Tags

, , , ,

This is how many players envisage their characters to end: Against a proper and dangerous opponent

After a recent discussion about “What do I do if the party suffers a TPK”, which I intend to also elaborate on the blog later, I brought up a rule that I first saw in 13th Age. In 13th Age one of the interesting optional rules is that the adventurers can’t be killed by random “Chaff” type monsters. Instead, they have a “Named villain/nemesis/antagonist” rule where only the essential, named enemies can result in dispatching the players characters. When I first read this rule about “Named villains”, I originally entirely dismissed it.

On thinking about it after the discussion about avoiding TPKs, perhaps I was a bit hasty as there is actually some merit to the idea – especially depending on how strongly you tie your PCs into the overall narrative.

The problem with untimely character death

Of everything that can happen to a roleplaying game, player characters dying to a random group of jobbers by poor tactics, ill luck or otherwise is often one of the most problematic ones. For one thing, it’s very rarely ever satisfying to know your character was accounted for by random chance and you couldn’t roll above a 2. Alternatively, the time where boulder trap #7 squished everyone (or just you) as the rogue didn’t come to that session – so nobody qualified could check for traps. Or that time someone (or just as frequently, everyone) died, because of a ridiculous monster that the DM under-estimated in difficulty.

I could go on, but suffice to say there are a multitude of reasons for why characters can die in your average Dungeon crawling based adventure. For a lot of games, which aren’t inherently focused around any particular storytelling and random coincidental death is a core part of the fun, this is generally fine. However, if you’re tying your players characters into the overall plot or world, from a narrative and gameplay point of view, untimely deaths often cause considerable issue for both the DM and the player. With the plot entwined with the player characters fates (or lack of them), it creates a big problem for your story and world, which now needs to account for their death in some way. This is also compounded by the time it takes to come up with and make a new character, which can remove that player from the rest of the entire session.

Tying players characters tightly into the games world and narration, while making them “mortal” makes every “general” encounter especially risky to the story. Removing this potential story risk is probably the main reason 13th Age’s designers came up with the “Named Nemesis” rule in the first place. Especially because 13th Age models heroes who already have an important destiny and place within the game world. Using this rule, the player characters subsequently don’t “die” to the regular jobbers or traps encountered, but take debilitating injuries or suffer other complications instead. At the same time named villains and antagonists suddenly get a more deadly “aura” about them, because the threat of death suddenly becomes real.

The argument for deadly encounters and character death

DEATH IS SASSY

Before I continue with the merits and potential applications of this “Named Nemesis” rule, it’s important I talk about why I initially disliked the concept. For one thing, I believe it’s definitely not a bad thing to let player characters die, including when they are important parts of the story themselves. In many ways, I don’t mind my games being a bit “Game of Thrones” – at least in theory – with potential main protagonist death around every corner.

This doesn’t mean I am particularly lethal as a DM, because in all my current games despite threatening death several times, I’ve not actually killed any of my players characters (yet). It is the potential of death that is important and the subsequent removal of that risk, well it makes me feel instinctively uncomfortable. This is because if the potential of character death is removed, it inherently changes the perception of an encounter to have less risk vs. reward. With no real risk, what’s the harm in engaging in more encounters and will players treat those same encounters seriously anymore? Is a battle with Yuan-ti archers on a hill, with rocks falling down around the characters really that dramatic anymore if nobody will potentially die?

Another issue with this concept, is that sometimes a hero standing alone against all odds and falling to a nameless horde is a great piece of story telling. The classic example of this is Boromir from Lord of the Rings, who essentially is killed by a large number of “jobbers” while trying to save his companions. This moment is an essential part of Boromir’s character arc and redemption, most notably because he attempted to take the ring (a powerful artifact) for himself in an earlier scene. Under a philosophy like the “named antagonists” concept above, this scene isn’t one that makes a lot of sense because you’ve established heroes won’t normally fall to jobbers. So you’ll need to kludge in a solution like (ironically) something similar to the LotR movie, where Boromir is actually finished off by a specific (named) enemy.

The other really big problem is the whole “What do I do if they all die to a random ochre jelly or otyugh, which by all accounts would just consume them?”. Now you’ve got to potentially put in an unsatisfying or verisimilitude breaking solution of some sort, like perhaps a last minute rescue. Even in the cases of non-TPKs and the more common “The wizard bit the dust, oops” moments, you still have to decide if there is some consequence for failure. Often tying yourself up in knots trying to explain why they didn’t die, is harder than actually just accepting the TPK/player death in the first place.

So let’s bear in mind these are my main criticisms of the idea, but let’s go over some of the ways this idea has some genuine merit.

Putting the spotlight on established villains

Climactic Battle

If there is one thing I think this rule accomplishes brilliantly, it’s that it inherently gives the most gravitas to the most important characters the DM has in the story: Their main named villains. For example, Gurok the Headsplitter, whose brutal raids on caravans up and down the Forest Road trade route, has an impressive bounty on his head from the local Trade’s Guild. When the adventurers take up this challenge, facing Gurok personally automatically has an aura of menace about it. As Guroks a named villain, any confrontation with him is going to have much more narrative meaning and higher stakes than his various underlings.

In fact it’s arguable that the entire point of the “Named Nemesis” rule, is to ensure all the adventurers do reach this climactic battle with Gurok. Once they are face to face with the Ork leader, the fact that he can actually kill the player’s characters means the players (ideally) feel more tension. Effectively it puts all the danger and risk into certain encounters, which does give you as the DM a big advantage. If everyone at the table knows that the battle with Gurok is the one where they can die, you can bet the players pull out their biggest spells/abilties there (or save them for it). Thus, you can predict roughly how difficult you want to make the encounter more easily. Especially because you know you won’t risk randomly losing a character to a bad encounter along the way.

Another big advantage of this system is that once in place, you have all the reason to safely eject my pet hate about Dungeons and Dragons like systems: Resurrection based magic. In many ways, the ability to raise characters back from the dead or outright resurrect them is actually a similar kind of narrative device as this rule. It allows players to “fix” an unfortunate death to a save vs. die (or similar spell), or being eaten by the Tarrasque or whatever it is they encountered. This is particularly because this magic tends to be available for higher level characters, with it being reasonably trivial to actually perform after a certain point. Have you ever asked yourself why that is? It’s because remaking a new high level character from scratch is a royal pain in the ass, so raise dead is trivial solely to keep the game going.

In games where I would use this concept that a character could only die to a named nemesis, I would equally back it up by removing all forms of resurrection or raise dead magic. Now when the players are battling Azerkhaban the Green Dragon, they know the stakes are through the roof because not only can they die, they will stay dead. The battle with the dragon, which is probably the climax of the adventure, develops an extremely high level of tension by default. Meanwhile the adventurers are encouraged not to use their best resources beforehand, such as fighting through her 20 odd kobold servants while exploring the lair.

Another good thing that this rule can provide is fixing a problem I have with the newest edition of Dungeons and Dragons. In 5th Edition DnD, new characters are particularly prone to death at the slightest bad die roll. While this is rather intentional as Mike Mearls on reddit described this as modeling a Game of Thrones “Anyone can die at any moment style”, it doesn’t actually hold up very well. New characters can trivially drop like flies, while more experienced adventurers can prove remarkably hard to actually even dent significantly. Once again that issue where resurrection and raise dead magic, especially for higher level characters, becomes rather trivial to access also becomes pertinent.

If you think about it, this means that 5E isn’t modeling a GoT style very well whatsoever. In GoT certain characters, without spoiling anything, would be considerably harder to ever kill off (due to being equivalent to high level adventurers). From my experience with the official modules, like Hoard of the Dragon Queen and Rise of Tiamat, or even my own campaign Curse of the Black Pearls, once characters live past level 5 reliably hurting them significantly can be challenging*. Actually threatening to kill them can be even more challenging in many ways, especially in the official modules where many encounters aren’t anywhere as dangerous to a well established group. In essence the “Anyone can die anytime” concept doesn’t last beyond a certain point, which simply leaves the game too lethal initially and not lethal enough later.

Using the named villain/nemesis rule in Dungeons and Dragons, it allows you to emphasize the core villains, safely remove all safety net of raise dead/resurrection magic and make climactic battles significantly more dangerous (safely, as players should also be carrying more resources into them). At low levels in 5E, if you don’t want player death to occur to just a handful of bad die rolls, just don’t use any named villains. This means low level adventurers can more confidently explore and take more “heroic” risks, without instant threat of potential death. Noting that this *is* a trade off between tension and letting a character suffer the full “consequences” for their actions**. When you want to threaten your players and make them worry about death? That’s when you introduce your monstrous villain.

In essence, this rule and mechanic is all about building up your core villains. It’s where the idea has considerable merit and application. While it’s definitely not suited for all styles of game and it does heavily reduce normal “regular” encounters importance, the emphasis on climactic encounters is appropriate from a storytelling point of view. In a game where you want your players characters more central to the story, say by having Juliash be the Elven scion of a lost empire, their death to a random gelatinous cube is… not ideal. Juliash giving their life to stop the villain Necroshant, who was moments away from destroying their newly rediscovered Elven empire, makes for a better “heroic” moment (and ending).

But then again now that I’ve thought about it, there is an equally good story about an elf who thought they were long lost royalty who got digested by a cube of acidic jelly….

*This weeks “Ask the Guild” will be talking about some issues around deadly encounters, low levels and how to get around them in 5Es module “Princes of the Apocalypse”.

**A future training day will explore how to make characters choices have more meaning and impact, outside of “You died horribly!”

Advertisements