The History of Traps In Dungeons & Dragons

In original Dungeons & Dragons, the three brown books only include one rule for traps. “Traps are usually sprung by a roll of a 1 or a 2 when any character passes over or by them.” That’s it. Except for the Find Traps spell, the rules never explain how characters can find traps. In D&D’s original play style, if you wanted to find pit traps, you just told your dungeon master how you pushed down on the floor ahead with your 10’ pole. Or you sent your hireling ahead first. See A Lack of Ability Checks Shaped How People Originally Played Dungeons & Dragons.

Book 3 The Underworld and Wilderness Adventures

Book 3 The Underworld and Wilderness Adventures

In Book III, Underworld and Wilderness Adventures, Gary Gygax lists a dozen tricks and traps such as slanting passages, sinking rooms, and one-way doors. All foiled a retreat from the dungeon. They threatened to make characters lose their way out, or worse, deliver them to a deeper level and more more dangerous foes. None of these traps need rules to play, just player ingenuity.

Undoubtedly, Gary had thought of other traps such as spring blades, poison needles, and warning bells, but his list conspicuously omits any traps that seem to require game-world finesse to overcome.

In the May 1974 issue of a fanzine called the Great Plains Games Players Newsletter, Gary Gygax presented the thief class. In his introduction, Gary tells how the class was suggested by a gamer named Gary Schweitzer (probably Santa Monica gamer Gary Switzer). “He mentioned that his group was developing a new class of character—thieves. Gary gave me a few details of how they were considering this character type, and from from these I have constructed tentative rules for the class.” In 1975, Supplement I: Greyhawk made the class official.

The thief class featured the ability to “remove small trap devices (such as poisoned needles)” At level 1, the thief boasts a 10% chance! So when your new thief says, “Don’t worry, I’ve got this,” the party should dive for cover. Original thieves enjoy no special ability to detect traps. Keep your hirelings in front.

The thief’s limit to disabling “small trap devices” seems to exist as an attempt to confine thieves to working on traps that require a character’s game-world knowledge and dexterity. For example, a chest rigged to release deadly gas requires a thief’s game-world aptitude, and a die roll. Big traps like pits and rolling boulders, which can be beaten through player ingenuity, remain outside of the thief’s skills. Players can tell the DM the steps their characters take to bridge a pit or to chock the rolling-boulder trap.

In the summer of 1975, Gary  brought the Tomb of Horrors to the Origins convention for a D&D tournament. One of the tournament’s players wrote a first-hand account of the event for issue 4 of the Alarums & Excursions fanzine. Even though the party includes two members of the new thief class, the Tomb offers virtually no place for them to disarm traps, and the Tomb [SPOILERS!] is loaded with traps. To determine when players get caught by traps, Gary fills the adventure with an ad-hoc system of saving throws, rolls of 1-2 on a d6, and verbal countdowns. (Player tip: If the DM begins to count down, run!) The Tomb’s legendary status comes from the mix of ingenuity, divination, and attrition required to bypass its memorable deathtraps, rather than the number of disarm checks needed. (DM Tip: if you run the Tomb and allow thieves to detect or disarm much, you’re doing it wrong. The Tomb of Tiresome Checks is a different adventure.) See Tomb of Horrors tests patience, but still ranks as Dungeons & Dragons’ best villain.

In Advanced Dungeons & Dragons, thieves finally gained the ability to locate traps. A low-level rogue’s odds remained dismal, quickly upstaged when the priest gains Find Traps at level 3. This thief ability implied that no one else could find traps—after all, other classes lacked a Find Traps percentage. Third edition set this limitation in the rules by allowing rogues (and only rogues) to find traps “well hidden” behind a 20 or higher Search difficulty.

The rogue or thief’s limit to finding and disarming small traps remained in second edition. “These include poisoned needles, spring blades, deadly gasses, and warning bells,” but do not include “large, mechanical traps.”

In third edition, traps gained a systematic treatment, complete with triggers, effects, and difficulty classes. The Trapfinding ability enabled rogues the chance to locate and disable anything that the DM categorizes as a trap, small or large, magical or mundane. This gave rogues more chances to shine, but heightened the tension between the traps a thief can find and disable and the traps that test player ingenuity. We have all encountered players who insist that a disable trap roll will enable their rogue to easily bypass some elaborate and cunning challenge. So does staying at home, but neither tactic leads to much fun.

When the fourth edition designers rethought D&D, they saw traps as posing two core problems:

  • Traps can frustrate players
  • Traps can slow play to tedium

Problem: Traps that challenge player ingenuity can lead to player frustration.

This problem arises when when dungeon masters limit the players to a preconceived menu of potential solutions. This approach riddles the Tomb of Horrors, which includes many predicaments that require curiously-specific recipes of spells or actions to escape.

In Traps!, fourth-edition designer Stephen Radney-MacFarland writes, “In the early days, DMs all too often felt compelled to demonstrate their cleverness and punish players for making ‘wrong’ choices—even a choice as simple and random as which passage to explore.” For example, Tomb of Horrors. See Player skill without player frustration.

Problem: Traps can slow play to tedium.

Regarding the problem of slow play, Stephen Radney-MacFarland writes, “The ‘right’ way to play the game was to slowly and laboriously search each 10-foot square of dungeon before you set foot on it, or to use magic that made traps completely pointless. Neither option was much fun.”

Radney-MacFarland never mentions that old-school traps require wandering monsters or some other time pressure to avoid grinding the game to a halt. Of course, if time pressure denied characters the chance to look for the trap that killed them, the hazard seems arbitrary and unfair. See Three unexpected ways wandering monsters improve D&D play.

Fourth edition gives traps a new design

Radney-MacFarland admits designers thought about “disappearing” traps from the game, but decided to try fixing them first.

The fourth-edition design sought to fix the problem of frustrated players by eliminating traps that only challenge player ingenuity. “We wanted to expand the ways in which you could counter a trap. Much like figuring out that sometimes you wanted other skills to allow a character to recognize a trap’s threat, we made an effort to design traps that could be countered with an interesting skill uses.” Skill checks became the core mechanic for resolving traps. The game invited dungeon masters to allow as many different skills as plausible so everyone could share the fun of making skill checks.

Many players prefer traps that require ingenuity to overcome, because such challenges make the players’ decisions matter in the game world. But not all players favor this play style. The fourth edition design aimed to please players who insisted that a disable trap roll enabled their rogue to easily bypass some elaborate and cunning challenge. Still, the designers recognized that turning traps into a cause for skill checks failed to offer enough fun, so they redesign went farther.

“Most traps work best when they ‘replace’ a monster in a combat encounter, or serve as a hazard equally threatening to both sides.” In fourth edition, traps become a sort of stationary monster that the characters can disable or attack. Like monsters, traps make attacks, grant experience, and have solo and elite varieties. In this new concept, traps add spice to combat encounters, allow rogues to strut their skills, and target monsters as well as players—a new tactical element.

Radney-MacFarland writes, “Don’t fret, rogue fans. That class and other characters trained in Thievery are still the party’s best hope to shut down traps quickly and well.” But fourth-edition rogues soon learned to approach traps like everyone else, by attacking. Fourth-edition rogues inflict so much damage that a series of thievery checks always took longer than just attacking a battlefield trap.

Justifying battlefield traps

In the game world, the battlefield trap always seemed hard to justify. I pity dungeon builders stupid enough to bother enchanting, say, an automatic-crossbow trap rather than an iron defender or other construct. Unlike constructs, traps (a) cannot move, (b) can be disabled, and (c) will attack your guards as well as intruders. The dungeon builder’s henchmen, hired to fight alongside their master’s indiscriminate death machines, should look for a job at a better class of dungeon.

The fourth-edition approach to traps never proved as satisfying as hoped. As the edition evolved, we saw a gradual return to classic traps, even with all their problems.

Fifth-edition traps

Although complex traps revisit the good ideas from fourth edition’s battlefield traps, most fifth-edition traps recall the ones from before fourth edition. The rules offer advice for avoiding the problems with traps. “Traps are most effective when their presence comes as a surprise, not when they appear so often that the characters spend all their effort watching out for the next one.”

Just like thieves in D&D’s original game, fifth-edition rogues lack any special ability to find traps. Now, to find a trap, any character can attempt a Wisdom (Perception) check. The rules specifically allow players to find traps by looking in the right places. “You should allow a character to discover a trap without making an ability check if an action would clearly reveal the trap’s presence. For example, if a character lifts a rug that conceals a pressure plate, the character has found the trigger and no check is required.”

Depending on the trap, the best way to disarm may be a Dexterity or Strength check, but player ingenuity often works. “As with many situations, you shouldn’t allow die rolling to override clever play and good planning.” If disarming a device requires a check, the rogue’s proficiency with thieves tools can help.

Ironically, rogues rarely have high Wisdom, so they rate as bad at finding traps. Lucky for today’s rogues, the class pivoted from unlock-and-disarm specialists to hidden snipers. See The Thief’s Strange Trip From Non-Combatant to Battlefield Domination.

12 thoughts on “The History of Traps In Dungeons & Dragons

  1. Ty

    From a game play standpoint, traps are just a terrible idea all around. Conceptually, in order for a trap to be a “good” trap, it needs to be massively unfair. It needs to kill outright or seriously maim. One minute you’re alive, and then boom!, you’re dead. No saving throws, no “noticing something off at the last minute”, no “jumping out of the way”, …nothing. An IED is most effective when suddenly your legs are just gone. A bouncing betty is most effective when suddenly your legs are gone. Punji stakes dipped in poo probably won’t kill you outright, but you’re not fighting fit anymore when you step on one. Plus, doctors probably need to amputate. Legs gone. “Good” traps aren’t sporting. “Good” traps don’t let you roll a saving throw. “Good” traps are savage. That doesn’t make for a very fun game, in my opinion.
    If, from a game play standpoint, you swing the other way and take the edge off your traps, you run the risk of the game feeling like wacky shenanigans. Your game becomes a Home Alone rip-off. Kobolds might be laughably inept, but any serious villains are going to look pretty stupid defending their base with something silly like “stinkpots.” Why hit them with stinkpots when you could have used napalm AIDS-inducing super acid.
    As a DM, I’ve never been able to find a happy balance when it comes to traps, so I generally avoid them altogether because of that.

    1. Ken W

      You need to take the edge off your realism. A trap shouldn’t be “instantly lethal” in game terms anymore than a strike with a sword or greataxe is. In real terms, if you get “hit” by a swinging claymore, you are likely suffering a severe wound. But the abstraction of D&D combat and hit points means that each “hit” represents a depletion of stamina, not a mortal wound. Only when you reach 0 hit points does it really represent that fountaining arterial spray we would otherwise expect.

      Traps operate in the same space as combat weapons in this regard. The only difference between a trap and an enemy combatant that gets a turn while the PC is surprised is… well – nothing. Except the trap essentially “dies” after its turn is over.

      1. Beoric

        Traps can be instantly lethal, or at least very dangerous, but they should be easy to find. The interesting part is the choices you make in order to deal with the trap.

        Easy to find means (a) in plain sight, like an open pit; (b) in an expected place like a treasure chest; (c) telegraphed, like by scorch marks on the ground; or (d) automatically discoverable by a change in procedures, like always probing ahead with a 10 foot pole.

        Knowing the trap is there is not the same as knowing its mechanism or how to circumvent it, so a found trap (or a suspected trap if you flubbed your find traps roll and haven’t been able to rule it out through testing) is a puzzle – and one of the options is always to walk away and find another route.

        The 10 foot pole is worth a special mention, because the procedure slows down movement (but not play, assuming you don’t roleplay every instance of poking the floor), so the choice is risking traps or risking wandering monster checks – and it makes running away through an unmapped area very risky.

        1. Ty

          There’s one problem with that through- a “good” trap shouldn’t be easy to find. That’s what makes it an effective trap. You don’t see it coming. This is barring “area denial” traps, like minefields, of course, but those are serving as more of a deterrent. If the trap-setter is truly that smart/competent, and really doesn’t want the PCs to find their “treasure” or whatever, they’re going to place traps in ways that don’t give intruders a fighting chance.

          I’d like to point out that mechanically, you described a great way to implement traps if you’re planning on including them in your game. I used to use them a lot more and that’s basically what I did, but I could never kick my own perception that traps were a zero sum game.

          1. alphastream

            Not necessarily. A trap can be a lot of found when found, if it requires engagement to disarm. As a DM or author I try to think through the point of the trap not just whatever creatures put it there, but for the game experience. It can be hard to find and that’s fun, or it can be easy to find and be fun as well. Think of the “only the penitent man shall pass” in Last Crusade; that’s fun because you know it is there and need to figure out a way past it. Similarly, traps can be found and that can be the beginning of the engagement.

          2. Beoric

            True, from the in-game trapsetter’s perspective a trap should be hard to find. But if you need to justify it (I don’t), consider that in a standard vanilla pseudo-feudal setting there are technological limitations. Parts for non-magical traps are made by smiths, not machinists. Ever looked at a medieval lock?

            Perfectly good traps can be suspected because the nature of the trap is not entirely concealable. Raiders of the Lost Ark style traps can be suspected because the tiles on the floor have no grout because they are pressure plates, or there are holes in the wall from which darts shoot.

            The trap may also be old, and detectable by signs of wear, like a layer of powdered stone on the floor or vertical gouges on the wall for a falling block trap, or soot on the walls or floor with a fire trap, or spent missiles on the floor with a dart or arrow trap.

            Also consider that some traps can be very well concealed if they are not being looked for, but still be detectible if actively searched for. A standard old-school pit trap was pretty much undetectable visually and could only be detected by tapping it. But tapping an apparently stone floor that is actually made of paper mache is a dead givaway. The modern equivalent is sweeping for mines with a metal detector.

            Traps on locks and chests are also virtually undetectable, but you presume their existence because of the nature of the devices.

            None of those are actually bad traps. They just have limitations because of their nature.

            There is a great discussion of this at the Hack and Slash Trick and Trap Index:

  2. alphastream

    Very fun read. A few thoughts, beginning with how in 5E, it’s still not entirely clear nor standard at tables whether Investigation or Perception is most commonly used for finding a trap. I have my thoughts, which I think are right, but I see it run many different ways. (In general, I think that if a trap is one that could be seen with the naked eye, then Perception would work. For example, a pressure plate that has slightly discolored stone, or which is slightly sunken. Otherwise, and in my game this is most of the time, the trap is not obvious and needs Investigation to be found. A well-crafted pressure plate is like any other stone. The only way to find it is to tap at it or otherwise determine what it is, which uses Investigation.

    I like the general tone of what you are saying, though I think we see examples of making checks to find traps across many adventures and many editions before 4E. In older adventures they may be ability checks, but it was certainly very common in 3E adventures to have adventure text prescribe a skill check to find or bypass a trap regardless of size or type. We saw this a lot in organized play. BUT, it varied greatly. One DM or one adventure might allow a tracking check to find a trap (because the denizens move so as to avoid it), while another might not allow any check at all. 4E really began to codify that.

    4E’s concept of “trap as monster” failed due to the underlying math, which assumed a check per round and 4 checks to disable the trap, which was supposed to equate how monsters were envisioned as taking 4 rounds to defeat. The problem is that this cold math doesn’t understand how that 4 round concept wasn’t very accurate – players focused fire on important targets and might take them down in 1 round, while ignoring others. The rules for traps (often requiring a character to move to the trap the first round) and the lack of clarity for players (the rogue might get there to find that Religion is the best check to defeat it) meant characters made traps the last priority. And, by virtue of ignoring most traps, they killed the other monsters faster and then a DM might hand-waive the trap.

    I like to think the 4E concept is still really cool, but it takes clever authoring to communicate to the players how to engage with it. It is awesome if the cleric immediately realizes that this trap is empowered by a rival deity and they can shut it down and greatly help the party by doing so. That feels really heroic. It’s awesome if the rogue can tell the party that interacting with the trap for two rounds will move the rays of lightning to the area where the enemy archers are standing. These are great cinematic concepts if you set them up right.

    5E DMs should still consider these “complex traps” (and the best ways to run them), as well as the more open style of traps and the simple traps. Tomb of Annihilation really showcases how open trap design can work well, such as in the House of the Crocodile. Players and DM should have a lot of laughs with that design, and it is worth copying. I tried my own hand at it with Dungeon of Doom. Nate and I designed a large variety of 5E traps in that adventure, and they provide a diversity of experiences (you can get the adventure free and also see people play through them, all at Thank you for putting up with the shameless plug, but it’s hopefully useful for people given this article.

  3. Vividantivirus

    Just wanted to chime in that I think your historical summary of traps is excellent and your assessments are accurate. It further illustrates that there still isn’t a clear-cut solution in the 5th Edition’s rules set that implements traps, both detecting and disarming, in a satisfactory manner. The current design, while looking similar to the earlier version of the game, has inherited their flaws to one degree or another. It might require a significant re-thinking of them to bring traps in line with other mechanics that, while simplified, make that portion of the game fun.

    The good news is there’s portion we know that works: how much damage they inflict, or what the results of the trap are when sprung — basically, the game mechanics portion. The detection, or at times, “auto-detection” via Passive Perception, of the trigger seems to have some faults. Also, allowing for flexibility on the manner of disarming traps, either utilizing “player ingenuity” or die rolls, makes things inherently murky. We don’t see that with Combat mechanics; there’s no way for players to narrate their way through combat. And as you mentioned, democratizing finding and disarming traps both diminishes the role of a Rogue while at the same time doesn’t allow them to shine. Why don’t we see the same treatment with Clerics turning undead? Probably because it’s a bad idea.

    I disagree with the conclusion of why 4th Edition’s traps failed. From my experience, it wasn’t the underlying math that caused the problem. It was the disparity on how gamers thought traps should work. The more elaborate they became, the less clear it became for players to defeat them. They were just too much to figure out. Nothing’s worse than a player not being able to know what they are supposed to do while something is killing their character. It became guesswork on the players’ end and a lead-by-the-nose “this is what you need to do” on the DM’s end. The solutions were often not self-evident. They were too clever for their own good.

    I do agree with Alphastream that the “traps as a monster” idea is excellent. 5th Edition’s complex traps are worthy a spiritual successor, but there’s still the niggling questions: are detecting traps better handled through investigative dialogue between players and DM player? Should disarming traps be handled through player ingenuity or should they be resolved through die rolls? If player ingenuity, then have the rules built a large enough “vocabulary” with players on how to foil the various trap mechanics? I suspect not.

    It seems that unless the necessary steps or actions to disarm each type of trap are spelled out, the only reasonable solution that countermeasures can provide is to require a die roll. But I’m getting into the subject of how to fix traps, and not the retelling of how they’ve been handled in the past.

  4. Randall Baumgardt

    I have found that different players handle traps well, differently. Some like the role play challenge of overcoming them through investigation and player intuition. Other groups like relying on the characters mechanical builds. Some groups like them others hate them. It is up to the GM to discover how the table wants to deal with traps or anything else for that matter and adjust for that style.

  5. Bruno Kopte

    Excuse me, mr. Hartlage. I’m a translator for the brazilian website rederpg ( ). We are interested in translating and posting your article there, if you allow us. Of course, you’ll be credited and a link to the original article will be included. Thank you for your attention and time.

    1. David Hartlage Post author

      Hi Bruno,
      Yes, translate and post the article. Thanks for your interest.



Leave a Reply