I want to talk today about a word very near and dear to my heart. No, it’s not a swear. It’s an actual gaming word for game designers. Kind of. I know it won’t sound much like a game design word, but it is. It’s something game designers have to know about, just like how chefs have to know about that poop-filled tube that runs along a shrimp’s back and isn’t actually a vein.

Yeah, you don’t actually devein a shrimp, you decrap it. Try not to think about that next time you eat shrimp. Which you shouldn’t do because shrimp are gross.

Anyway…

The word is screwjob and, like a shit-filled shellfish intestine, you really don’t want to find one in your game.

Crap Traps and Broken Weapons

A month or two ago, I mentioned screwjobs in a fight that was raging about traps in my supporter Discord server. Actually, traps and screwjobs often come up in the same conversations, and it’s probably worth having a whole long-ass discussion about designing and adjudicating traps. Especially given how many of you revealed recently that you don’t understand my years-old Click! rule either.

Probably time to revisit all of that shellfish crap.

Meanwhile, also in Discord, this Game Master was trying to come up with a system for weapon breakage to force his players to switch to backup melee weapons once in a while. Like most weapon-breakage systems, it would have ended up a total screwjob and, as is my personal idiolect, I wasn’t afraid to say it. Loudly and repeatedly, and using some of my favorite non-game-design words.

But you probably remember the word screwjob from my recent Feature, Order Matters, which was all about how it’s important to show players things in the right order. Especially given that I told you to remember that particular screwjob as we’d be talking about it again.

I wasn’t lying.

The Screwjob Letters

Imagine you’re playing a fighter. You’re leading your party through an evil temple of evilness. In the temple’s primary vault, you encounter an emaciated, bony fiend brandishing four vicious, serrated weapons in its six arms. Two of them are two-handers, so the math checks out. You’ve never seen anything like it. Maybe it’s a devil; maybe it’s a demon. No one can say because that’s a stupid-ass distinction that needs to get the frigging ax. Without an ‘e’. You know it’s nothing from any of the Monster-o-pedias, though, because you’re one of those disgusting players who reads the books you’re not supposed to read. Obviously, the Game Master is going all homebrew on your ass. Maybe an osyluth and a marilith got freaky in some Tanarrian pleasure pit, and this is the result. Doesn’t matter. What matters is that you’re a fighter and you have a mighty two-handed ax and you’re in charging range. So charge you do and hit you do, but then there’s a crack and discharge of purple-green light, and your double-bitted beauty crumbles to dust in your hands. That’s just how this devil-demon works. Apparently.

So your signature weapon is just gone. The weapon that you picked all your feats and fighting maneuvers for. The one you put all your upgrade points into. The one that makes you the double-bitted ax guy. The one that is literally the reason why you named your character Labryx and paid extra for a helmet with a bull motif. Oh, sure, you’ve got a backup baby ax on your belt, but that just means a smaller damage die and forgoing your Big Honkin’ Weapon Master feat and your Cleavealicious combat style.

You’re just screwed. And you’re pissed off. Not character you; you you. You are pissed off.

“What the hell was that?” you demand of the Game Master. “How the motherloving crap was I supposed to know that would happen?” The Game Master responds with a gurgle, and you realize you’ve somehow got him by the throat, so you let up. Eventually, he explains that, had you taken a different path through the random Gygaxian spaghetti-dungeon that he calls a temple, you’d have found the Libram of Demonology. You’d have discovered a bookmark that would lead you to the demon’s entry and have seen a sticky note reading…

This would be perfect for guarding our vault. Let’s get a summoning circle together. -Cultward Fiendcaller

That, my friends, is what a screwjob looks like. If it doesn’t seem that bad to you because you main spellcasters like a pussy loser, imagine instead that when you cast spells at the demon, they fail and get erased from your spellbook and from your memory, and you can’t use them again until you reconstruct them in your wizard library or some shit like that. See? Screwjob.

Now, let me politely suggest that you read this entire Feature before you leave a comment explaining how it’s totally a fair challenge to ask a melee-focused player to find an alternative to melee attacks once in a while and how there’s nothing wrong with breaking someone’s weapon in modern Dungeons & Dragons and how, back in your day, weapons broke all the times and players sucked it up and modern games are for little baby video gamers masquerading as real gamers. Unless you’re that last guy. If you’re that last guy, by all means, type up a very long comment about your old-school experiences. Just remember that such comments are best typed in an unventilated garage with the door closed and your car running. Go ahead and find your keys; I’ll wait.

The Anatomy of a Screwjob

Unless you’ve got a writhing mass of shrimp veins where your brain should be, you know the above is a total screwjob. But why is it a screwjob? How do you know it’s a screwjob?

You see, there really isn’t anything wrong with occasionally presenting a monster that the players can’t fight the normal way. That’s why gray oozes and rust monsters exist, and it’s why save-or-die effects existed before pussy little baby video gamers took over the hobby. If you’re a fighter, a monster you can’t use your primary weapon against means you need a fallback strategy. Maybe you literally fall back and pelt it with throwing weapons. Maybe you turtle up and just keep the monster’s attention on you while the spellcasters warm up a nuclear option. Maybe you do something really cool, like drop a chandelier on it or wrestle one of its own weapons away to use against it.

So why is the scenario above most definitely a screwjob? More importantly, what can we learn from it that can prevent us from including less obvious, but no less shrimp-veiny screwjobs in our own games? How can you tell you’re setting your players up for a screwjob, and how can you fix it?

It’s actually pretty simple; it’s down to just three questions…

• What the hell could I have done to avoid that?
• How the hell was I supposed to know that?
• I’m still screwed, so who gives a shit?

If a player’s asking any of those three questions and you don’t have a really good answer, you screwed the player. It’s that simple. Except it’s not. Those questions hide a lot more complexity than you might think. So let’s look carefully at each.

“What the Hell Could I Have Done to Avoid That?”

The first question that identifies a screwjob — and the first question to ask of any situation you’re designing that has screwjob potential — is, “What choices could the player make to change the outcome?” Simple, right? I mean, with the Screwjob Demon, it’s obvious that the player could have chosen not to attack with a melee weapon, or he could have chosen to attack with his backup weapon so his signature weapon wasn’t at risk. Fair enough, but we’ll come back to those answers and why they’re shit.

First, though, I want to note that a few things don’t actually count as choices. One thing that never counts as a choice is “the player could have come up with some creative alternative strategy.” Some of you will recall me referring to that shellfish crap as an Open-Ended Creativity Test. While in roleplaying games, it is always possible for players to come up with brilliant, outside-the-box solutions, it isn’t actually a choice to do so. It’s down to luck and a stroke of in-the-moment inspiration. This shit doesn’t test player cleverness or creativity or intellect, especially in tense situations.

Honestly, that’s no different from testing a player’s ability to roll the right number on a die, which is also something that doesn’t count as a choice. Let’s all say it together, “rolling dice isn’t gameplay; choosing what dice to roll isn’t gameplay.” Allowing a saving throw doesn’t make something a choice. Nor does letting the player choose which saving throw to roll. If the player’s choice doesn’t change the odds or the possible outcomes, it isn’t a choice.

Obviously, this is where lots of traps fail the Screwjob Test. Most traps just ask you to roll a die to avoid a bad thing happening because you stepped in the wrong place. What could you, a player, do differently that would actually have an impact on whether the bad happens or not? You could have searched for traps, sure, but even that barely counts as a choice. That’s just another way of rolling to avoid the screwjob. Admittedly, it does change the odds because two rolls are more likely to succeed than one, but it’s still pretty weak as choices go.

Meanwhile, you have to consider the knock-on effect of the choice. Imagine your players are walking down a random hallway in some random dungeon. Suddenly, there’s a click and a blast of flame, and all that’s left of the meatshield up front is a pair of smoking, bull-skin boots and a well-maintained bronze labrys. “What the hell were we supposed to do about that?” the players shriek. “You should have searched for traps,” you say. “It’s a dungeon. Dungeons have traps.”

What’s your game like from now on and forever? It’s weighed down by the players searching every chest, door, room, hall, and graph-paper square for traps. Is that really how you want your game to go? Are the players even making a choice there? No, they’re not. They’re responding automatically to the idea that every square inch of the dungeon is randomly dangerous and throwing dice to make it safe.

Which brings up the issue of whether it’s actually meaningful or reasonable to expect the player to make the choice. Is it reasonable, for example, to ask the party’s primary, front-line fighter to choose between destroying his signature weapon or just not participating in the fight? I ain’t saying that’s what the screwjob demon was doing, per se, but I can imagine a situation where that really is the choice. Imagine, for example, a spellcaster in battle against a foe that’s immune to all magic. Or battling in a room where casting a spell makes your head explode. What can a spellcaster do differently? Pretty much nothing, and when I say that, I don’t mean that there are no other choices, but that the other choice is to literally do nothing. Wading into physical combat is just foolish; for most spellcaster mains, it costs more than it gains, even if they’re carrying a knife or stick. If all they have is attack magic, all they can do is hide and watch. You can say it’s their own fault for not bringing a more varied magical palette, but, again, once the situation arises, it’s just a screwjob.

You see how quickly “what the hell could I have done to avoid that” can spiral?

But that shit about traps and spellcasters brings me very neatly to the next question. Which is exactly how I planned it.

How the Hell Was I Supposed to Know That?

Here’s a fact for you…

It is unreasonable and irrational to expect someone to do something different if they have no reason to think they should do something different. That’s where the screwjob demon totally fails. It’s not that Labryx of Schminoa couldn’t have done something different; it’s that he didn’t know he needed to do something different until he was already screwed.

Of course, the Game Master explained that away, didn’t he? “Somewhere in the dungeon,” he said, “was the information you needed. Had you but taken a completely different random path, you would have known better than to go all Axe Cop on the demon.

Let me ask you something…

Given my stance about random flashes of inspiration and random plastic math rocks, do you think I consider randomly wandering a path through a Gygaxian graph paper maze to count as choices and gameplay?

It ain’t enough that the necessary information to avoid a screwjob exists somewhere in the world. The game must be set up in such a way that a reasonable player would have had the information before ending up in the screwjob situation.

This brings me around to something lots of you struggle with: the Reasonable Player Standard. Some of you hate this shit. Some of you refuse to accept that it’s even a thing. It’s too subjective, you say, and too variable, and too down to judgment calls. Well, suck it up, buttercup, because this is making games for humans for you. The Reasonable Player Standard is what you’ve got.

To apply the Reasonable Player Standard, you just imagine an average player of average skill and average attentiveness and ask what it’s reasonable to expect of that player. Whenever you say, “the players should have known better,” you’re just applying the Reasonable Player Standard.

If your dungeon really is a random, graph paper monstrosity, is there any reason to expect that a reasonable player would choose one path over another? Why? What is it about your maze’s design or presentation that’s leading you to say a reasonable player would have gone left before going right? What should the players have recognized to make a more informed choice?

The answer, by the way, need not always be as obvious as spelling shit out. It need not always be a book. Context clues count here. If the demon’s room was full of armored corpses holding the hilts of disintegrated weapons and you made that an obvious and unmissable part of the flavor text and maybe also the demon was sweating brilliant green acid that fizzed wherever it dropped on the floor, then you’ve got a case when your player says, “How the hell was I supposed to know that would happen?”

Believe it or not, by the way, that’s the right way to think about this shit. Sometimes, your players are going to fuck up. They’ll miss something or they’ll fail to put the clues together. The Reasonable Player Standard isn’t about what your actual players actually do, but what they should have been able to do. So it’s a good idea to think about how you’d defend your dungeon design if your players brought you before the Game Mastering Tribunal on charges of screwjob in the first degree. What’s your case?

In fact, if you’re a decent designer, you’ll see your players dismiss their own cases at the table. You’ll hear them say, “You know what? I should have that coming.” You cannot ask for firmer evidence that you didn’t set your players up for a screwjob than one of the players admitting they shouldn’t have done what they did.

Imagine, for example, that the players somehow knew the room they were entering was the demon cult’s central vault. Imagine they had it on good authority that the demon cult had summoned a powerful guardian to protect their vault. You know the players are going to sit on that threshold arguing about whether to plunge into the room that obviously contains their goal and also obviously contains a powerful boss, or whether they should explore the rest of the dungeon first, in case they discover useful information or powerful tools to help them deal with it. Thus, when the players do plunge in and get screwed, someone’s gonna say, “I told you we should have explored the temple more before we came in here.”

Note that I did exactly that in my module, The Fall of Silverpine Watch. The party knew where the boss monster was and got to choose when to confront it. Only the thickest of thickos tripped over the boss by accident. Everyone else made a conscious choice to start the final fight, and thus everyone who wasn’t ready for it had an answer to, “What the hell could we have done differently?”

You have to be careful, though, about applying the Reasonable Player Standard.

First, remember that, for the most part, you can’t assume the players know anything that you didn’t teach them, tell them, or hint at them. You can’t just claim the players should have realized that a demon cult would summon a powerful demon to guard their main vault and that they should have recognized that the thing behind the locked door was the main vault and that they should have guessed the demon would have some weird ability requiring them to tread carefully and feel out the foe. The Game Mastering Tribunal won’t buy it. They’ll want to know what specific elements in your adventure would have made it possible for a reasonable player to know — or conclude — those things.

“You should have known there’d be a tricky boss because games have boss fights and bosses are tricky,” is bullshit.

Does this shit sound familiar, by the way? Yeah, it sounds a hell of a lot like “you should have searched for traps because dungeons have traps.” You see that, right? Unless you’re completely frigging nuts and actually do want your players searching literally everywhere three times for traps, then the question you’ve got to answer is, “Why would a reasonable player expect to find a trap on this particular thing at this particular place? What’s signaling that?” If you don’t have a good answer, you have a screwjob.

Except…

I hate giving you permission to do bad design, but most traps actually can get away with some screwjobbiness. You shouldn’t lean on that, but it’s true, and I’d be lying if I said it wasn’t. That’s down to the second reason you’ve got to be careful with the whole Reasonable Player Standard thing.

Reasonable players are theoretical constructs. They don’t exist. Your players ain’t reasonable. They’re irrational, emotional idiots. On the one hand, that means that if you don’t piss them off too much, they won’t get their panties in a jimmy when you do hit them with a screwjob. On the other hand, though, it means that, if you piss them off too much, they won’t care whether your screwjob wasn’t a screwjob and they could have avoided it.

When that happens, they ain’t going to care what the Game Mastering Tribunal says. They’re just going to call you a dick.

“I’m Still Screwed, So Who Gives a Shit?”

There are two different ways to talk about making and running games. First, there’s the dispassionate and academic way that I use here on my site. Well, academic, anyway; I’m passionate. You don’t scream, swear, rant, and insult if you ain’t passionate. Dispassionate designer speak involves fancy game design terminology like inertia and abnegation-seeking behavior and screwjob. It’s what’s in your head when you make games to run for your irrational, emotional idiots.

The other kind of talking about making and running games is in the stuff your players scream at you while they flip the table, and what you scream at their asses as they walk out the door.

The fact, plain and simple, is that some things suck badly enough that reasonable players will feel screwed no matter how much they’re responsible for the suck. To put that in dispassionate, objective game designer speak, the more painful the outcome, the higher the standard against which the players will judge the screwjobness of it all.

When you get down to it, traps ain’t a big deal. Yes, they’re mostly down to damage you get to roll to not take after stepping in the wrong innocuous place there was no reason to be afraid of, but they also mostly just spring up, slap you, and go away forever. Big picture-wise, unless a trap takes a character’s last hit point, most players just shrug and get on with their lives. Similarly, if you discover a monster has an aura of damage by blundering into it and taking the hit, it’s no big deal. You take the hit, you learn your lesson, and you adjust next round.

It’s okay for a player to learn the stove is hot by touching it, as long as it doesn’t burn off their whole hand. They’re technically screwjobs, but they’re ones you can get away with.

But once the price of being screwed goes beyond a bit of damage, a manageable penalty, or the loss of a fairly cheap piece of unimportant gear, then players’ natural, irrational, totally human loss aversion kicks in. And it kicks in hard. It can even make a totally fair non-screwjob feel like a screwjob.

Some of this shit’s really obvious. If a character dies, the players are going to hold the whole situation to an extremely high screwjob standard. In fact, many players will judge anything that removes a character from play for more than a single round against a pretty high screwjob standard. This is why, instead of teaching Game Masters how to properly avoid screwjobs, the designers just removed save-or-die effects from the game. The assholes.

Players hate not playing the game, and they hate losing their characters permanently. Actually, they hate losing anything permanently, but they especially hate losing anything they’ve earned through play, like experience and levels, for example; anything they’ve invested resources into, like weapons they’ve put feats or character abilities into mastering, spells they’ve researched, or items they’ve crafted or upgraded; and they hate losing anything they feel is a substantial part of their character’s identity.

Why the hell is it so hard to get Game Masters to grasp this shit?

I ended up in this fight in the Discord server about why taking a melee character’s signature weapon away in modern Dungeons & Dragons was very likely to hit as a screwjob even if it technically, objectively wasn’t one. It ain’t just a weapon being lost, it’s a chunk of the character’s identity and investment combined with a nasty choice between playing at your weakest or returning to town to replace the weapon. Which, by the way, is another thing players hate. They hate failing adventures, they hate losing encounters, and they hate setbacks, major and minor. People hate failing as much as they hate losing.

That’s why Labryx the Ax Guy’s player got so pissed off.

By the way, I’m totally aware that it was different back in 1974, and players didn’t take weapon loss as seriously. That’s because everything was screwjobs back in 1974, and enjoying combat was playing wrong, and players didn’t know that roleplaying games could actually be good, so fuck off with that.

In the introduction, I mentioned a discussion about weapon degradation and damage systems. Someone in my Discord server proposed a system for such because they wanted melee players to sometimes have to switch weapons, which is a motive I still don’t understand, even though people keep spouting it. Why is that so fucking important to so many people? What does it get you?

Anyway…

The system was based on fumbles. Roll a fumble, your weapon is now at risk. Fumble again with that same weapon before you get it repaired, and it breaks. I first offered a statistical analysis of how surprisingly high the odds of rolling at least two fumbles in a given adventuring day actually are based on the average number of attacks a melee character rolls in D&D. I then pointed out that there’s nothing players can do to avoid weapon degradation because it’s just a random crap shoot and after the players’ weapon breaks, they’re just going to switch to their secondary weapon and hate it, which isn’t actually a strategic choice either. Then, I explained that, in modern D&D, melee main players don’t think of losing their signature weapon as a minor inconvenience, but as a screwjob.

That last point was the one everyone fought me on. Incorrectly. Game Masters just don’t get this shit. They need to get this shit. You need to get this shit.

A Good Screwjob is Expensive

It’s important to avoid screwjobs because fairness and good game design and proper challenges and blah, blah, blah, objective game design bullshit…

Screwjobs ruin games. They destroy player trust. When that happens, the absolute best case scenario is that the players start playing in really unfun ways. Every Game Master knows what happens when one too many NPCs betray the heroes; the players stop trusting anyone ever. They don’t even treat the game’s characters as people anymore. It’s exactly the same when the players start to think every flagstone in every dungeon might hide a trap. They won’t walk down a hall unless accompanied by the clatter of search-check dice. They ain’t thinking, they ain’t engaging, they’re just behaving like wounded animals. They’re afraid of everything.

Me? I’d much rather hear my players say, “We should be careful here because this whole setup smells like a trap.” That’s thinking. That’s engaging. That’s actually playing the fucking game.

I don’t want to piss and moan about Hollow Knight: Silksong here, but it’s a perfect example of ruining player trust. Several of the game’s benches — those are the game’s safe recovery respawn checkpoints — several of the game’s benches are screwjob traps. You use it, it hurts you. Maybe it kills you. Some can be made safe, while others just screw you over and disappear forever.

Miyazaki — the Dark Souls one, not the one who made Akira and Avatar the Airbender — Miyazaki was asked once why he never put a trapped bonfire checkpoint in Dark Souls. He said, “If I did, the players would never trust the bonfires again.” In Dark Souls, as in Silksong, the checkpoints are your sanctuaries, your earned victories, they’re where you lock in your progress, and they release some of the tension you build up playing through the game’s challenges. They’re too important to the gamefeel to turn them into screwjobs, and once is too much.

This is why the last screwjob factor I mentioned is so damned important. If a player ever says, “I’m still screwed, so who gives a shit,” that’s a serious problem. In reality, there isn’t a Game Mastering Tribunal to exonerate you if you’re falsely accused of screwjobbing in the first degree. The objective game design reality is just spherical chickens in a vacuum. The only thing that’s real is your players’ relationship with you and with the game you’ve asked them to play.

Unfortunately, while facts may not care about your feelings, it’s the feelings that will determine whether your players come back next week or key killer GM into your car when they leave forever.