these can become game-breaking if you are too liberal with them, so be cautious.

Eliminating Save-or-Die

I really, REALLY hate “Save Or Die” mechanics. I have four solutions that apply to individuals and one that applies to a group situation to suggest as alternatives. Make no mistake, the party will be hurting almost as badly after one of these. To make up the difference, I also offer a fifth suggestion below.

1. Percentage Damage

Option 1: “Lethal” traps are rated for percentage damage done. An 80% chance removes 80% of a character’s hit points. A 95% trap removes 95% of a character’s hit points. Always round in the character’s favor.

2. 1 HD

Rate traps by HD as though they were any other encounter. If it goes off, roll that many dice for the damage total. If the character’s HD are d10s, use d10s; if d4s, use d4s.

Since characters get CON bonuses, whereas the trap does not, their capacity exceeds their level if they have high CON. But this can still be lethal, it just won’t happen as often.

3. 1 HD smaller

This works off the character’s level and the size of their hit die, which means that it presupposes that the increase in hit points per level reflects better ability to dodge and so on. If the character gets a d10 hit die, use a d8 for the damage; if a d8, use a d6; if a d6, use a d4; if a d4, use a d3 (i.e. use d6 and halve the total).

Even if the character has a negative CON modifier, which would be rare amongst PCs, he has a genuine likelihood of surviving such a trap. A more normal character, with a +1 or +2 CON bonus, would probably have about 3HP per level left. We’re still talking massive HP losses.

4. 1 HD left

The simplest and most brutal alternative. Characters hit by the trap have 1 HD + CON Bonus left. If generous, you might maximize that HD, so that a low CON bonus doesn’t risk immediate death.

5. Damage Pooling

Rate the trap in HD as per option 2. If the largest HD in the party is a d10, use a d12; if a d8, use a d10; if a d6, use a d8. Heaven help your PCs if the largest HD in the party is a d4. Kill them immediately.

If the trap gets triggered, roll the HD worth of dice of the indicated size and add the number of PCs as though it were a CON modifier. This is the total damage inflicted on the party. How they split it up is up to them – but the more HP you have, the more you can soak up. Only one rule: you can’t suicide a character by taking a full compliment of HP damage without GM permission.

This only works if the trap’s damage affects multiple characters – EG poisoned gas, rain of arrows, hail of boulders, flooded chamber with sealed exits, etc..

Supplementary Stat Loss

To make up for the lack of immediate death, ding the survivors a permanent -1 to a stat. This is NOT something that can be healed or restored by any means. If generous, let the player roll for which stat is affected. If mean, make it their primary stat (players can choose if their class has several). If really generous, let the player pick – but put a cap on how many times the player can ding the one stat.

A Common thread exists to several of these: what does gaining a HD with a level advancement actually mean? I’ve hinted at this in the text above, but thought I should state it explicitly.

After all, changing the interpretation of what a Hit Point actually represents has massive repercussions on how Healing works, as shown in the All Wounds Are Not Alike series which looked in detail at the phenomenon.

Broadening and Limiting Rogues

This article is wending it’s way toward a conclusion. To close it out, I have some general advice to GMs on how to handle rogues in their campaigns. I’m offering three specific suggestions:

1. Return of the Direct Observation rule-of-thumb
2. Moving from shadow to shadow breaks cover
3. A free HD for Rogues (and mages) at 1st level and Level-as-a-stat AC bonus

1. Return of the Direct Observation rule-of-thumb

I’ve discussed this in the main text. I strongly recommend a return to the “You can’t Hide In Shadows if you are under direct observation when you attempt to do so” rule of yesteryear.

That doesn’t mean that if you’re already hiding, you become apparent to someone glancing your way; once hidden, you stay hidden until you do something to “break cover”.

Which brings me to item number two:

2. Moving from shadow to shadow breaks cover

Again, this is something I’ve discussed in the main body of this article. That said, characters can move and then act, so a rogue can move (breaking cover) and then make a fresh attempt to Hide In Shadows – so long as they have moved from one shadow to another, and don’t fall foul of rule 1 above.

3. A free HD for Rogues (and mages) at 1st level and 10+level-as-a-stat AC bonus to 10th level

Both of these rules reduce a rogue’s combat capability. I want to put that ability back, but in a form more suited to the nature of the character; they aren’t fighters and shouldn’t act like them, even in melee combat on the front lines. I came up with two solutions, neither of them quite enough on their own – and liked one so much that I’m recommending extending it to mages.

Let’s break them down:

A free HD for Rogues and Mages at 1st level:

What the heck, make it a d6. They don’t get any CON bonus or penalty, they just get a small handful of extra HP. Just enough to mitigate the likelihood of getting snuffed at 1st or second level.

10+level-as-a-stat AC bonus for Rogues to 10th level

Giving rogues a bonus to AC makes them slipperier, harder to hit, more adept at dodging. It doesn’t make them any better at hitting an enemy. The question was always how to interpret the general principle?

The solution was to treat Level as though it were a stat and apply the stat modifier as the AC bonus. But a strict interpretation of that means that the “bonus” would be a negative value for the first 10 levels – they have different zero points.

So, a 1st level rogue looks up 11 on the stat table and gets a bonus of +0.
A 2nd-level rogue looks up 12 on the stat table and gets a bonus of +1.
and so on.

A tenth-level rogue looks up 20 and gets a bonus of +5.
An eleventh-level rogue still looks up 20 and gets the same +5 bonus.

As a bonus, this means that a rogue is harder to hit than a fighter of the same DEX, but not excessively so.

If you like this approach, it’s also worth contemplating using 11+Level instead of 10+level. This gives the rogue an immediate +1 bonus to their AC at 1st level, so that the bonus observation is true even at 1st level. It means that the bonuses kick in at odd-numbered levels instead of even-numbered ones, and it means that at 10th level, the bonus maxes out one higher at +6. None of these seem game-breaking to me.

Whew! It’s been a journey – we started talking about Traps and ended up talking Rogues. But that’s because the two subjects are inextricably entwined; and that’s part of the problem with Traps.

That’s why I’m such a big fan of the four-way Skill Split for disarming traps. Making it a process and providing rules for the interaction of other characters with that process makes dealing with traps more engaging, more interesting, and far more interactive. Perhaps not to the same extent as combat, but right up there. That can only be a good thing.

As one more side-benefit, that list of standard actions goes away. But the detailed write-up by the GM is back; so there is a price to pay. But it’s a small one, and worth every penny.

One final question, then: did you find the Bear-trap?