I've been GMing for a couple of sessions now and I'm still struggling at making encounters or situations more difficult. I will describe dangerous situations and encounters in fiction but the mechanics make it too easy for my players to succeed.
For instance no matter how big I describe the Troll, if the fighter of the party charges him and swing his sword, roll+2 for Hack and slash, he'll easily hit the troll. Almost 40% chance to hit without consequences.
One thing you can do to make a big boss dangerous is just make normal Hack & Slash useless. Imagine they are fighting a storm giant or something the like, and they normal Hack & Slash, you can just say "ok, you are just chipping his toenails, that is not going to work."
Force them to be creative: climb the giant (defy danger), try to out maneuver it, use terrain to get higher, make it kneel, whatever. It's a giant, it makes sense a normal attack on his foot with a sword is not going to work (even if it works in other games).
About hard moves, I would say the troll hitting is a consequence of them failing. For example you say:
— The troll comes at you with its huge club, what do you do?
— I tumble to the side to avoid it. (player rolls and fails)
And now is when you decide a hard or a soft move. You could do damage, or you could just put the player in a harder situation (you are not fast enough and the club grazes you taking you out of balance, your weapon goes flying and you are now lying on the floor, while the troll prepares to step on you!).
The narrative at hand is a perfect reason to make things difficult for the player. You should always ask questions like crazy, ie. ask for justification from the player about how he does what he does. He may want to hack and slash at the troll but he doesn't decide on what move his intended action corresponds to. Feel free to challenge his intent by clarifying details and then asking.
I can imagine a conversation going like:
— I hack and slash at the troll!
— Cool, how do you do it?
— I just swing my sword to chop his head off!
— Well, he's swinging a huge tree trunk around, how do you intend to get close enough to bring your steel toohpick in striking range?
— Well, I try to look for an opening and dive in.
— Looks like discern realities, please roll+wis.
— That's an 8, I get to ask one question, ok, what should I be on the lookout for?
— His swing has a semi-random looking pattern but you can guess it out.
— So I can just run in with perfect timing to avoid his swing.
— Cool, that looks like defy danger with roll+int to outsmart the troll.
— What? Not dex? Why?
— Because you can see that even with speed and reflexes, you'd eventually get hit by the trunk unless you figure out his rythym.
— Well, here goes… a 4, ouch!
— There's no way you're getting in by timing it, sooner or later that trunk will push your skull into your ribcage unless you keep your distance. That's what you have been doing instinctively anyway without realizing that you have been cornered, with nowhere to retreat anymore. The troll seizes the opportunity to deliver its skull-to-ribcage promise by swinging the trunk straight down on you. What do you do?
— Whoa, I duck and tumble inside his guard.
— Sure, defy danger with roll+dex…
— 9 this time. What now?
— Well, you're obviously not going to pull that off with a sword in your hand. Either let go of it or feel the impact of the trunk on your back, your call.
— I need the HP, so ok, I let go of the sword and tumble.
— Cool. You're now inside the troll's guard without your sword. The troll grins to show you his rotting teeth, drops the trunk which just falls on your sword, and lunges to grab you by your throat. What do you do?
Here, the GM diverts the player's action so that he's on the defensive rather than the offensive he intends. The player had to go through three potential points of failure and he still couldn't take a real swing at the troll.
That, in my opinion, is what makes this a difficult adversary. Not how improbable your chance of succeeding is, but how many other things you have to go through before you can take a shot at your quarry.
Dungeon World puts difficult after the roll, not before.
There are two misunderstandings combining when someone wants to start fiddling with the mechanics to make Dungeon World harder (or easier).
These two assumptions that are true in many other games are false in Dungeon World:
In another game, you might do this:
DM: "Jumping the chasm is a Hard check. Roll to see if you succeed."
- Success: "You leap the chasm and land safely on the other side."
- Failure: "You fall!" (rolling) "You suffer 23 damage from the fall. You're lying in a shallow stream at the bottom of the chasm."
This is not how Dungeon World works.
In Dungeon World, the corresponding truths are:
In Dungeon World, what happens is more like this:
DM: "Jumping the chasm is dangerous, how are you going to defy that danger?"
Paladin: "I'm going to take a running start and put my all my strength into it."
DM: "That sounds like Defying Danger by Powering Through, roll plus strength."
- Hit: "You leap the chasm and land safely on the other side."
(no GM move)- Partial hit: "You go to leap the chasm and then pull up short. You can feel that you're just not going to make it with your panoply of plate. You can remove your amour and leave it on this side and make the leap no problem. What do you do?"
(executing the 7–9 result of Defy Danger: the PC hesitates and the GM offers an ugly choice)- Miss: "You leap the chasm, but only just. You slam into the far edge and barely manage to grab on to stop yourself from plummeting. Ranger, you can tell that Paladin is stuck, unable to pull herself up due to the weight of her plate armour. She's slipping, what do you do?"
(GM move: put someone in a spot, along with the advice to move the spotlight around)
Let's unpack that.
First, there is no success or failure, just hits and misses. Both hits and misses need to be translated through the rules and some more choices first, before you can judge subjectively whether the outcome was a success or a failure.
Hits of either kind always become fundamentally successes, but only after you follow the move's rules, and we don't know about misses until the GM makes their move. The success of a partial hit can carry consequences that can hurt enough that it's still a success, but a Pyrrhic one. We always follow the rules of the moves to find out what happens exactly first, we don't just say "yay, you did what you were trying to."
For misses, we need another step of transmutation before we learn about success or failure: the GM move. Most GM moves are some kind of consequence that is tangential to what the PC was trying for: losing things, getting split up, being caught in a bad spot, noticing a temptation, etc. None of those moves say "you fail", so when the GM makes those moves, she can say that you succeeded at your main intent anyway. The only time the GM has to say you failed at your intent is when she picks a GM move that has in-game effects incompatible with whatever the PC was attempting to achieve. If the GM is making a soft move, succeeding at the original task is one of the ways of making a GM move softer. If the GM is making a hard move, then failure is likely part of the package. Success and failure, then, depend on what the in-game effect of the GM's move is. The miss isn't the failure, it's the consequences of the move the GM chooses.
Second, mechanical difficulty isn't determined before the roll. Instead, the situation has already been set up using soft (or sometimes hard) GM moves in response to the players declared what they're doing, saying, and looking at, and whatever else is happening at the same time. That sets the in-game difficulty of the situation, but the difficulty only gets applied to the task at hand after the roll, and only on a miss or if the move's rules says the GM gets to add something unfortunate. On a success, the difficulty simply didn't matter.
Defy Danger seems like an exception to this because it often happens before other moves, but it's not. The GM can't just call for Defy Danger on a whim, it has to be because the PC is doing something that puts them in danger—so it triggers independently of the GM's wishes, based on the fair judgement of the situation instead, just like every other move. It's the choice to do something that exposes them to danger that increases the difficulty, not the GM deciding it should be harder and assigning a mechanical modifier.
When a player declares their PC is attempting something that matches a move without exposing themself to danger first, the group just does what the move says. There is no difficulty involved at first. The roll is made, and a hit means they did what they were trying to do, easy as pie. Only when the roll gives the GM an opportunity to make a move, or to say something as part of the move's instructions, does the GM bring difficulty into the picture after the fact. If the roll "should" have been harder in some other system, now is the time to show just how hard it was by using a hard move to show the hard consequences of the risks taken. If the roll "should" have been easy, now is the time to show that it was easy by making a soft, kind move that might even include success, despite the miss.
A quick example might help demonstrate the difference: Imagine the Fighter is Hack & Slashing against an orc, and misses. The GM chooses a move that follows (a GM principle), choosing to have the Fighter slip and go to one knee as the orc grins and winds up for a strong overhand chop (soft move: show signs). Now imagine the Fighter is Hack & Slashing against a roper (a toothy tentacle monster), and misses. The GM chooses a move that follows, choosing to have the Fighter grabbed and pinned by three wrapping tentacles (hard move: put them in a spot). That's much worse! Fighting the roper is harder than fighting an orc, but it's harder because it's more dangerous when the Fighter messes up, not because of plusses or minuses.
This actually goes for hits too: even when you roll a hit, what you were trying to do... maybe that actually makes your life harder, and you did that all by yourself without the GM's help. Again, it's the situation caused by the result of the chain of moves that makes things harder or easier after the roll. That's not something that could be modelled by a bonus or penalty to the roll anyway, but it neatly flows from the non-bonus-based difficulty rules of Dungeon World.
You could make things mechanically more difficult by making custom moves, but I would not start there. Instead, narrate in the fiction greater consequences, via the GM's show an approaching/looming threat moves and follow up with hard moves, tougher monsters, and more damaging challenges.
For a boss with giant sword, lead up to the fact that he's dangerous: share that he's killed twenty knights in one-on-one combat, show carnage on the way to him, have him be encrusted with gore. The 16 HP Dragon is an example of what can be done. You could also make him mechanically tougher, of course.
I wouldn't make the lockpicking roll harder, but I would just ramp up the consequences of failure—again, appropriately telegraphed so that when they lose a finger or hand when they botch (6-) the roll they aren't completely surprised.
I'm also having a hard time saying to my players as part of a hard GM move: You are hit by a the troll holding a tree as a club and you fly across the room and land on the wall. You take 4 damage that ignores your armor. I can already imagine my players asking:
- Can't I try to dodge the attack?
- Why do I get damage automatically and why does it pass my armor?
Explain when you, as GM, are allowed to make hard moves like this. If they ask why their armor doesn't help, ask back: Do you think it's going to help? You got hit by a tree and landed on a wall. Do you think it would help you if you fell down the stairs? Remember the GM principle "Ask questions and use the answers", but also talk to your players so that they understand how the game works.
The simple answer is use hard moves.
When you have a chance to make a move, you get to choose whether it is a soft move, announcing danger; or a hard move, delivering on the promise of that danger.
So if you think fights are too easy, use more hard moves.
My experience is that players like it when they succeed, but bleed. That means they get through just by the skin of their teeth - they face dangers that do them harm and make tough choices and live with the consequences - but they achieve their goals. Maybe with a price they hadn't anticipated, but they succeed just the same. They will deny it, they will tell you that they like the succeed part just by itself, but they are wrong or they are lying. They like encounters that balance on the razor's edge - and why not? If the encounter was too much for them, they'd have been smarter to find a way around it. If it was a cakewalk, why even bother playing it out? It's the instances where it can go either way that are worth playing.
One of the great things about Dungeon World is that it allows you to dynamically adjust the difficulty of encounters instead of having to set everything up exactly beforehand. If your PCs are bowling over the goblins, they're going to succeed, but if you need to make them bleed before they do, use the horde tag to make them call for backup from their kin! And use organized to make sure they fight like intelligent creatures fighting for their lives - they'll volley with arrows and divide-and-conquer!
I highly recommend reading The Dungeon World Guide - it provides lots of great examples specifically on combat and hard-vs-soft moves.
