I recently read through The Derelict, a one-shot scenario for Call of Cthulhu released for Free RPG Day (I bought my copy the normal way, though). This post will contain massive spoilers for that scenario, so be aware; proceed with caution.
Some of you may remember my post about Monster Balance in Call of Cthulhu, which is also relevant here.
So, the scenario features a group of Investigators who rapidly find themselves trapped on an iceberg with two wrecked ships.
After finding signs that the crew of the other ship were massacred in a way that doesn't look like either humans or polar bears, they eventually realise they're being hunted by an inhuman monster.
The villain of the piece is a "sciapod". This is basically an Ice Yuan-Ti, for those of you with a D&D background. Okay, it's a huge ice merman covered in chitinous armour, which is invisible (with one specific exception), carries a bow that conjures steel-hard ice harpoons, and likes to eat humans.
The rest of the scenario will likely consist of the Investigators cautiously making their way around the ships finding some more horrific scenes, and potentially a couple of mildly useful items. They will almost certainly be picked off one by one, though a couple might escape to dubious safety on the ice waters.
Prompted by my reading and my reservations about the scenario, what I actually want to talk about today is the use of single entities as antagonists. Specifically, I want to consider some of the drawbacks and difficulties presented by this situation.
In many ways, I think you can sum this up simply as having all your eggs in one basket.
The eggs take a variety of forms: psychological threat, mechanical threat, persistence of the problem, capability of the antagonist, and so on. Regardless, packing them into one bundle tends to give you a blunter and more fragile tool to work with.
Coming back to my previous article, the Monomonster is going to be a Beast, Hunter or Final Horror because as a lone operator it can't really be anything else. In theory it could have a dual role as a Gatekeeper or Messenger as part of a larger storyline.
One Man Army
The most prominent of the issues is physical confrontation, so that's where I'm going to start. For this section I'm focusing on scenarios where the monster* presents a direct physical threat to the survival of the PCs. This may not actually be its primary danger, or its preferred way of interaction, but in a direct confrontation the monster is a serious threat.
A lone monster of this kind is significantly more powerful than any one PC, in terms of whatever combat-equivalent features in the game. Usually, it is at least as powerful as all of the PCs combined. There's often a straightforward reason for this, which is that if the monster wasn't this powerful, it wouldn't present a credible threat to the party.
*I'm going to say "monster", which should not be taken to assume a D&D perspective on all this.
From a structural perspective, the danger of the situation is twofold. Firstly, the monster might win. Secondly, the monster might lose.
Yes, I am feeling glib today! How did you know?
To put it another way, every combat* encounter between the monster and the PCs takes on an existential nature. If the PCs do somehow manage to defeat the monster, then the scenario is effectively over. The psychological threat is lifted, the mechanical threat ceases to exist, and although the GM may have some narration or investigation to offer in closing, this tends to feel like a wrap-up. The PCs can achieve a sudden victory, but with only a single opponent to confront, it's difficult for them to slowly gain the upper hand or see the odds growing in their favour.
*I'm just going to say "combat"; there might be other forms of confrontation that work in equivalent ways in specific games.
For the GM, this poses a challenge, because if there's any possibility that the PCs might defeat the monster through a series of inspired decisions and lucky rolls, well, there's a possibility that the PCs might win through a series of inspired decisions and lucky rolls. And if that possibility crops up five minutes into the scenario when the PCs first encounter their opponent, that will make for a very short gaming session. In the case of The Derelict, if the PCs somehow manage to kill the sciapod shortly after reaching the wreck, there really isn't anything left except gathering some backstory.
An obvious and common solution to this problem is to ensure that the monster cannot, in fact, be defeated through a series of inspired decisions and lucky rolls. The sciapod (to continue with our example) has 45HP and deducts 5 points from any damage inflicted; the PCs have three firearms with 6 bullets apiece dealing roughly 1d10 damage. Now, hypothetically, this means the PCs could deal 90 damage to it, killing it twice, without even taking extreme successes (extra damage) into account. The chances are astronomical through. If the PCs understood what was up very quickly, were all together with weapons drawn, and managed to launch a surprise attack on the (invisible!) creature before any of them were killed, it might just about work. Or it could continually roll 100s. I'm not going to say never. But it's incredibly unlikely.
If that thousand-to-one chance comes up, though, the scenario may fall flat. It doesn't actually have to happen that early, either. It just needs to happen at a point when the narrative momentum points in another direction.
Hero Al-Jaf and Bugsy Warburton have finally discovered the horrific experiments behind their predicament, and realised they need to ransack the Science Building where Dr. Sinister's notes and equipment may provide them with a faint chance to escape alive. As they prepare to leave the office, they hear the terrifying sound of the Slatcher gnawing at their door. They turn tail as the Slatcher bursts through, firing wildly in a desperate attempt to buy time.
In a stunning display of serendipity, three bullets strike the Slatcher in the sensitive nostril region, overloading its nervous system and destroying its brain. With the hulking creature dead, our heroes have no particular reason to go to the Science Building, and begin to shuffle awkwardly home.
If you have a number of monsters stalking the PCs, this isn't a particular problem. Even if each monster is supposed to be extremely tough, the players will probably understand that they got really lucky and still shouldn't try to confront the remainder outright. But with only one, you have nothing left to play with.
We're All Dead Here
The parallel problem to this is that the monster can win too easily.
Remember, because we have only one monster in this scenario, that monster is overwhelmingly powerful. It is capable of defeating at least several of the PCs - which, given mechanical variation and raw luck, means that it's almost certainly capable of killing all of the PCs at once. Often this is deliberate; the intention is that PCs will either escape the monster, or work out a way to overcome it indirectly using the environment and information they can accumulate, or increase their capabilities (either mechanically or in terms of knowledge and resources) to the point where they can in fact defeat it.
Because the monster is so powerful, any time there's a confrontation with the whole party, there is a risk that the monster will kill them all. While that risk may ultimately be part of the scenario, once again it can feel very unsatisfying if it happens too early, or at a time when the narrative momentum isn't pointing that way.
If it does come down to combat, then, the GM has the difficult job of reining in their beast. They need to calculate fairly precisely just how much damage it should wreak. If they simply follow the mechanical potential of a rampaging monster capable of killing the entire party in a few rounds, there may come an abrupt end to the scenario. Conversely, if they restrain it too far, the psychological threat can be lost; this not only affects the narrative, but can cause further complications because players become tempted to keep fighting. After all, if the creature doesn't attack as ferociously as they expected, this may be their best chance to defeat it!
The psychology of the players, and indeed the player characters, can get convoluted, especially with metagame considerations coming into play. You can of course resort to saying "hey, this monster could easily kill you all, but I'm giving you a chance to get away," but there's a real risk that this will puncture the immersion too far - players may end up feeling that everything is under the GM's sufferance and become dissatisfied.
Basically, then, the difficulty for the GM is knowing when to stop. On top of choosing a narratively and mechanically appropriate amount of havoc for the monster to inflict on the characters, bearing in mind the psychological effect of that havoc and the tropes that will be evoked by it, the GM also needs to have in mind a pretext for stopping that feels satisfying given the nature of the monster.
Players can be surprisingly stubborn about backing down. It's not super surprising though. After all, often the whole point of encountering a monster in a game is to fight it.
I think one of the particular difficulties of a monomonster is that it's difficult to give a useful sense of scale. By which I mean: exactly how much of a threat does this thing pose?
Of course, there are some exceptions. If a monster is genuinely meant to be overwhelmingly dangerous beyond anything the PCs could dream of, you can generally find a way to demonstrate that just in description. In that case, the scenario is about staying as far away from it as possible.
When as a GM you have many monsters, you have options. The most straightforward is that you can have some of them attack the PCs (or some NPCs) and show roughly what kind of threat they pose. A Venusian Gnarker might charge a group of guards, only brought down by a relentless barrage of gunfire after smashing several aside. Several Chibblers might spring upon a PC, and when the party can wrestle them off, it shows that four PCs can cope with about four Chibblers without getting too seriously hurt - but the hundreds encroaching on their base will be a problem. They might know that one unarmed cultist is a relatively manageable threat, but that they can't really handle the fifty who are meeting downstairs.
In particular, you can expend some of your monsters to demonstrate the scale of the threat. Was it easy to kill them? Hard? Terrifyingly difficult? Okay, now you know what you're dealing with. And that was just the hatchling...
What scaling is about is informing the players of the situation. You are trying to convey a sense of appropriate threat, bearing in mind the tropes and conventions of the genre, so that the players understand what it means for them to take various courses of action. This is all part of the general GM's role, just as in portraying an NPC you try to show the players what kind of NPC and situation they are dealing with (with reference again to genre tropes).
With a monomonster, you can't easily do this.
You can have the monomonster attack someone else, which can work well, though it depends on how the players interpret the power of that victim. There's also a risk that the PCs think this is a "rescue the NPC!" situation, because they haven't yet worked out what they're dealing with. Finally, some kinds of monster don't work well here because they should really move on to killing the PCs.
Another classic is leaving evidence of the destruction wrought by the creature. This is potentially very potent; in particular, it allows the PCs to study the scene at their leisure and learn a lot about what the creature can do. It can include witness accounts and scenes of carnage. However, it can be difficult for players to grasp what these mean in mechanical terms. Is being able to wipe out a roomful of police officers evidence of appalling power, or just that the police are a bit useless? Are we supposed to be scared about this for practical reasons, or aghast at this tragedy, or is it just the kind of thing that happens? How did those police match up to us, anyway? Of course, this depends on things like system. In D&D, you don't know whether a dead NPC is 1st or 20th level unless you're told. In Call of Cthulhu, people are mostly fairly equal, but skills can make a huge difference.
It is possible to have the monster simply attack, working on the basis that the party will mostly escape. This might work absolutely fine, but it's risky. There's that aforesaid worry about killing everyone, especially if they don't realise they're not supposed to fight back. Or the monster getting killed by a fluke.
You could also end up with more complicated situations. For example, killing off even a single PC early on might leave the game feeling flat for their player (even if you get a replacement, your investment can be diminished). Characters might be sufficiently injured or traumatised that it's actually difficult for them to do very much, especially in a game with injury mechanics. For example, a sprained ankle or fear of darkness could leave a player struggling to contribute. If the game has resource management, such as mana or even ammunition, it's possible for someone to burn through a lot of their resources in this kind of preliminary encounter and have little left for the main scenario.
In addition, while a monomonster can manifest new abilities or increased power over time, it can be very difficult to plausibly do the opposite. Monsters achieving new stages of evolution, greater manifestation onto the material plane, absorbing enough souls to breathe fire? There's precedent for that. Monsters losing the ability to breathe fire, or weakening, except due to protagonist intervention? That just feels like pulling punches. And if you try to repeatedly scale up and down, it will certainly begin to feel arbitrary.
Somewhat strangely, I think another issue with this kind of scenario can be player agency. Now, it's tricky because there's also a loss of character agency in many of these scenarios, which relates to most of them being horror. But I think the monomonster setup does tend to limit player options as a consequence of the limited range of narrative resolutions.
Let me try that again in less portentous language. As I've said, there's a significant risk of the narrative falling flat because the monster, or the party, are killed off at an unsatisfying time. The power of the monster tends not to leave a lot of leeway for error, which tends to mean limited scope for players to test ideas, explore and take risks. This can either result from player caution (including playing strongly into trope on the assumption that this is the only way to meaningfully engage with the scenario), or from GMing that steers into Quantum Ogres and railroading territory.
Let's take the sciapod as our example.
Freedom of Movement
The scenario obliges them to land on the iceberg. It then mandates that their boat is disabled, and the radio destroyed. This immediately removes two choices: the players cannot decide to have the characters call for help, including further information, nor can they decide to try and escape. This is a story about being hunted by a monster on a shipwreck, not about being chased by a monster across the Arctic and beyond. Fair enough, unless you're running a total sandbox you do need to decide what sort of thing you're trying to do in a scenario, especially in genres where GMs are expected to devote effort to building up backstory and providing realistic details, which require planning and research.
The players may wish to gather information about what is going on. This is a sensible choice and also in-genre; protagonists faced with an unusual situation do generally try to puzzle it out, and in horror genres this often overrides caution. The players aren't prevented from information-gathering, which is good. However, the majority of the information is generic atmosphere-inducing detail, demonstrating the power of the sciapod and the danger they face: the crew thought up several logical ways to protect themselves against something, all of which failed.
It's interesting to learn what happened here, but it doesn't actually help. Notably, they can't find anything that explains what the sciapod is, its weaknesses (spoiler: none) or how it behaves, any of which would increase their ability to make effective decisions. A very notable omission is any clue to the secret means of seeing through its invisibility, which is very much out of left field.
I suspect this is unusual. Stories of this kind often involve the protagonists observing the monster's behaviour (or the clues it leaves behind) and slowly learning of its quirks and vulnerabilities, eventually equipping them to escape or defeat it. This helps them make informed decisions, offering agency. On the flipside, there is a risk that the scenario becomes a series of scripted reveals, with the players needing to progress though a whole clue chain before they are allowed into the endgame. Again, this is a bigger risk with a monomonster, because a monomonster is a bigger threat allowing less margin for error. With multiple monsters, they can gradually gain the knowledge to take on increasingly powerful monsters, or test out leaps of deduction and survive being wrong.
Other possibilities are significantly curtailed by that invisibility. It deprives them of information about what's happening, including the ability for characters to observe the creature and interact with it. It's technically possible to overcome the invisibility, but the solution is bizarre and isn't spelled out anywhere, which means the players don't really have the option to work towards discovering it.* The invisibility is an unusual trait for a monomonster, though. Shapeshifting and similar abilities can produce some related issues, but typically you can at least see the monster when it's attacking you.
Technically you could try and work it out by experimentation, but it would be insanely dangerous, and I'm still not sure players would work it out. It's also very hard for the GM to help out without just telling them the answer.
Judging the Odds
The sciapod's superhuman strength and toughness, and ability to kill at long range while being totally invisible, make fighting it largely pointless. My impression of the scenario was that by info-gathering, players would mostly judge that the monster is far too powerful to confront. It's made clear that a large number of hardy sailors taking sensible precautions, including some decent improvised weaponry, were unable to stop it.
While that doesn't stop the players trying, I think that is a reduction in effective agency. Within the context of horror, signs that an enemy is overwhelmingly powerful in a fight is typically a flag that you are not expected to fight it. Players may not mind their characters dying, but if you're confident the attempt will be fatal, it turns the choice to fight from "part of your range of interactive options" to a specific narrative decision or a last resort.
The players' choices seem very limited here, because there seems very little they can meaningfully do either about the sciapod, or about their own survival.
They can't call for help. There's nowhere to run to, and they can't actually see what they're running from. They can't really hide and it wouldn't achieve anything. They can try to fight back, or just defend themselves, but the scenario makes a concerted effort to show them how ineffective it would be against this inhumanly powerful and resilient creature (which is, let's remember, able to kill them with one shot at long range while remaining totally invisible). They can't observe the creature's behaviour and exploit that, because it's mostly invisible and doesn't seem to have any behaviours other than killing, and also trying to watch it is insanely dangerous.
Why is this about monomonsters again?
Okay, so I'm using one creature from one scenario as an example. They're not all the same.
Basically, a monster story tends to have one of two satisfactory narrative structures. In the first, the characters begin by fleeing and hiding, gradually gain knowledge or resources that improve their odds, and finally defeat the monsters in a series of climactic encounters. In the second, the characters begin by hiding and fleeing, gradually gain knowledge or achieve small victories that improve their situation, and finally find a way to make their escape. In both cases, what essentially happens is that over time, the monster's relative threat decreases until it is less than the characters' growing ability to achieve their goals.
In an interactive roleplaying game, things tend go to in unexpected directions at best. There's no special reason for a confrontation with a single monster to escalate nicely to a dramatic finale. This is especially true as players do not usually have perfect knowledge of the relative capabilities of their characters and the monster, or the odds in play, and are not necessarily aiming to construct a satisfying narrative. You wind up with situations where the players are expected to encounter the monster early, to identify that it is a threat and that they cannot defeat it, but for them to also somehow survive the early encounters, despite the fact that the monster is motivated to kill them, and capable of killing them.
Achieving this balance is particularly difficult with a monomonster because the GM cannot easily use convenient variation in the number of opponents, or their maturity or armament, or their apparent intelligence, to vary threat levels. Instead, they have to control the monomonster's behaviour to ensure it does not kill the players in the early encounters. Because it is controlled in this way, the significance of the players' choices is reduced, and thus their agency. In addition, as I mentioned in scaling, if the GM tries to vary the threat posed between encounters (and in particular, specific abilities) the players will struggle to make informed choices.
With multiple monsters, the GM can open with a Chibbler Larvae encounter to demonstrate the threat. It doesn't need scripting: they can design the Larvae to be moderately threatening and leave the players to respond as they wish, with confidence that the characters will survive and emerge with some understanding of the threat. When the characters encounter apparently more intelligent Chibblers, or find signs that a huge Chibboth has passed nearby, their players can make informed guesses as to the consequences of their next actions (run, fight, hide?). The GM can also scale threat up and down: once they've avoided a Chibboth, they can find a couple of Chibblers in the next corridor.
Except, I'm missing something. Quite often when monomonsters are used, helplessness is the point.
We're All Doomed
The story structures I mentioned above are either defeating or escaping the monster. I suppose third and further structures include the monster gradually picking everyone off, typically except for one surviving female character and possibly a cute pet. You can potentially view this as a Final Horror incarnation of the monomonster, from my Call of Cthulhu typology.
Some of those are effectively lone protagonist stories with a supporting cast, which you can view as working the same way as my two above, but being less appropriate for group play. Others are all about the characters' helplessness, which is fine if you enjoy that sort of thing, but rarely leaves much room for player agency.
I would tend to view these as being more suitable for a game specifically designed to produce a particular narrative - a storygame, in other words. I don't think they're a great match for more traditional roleplaying, because when characters don't have much agency, the GM needs to work very hard to give players agency to set short term goals they can achieve. These may be in-game goals like discoveries, acts of heroism and so on, or player goals relating to bringing about particular scenes or narrative elements they feel invested in. These are necessary because more traditional goals like "defeat the monster" or even "survive" aren't really within reach.
Players need to be thoroughly on board for this sort of thing, because if you find yourself in a monster scenario and try everything you can think of to defeat or escape the monster, only to learn that neither was physically possible, it will be massively frustrating. Essentially, it leaves you playing the wrong game. You want to be playing with the grain of a game: doing the kinds of things that are meaningful within the paradigm you have adopted.
If the monster is unstoppable, you want to be setting and achieving goals like "I stay alive as long as physically possible", "I radio HQ with as much information as we can get our hands on so they can stop this spreading", and "I make one hell of a last stand".
Of course, it's also quite possible for this kind of narrative to be a natural outcome of a monster story where the characters just get themselves killed, through bad luck or bad planning.
On the plus side, this is probably the genre best suited to a monomonster - it actually works quite well. The monster seizes a single victim at a time, and then stops to consume them, torture them, meld with them and assume their identity, and so on. The others can flee while it ignores them. In some ways this is actually better than with multiple monsters. Since you aren't worried about PC deaths if you're using this structure, there are relatively few problems. The main ones are ensuring players are on board with this, and that they don't get too bored if they're eliminated ten minutes into a six hour session.
The Great Escape
So you're faced with a terrifying monster, and just want to escape.
This one comes down very much to the other parameters of the story, because those determine how difficult escape is, and indeed what "escape" means. Are you trapped in a submarine? On a base in the desert? In a forest? In a city?
Monomonsters, being incredibly powerful, are a natural fit for the escape plan. However, they do also have some obvious drawbacks, the main one being that escaping from one thing is generally an awful lot simpler than escaping from many things.
Very broadly speaking, if the characters can determine where the monomonster is, they can sensibly proceed by getting as far in the other direction as possible. Similarly, it's well worth making Herculean efforts to lure the beast away, conceal their trail, block its passage or trap it even temporarily. It may even be reasonable for one character to sacrifice themselves so that the others can escape pretty much unhindered.
For example, let's say the creature is basically mortal, just incredibly tough - it can't pass through solid matter. If there is a way for the characters to separate themselves from the creature with a nigh-impermeable barrier, it's well worth doing even if it's extremely difficult, because achieving it will solve most of their problems. If there are two parts of a space station, and they can sever them using a complex series of explosions jury-rigged from food supplies, they are safe from the monster. While a perfectly good solution, the players may end up focusing on this to the exclusion of whole swathes of a much more complex scenario.
Of course, this isn't always a bad thing. It may be very effective for the players to feel that they're making good progress in their escape at times, and other times have the psychological pressure of knowing the monster is near. The GM may well be able to come up with ways for the monster to eventually resume the chase, hopefully without leaving the players dissatisfied.
The advantage of multiple monsters here is that the players know they do not only have to content with one enemy. This means concocting (potentially) cheap shots is not worthwhile; they should pursue more general escape plans. They can't be as confident in their safety simply because they know where one monster is, or even because they trap one. In addition, the monsters can (deliberately or accidentally) herd them via their own movements, or else force them to take risks to avoid being driven in the wrong direction. The players have to consider pincer movements, or the risks of fleeing one enemy at high speed only to run headlong into another.
Plus, if the numbers of the monsters are uncertain, it may never be clear when they are actually safe.
A second advantage here is the possibility of tiered success. For example, the players could make definite progress by cutting themselves off from part of a monstrous horde, then gradually see more and more of their pursuers drop away as their plans succeed. Or they could make the difficult decision to leave a holdout where supplies are running low, pass through a densely-populated region to make quicker progress, and then try to shake off pursuit. They might accumulate pursuers as they make mistakes, and have to shake them off. They might even be able to turn some of the monsters against each other, or otherwise have the monsters' sheer numbers become a disadvantage - "surely that bridge won't bear the weight of four such gigantic beasts?"
The Nature of the Beast
The personality of a monomonster is an important facet of the puzzle. Because it's the main (or only) challenge in the scenario, its nature will shape the kinds of encounters the characters tend to face.
A fairly basic monster is simply murderous: it has an instinctive drive to kill. This can make it both simple and difficult to deal with. When it notices the PCs, it attempts to kill them in a fairly direct way. Since our monomonster is also very powerful, this tends to mean it will kill every PC in sight unless they manage to escape it.
The difficulty here, then, is making the murderbot clever enough to present a practical challenge, but not clever or fast enough that the PCs cannot escape it. It probably shouldn't just fall into pits because the PCs stand on the other side, but it shouldn't be able to unerringly hunt them down. If the murderbot is both overwhelmingly powerful and inescapable, there probably won't be much to the scenario once they first meet - it will chase them down and kill them all efficiently.
Useful options here include weaknesses and incapabilities. If the murderbot can only come out in the darkness (a fairly common trope), the PCs have some limited opportunities to move about, gather and use resources, and make preparations in relative safety before their next encounter. Similarly, if the murderbot is too large to fit through vents, the PCs may be able to escape from a confrontation, while keeping them highly restricted: they can't afford to spend long in one room, and they're limited in terms of where they can safely go at all.
Not all Murderous monsters are instinctive or even stupid - you can use a ruthless assassin, for example, and The Predator is probably a decent example - but the impulse to kill is typically stronger than any other motivation.
In some ways, the Hungry beast is even more simple, but it's also potentially more complicated. That's because it actually has two motivations: kill things, and eat them. These can interact in interesting ways.
When a Hungry beast detects the PCs, it will tend to try and kill them. However, if it does kill someone, its desire to eat will often trump its desire to kill. This gives the PCs an opening to escape, counterattack, or attempt a cunning ploy. Whereas the murderbot would simply move onto the next victim in an efficient flurry of death-dealing, quickly killing the whole party, the Hungry beast has no reason to do so until it's sated its hunger.
Hungry beasts can be cautious, like many real-life predators who are wary of prey's defences and the simple risk of their food getting away through their carelessness. Others are so driven by appetite that they have no such qualms, and fling themselves ravenously on any potential food. Either behaviour can give PCs opportunities to trick, distract or otherwise control them. Similarly, many Hungry beasts aren't exclusively PCvores, so the players may come up with ideas to use food stores, wildlife or even NPCs as distractions and as bait.
In my limited experience, monomonsters often seem to be treated as Sadistic. I think that's because this is the most convenient for GMing purposes: just as the GM is playing a game with the players, the Sadist is playing a game with the PCs. The GM is trying to create a sense of threat, and the Sadist enjoys scaring the PCs. The GM doesn't want to end the scenario too early because it's fun; the Sadist doesn't want to kill the PCs too early because it's fun.
Sadists tend to have at least a moderate level of intelligence because it's sort of hard to be sadistic otherwise. They deliberately make things more challenging for themselves, allowing victims to escape and avoiding the use of their full capabilities. While a monomonster can wipe out the whole party at once, the Sadist simply doesn't want to.
To a large extent, this works well. The Sadist has a perfect excuse for being irrational, inefficient and frankly slipshod about killing off the PCs, even though it's perfectly capable of doing so. However, you can reach a point where this begins to fray under pressure. If the PCs begin to demonstrate that they can in fact take on the Sadist, it should begin to make more of an effort and fewer allowances. Unless the Sadist's goal is quite specifically to play a complicated and genuinely dangerous death game with the PCs, its behaviour should change in response to their capabilities.
For example, a Sadist who resents the intrusion of puny mortals into its lair, and is now sadistically terrifying them before devouring them, might logically stop to think after a mortal successfully hurts it, and move on to simply devouring them. A Sadist that relishes the sense of power over helpless fleeing victims would lose that satisfaction if the victims turn out to have a very sensible plan to destroy it, and abandon the game (it's basically a bully, after all).
One of the difficulties I had with The Derelict was that it's ambiguous about the nature of the sciapod. It's described as wanting to kill everyone, but also as eating human flesh; however, it doesn't seem to behave as though driven primarily by hunger, but if it wants to kill everyone it can do so almost effortlessly. The designers perhaps intended it to be sadistic, but this kind of thing needs to be clear to the GM. I also feel slightly that making a single powerful and sadistic enemy is perhaps a bit overdone.
A clever monomonster, sentient enough to act like a human, can have very complex motivations and conflicting goals. For example: the Slaughter Spirit wishes to kill you all because it thirsts for death, and will actually try harder if the situation is challenging out, of pride in its power and a wish to show contempt. On the other hand, it resents having been summoned by a wizard to guard this place, and sometimes the wish to sabotage its master's goals overcomes that bloodlust. It also hates wizards in particular, so magic-using PCs will earn particular ire and frenzied efforts to kill them. But it's also afraid of wizards. It is bored after centuries of guarding, and doesn't really want this brief excitement to be over. And so on.
This is potentially very good from the GMing side - it gives you a lot to work with. However, it can also be hard to put across this amount of detail to the players. This can make it hard for them to usefully interact with all your nuances, and the ability to interact sensibly is pretty vital in the survival-type scenarios that tend to feature monomonsters.
I suppose a final type of monomonster is simply Erratic. While it does have goals and motivations, it doesn't pursue them single-mindedly. This can explain why a monster doesn't relentlessly attack, pursue or even pay attention to the PCs.
An Erratic monster might be stupid, animalistic, mad, malfunctioning, struggling against a controlling force, or just alien. In some ways this can be useful to a GM: they can change the monster's behaviour as necessary to ensure the scenario doesn't end with an unsatisfactory splat.
The downside here is that as I said, learning how the monster behaves and how to interact with it is one of the few things players can actually do in most of these scenarios. After all, monomonsters are overwhelmingly powerful. "Winning" the scenario normally boils down to either tricking the monster to death, evading it long enough to escape, or learning enough about it to defeat it. Either option really calls for the players to work out how the monster ticks; otherwise it is liable to feel like GM fiat.
Stunlocks and the Action Economy
I've been talking with a fairly Call of Cthulhu mindset, but let's take a step back here. Some games have more of a tactical combat approach.
The classic problem with monomonters here is extremely well-known: it's basically one of actions.
The PCs are typically a group of 3-6 characters who each get to do one(ish) thing on their turn. The monster is typicallly a single entity that can do one thing on its turn. Although the monster's individual actions are typically more powerful and can often affect multiple PCs, it's still easy to end up with the PCs running circles around it due to their ability to do multiple things at once.
As a basic example, while the monster is fighting one PC (100% of its actions) the other 3-5 PCs can act freely. They may choose to contribute to the combat, but they might also be healing, preparing equipment, executing a complex plan to trap the monster, performing a powerful ritual, summoning help, running away, bombarding the monster with hindering effects, and so on. Generally there are multiple PC actions available to counter the efforts of the monster, while the monster has one set of actions available to counter the efforts of the PCs.
The archetypal problem is the Stunlock. An effect reduces a target's actions or even prevents actions. If PCs are able to create such an effect, they can use it against a monomonster to deny 100% of its actions for the turn. While they are totally unopposed, the PCs either unleash a flurry of powerful abilities, or carry out a difficult plan. In some cases, PCs can repeatedly inflict these effects to keep the monomonster Stunlocked and make the encounter flat and boring.
The issue here is that abilities in a tactical combat game tend to be weighed on the assumption of multiple opponents. Stunning powers are not inherently overpowered, but can become situationally overwhelming (I've discussed this sort of thing in my many posts on soft attacks). Against a single powerful opponent, any impairing effect is generally hugely valuable. Alternatively, designers react by making the monomonster highly resistant to such effects, to the point that many abilities are largely useless. A common result is the Save or Suck - depending on the your roll, your soft attack either bounces off harmlessly, or is devastatingly effective.
This sort of thing alters the relative effectiveness of different characters depending on the types of powers they can bring to bear. If a monomonster can be stunlocked, characters with potent soft attacks are devastating. If it can't, those characters who aren't designed for massive damage or tanking can feel very underpowered against them.
Games may try to address this by giving special rules for monomonsters. For example, recent editions of D&D allow designated powerful monsters to take additional actions during the turn - perhaps one reaction for each PC present, or a secondary attack that happens at a different point in the initiative order. I've also seen auras used as part of this strategy, with PCs potentially affected each round by some harmful or penalising effect.
Big Sack of HP
A related issue is that where games use a hit point-type system, monomonsters have a different sort of power curve from fewer weaker opponents.
Let's say we have two combat encounters: one against ten ogres with 1n HP, and one against a giant with 10n HP. Let's say these are equally challenging in some game-mechanical sense.
In the first case, the ogres can make up to ten attacks each round. If the PCs can inflict 1n HP to an ogre, the number of attacks drops by 1/survivingogres. There is an interesting tradeoff of attack types: a fireball might inflict substantial damage to five ogres but kill none of them, for example. The PCs must decide whether it's worth spending a turn injuring several ogres so that they'll be easier to kill in future, or whether it's better to cast a single-target spell that will probably kill one ogre and reduce the incoming attacks. The fireball also tends towards averages (some ogres will probably be damaged) while a single-target spell might miss completely.
With classic D&D mechanics, the giant works very differently. It makes one attack per round. Okay, it might make five, or it might make a single multi-target attack... let's just say it makes Y attacks. It continues to make Y attacks until the PCs inflict 10n damage to it, at which point it makes no attacks because it's dead. The PCs do not have tactical decisions about concentration of fire, because they cannot split fire. They can choose to use their most powerful attacks, or they can be less effective for no reason - it's a simple calculation.
Yes, they could do things other than make damage-dealing attacks, but I'm only concerned with damage here.
Not only is the combat less interesting in that sense, but the PCs also have less control of the situation.
In the ogre example, the PCs can choose to whittle down numbers as soon as possible, or use tactics (like area attacks) that kill the ogres faster overall but allow more incoming attacks in the short term. They can also do things like try to distract, bottleneck or otherwise impair the ogres' ability to bring all their numbers to bear. It's within the PCs' (and players') ability to try and shift the odds in their favour, rather than simply attacking whatever's closest.
In the giant example, the PCs can't use these tactics. They can't choose to focus on reducing the number of incoming attacks right now versus winning the combat efficiently. They can't necessarily choose to eliminate the threat to the squishy wizard first - it might be possible to distract the giant from attacking the wizard, but perhaps its attacks affect the whole party. They can't divert some proportion of the giant's HP and an associated slice of its attack effectiveness away from the combat for a time.
Even in Call of Cthulhu, which has a very basic system designed to model plausible events believably, there's an extent to which combat becomes a resource management exercise, which makes direct confrontation with a monster very tricky, because it often does come down to "can the party's 5 actions a round attached to vulnerable bags of hit points overcome the monster's 1-2 actions a round attached to a much less vulnerable bag of hit points."
Games do try to address these things. I've seen 4E D&D solo monsters built with several "forms", so the monster changes its behaviour and power based on injuries. Some game systems do allow specific injuries, which can reduce the power of the monomonster's attacks in a vaguely similar way to killing off some of a monster group. GMs can also use straight-up roleplaying to have monsters become wary, frightened or carelessly enraged by injuries.
I've Been Saving That
The last point I think I want to make about monomonsters is that they create an extremely strong incentive for the use of one-shot nova abilities, because killing or hampering 100% of your foes is generally the best outcome you could hope for.
I think this may be more of a problem in games which make less of an attempt to offer tactical combat, simply because these games are more likely to include flavourful abilities that can be devastatingly effective but are extremely limited in their use. This may be a deliberate decision (just once in a while, or even just once, the character can take someone down hard), or simply an unplanned organic result of the interaction of abilities, or careless writing.
The main contenders here are, I think, of two kinds. There are danger-balanced abilities, which are extremely powerful but balanced by a very high risk, so that it's foolish to use them often. There are also limited use abilities, which you just cannot use often.
The easiest example of a danger-balanced ability is psychic powers in Warhammer 40,000. It is (okay, in theory) extremely dangerous to use psychic powers, especially if you do so at full power, and so it's best to save them for extreme situations. However, they can be devastatingly powerful. So, if you can invoke a psychic power to pretty much guarantee defeating one target, but it's very dangerous, it is well worth doing so against a powerful monomonster.
The classic limited use ability is actually equipment. A lot of games allow you to occasionally pick up very powerful single-use items. Those such as vortex grenades, arrows of X slaying, scrolls of celestial annihilation or blade oil of banishment may give you a very strong chance of destroying one target outright or at least inflicting massive damage. Another category is single-use buffs and debuffs: you can become invulnerable for ten minutes, or reduce one target's strength to that of a quail, or turn the whole party into frost giants, or trap a target in temporal stasis while you pile up vast quantities of explosives.
There can also be abilities or effects that aren't intended to be near-flawless kills, but can be used that way because of the way they're designed. If players routinely try to pull these off GMs tend to either houserule or find ways around it, but it's often when faced with a desperate situation that the player first tries to pull it off. For example, there are several powers in various White Wolf games that are vastly more powerful than the designers seem to have intended. A GM will not necessarily have a comprehensive strategy for implementing these in a balanced way, especially if the PC first comes up with the idea during a boss fight.
Unfortunately, if you have a much-foreshadowed climactic encounter with the elder dragon Droom, what usually doesn't happen is the party valiantly battling it for several rounds, both sides escalating their attacks dramatically until a final strike with the Arrow of Dragon Slaying brings down Droom. Players are usually pretty canny, and it's obvious that the sensible thing to do is open up with your most powerful attacks in the hope of ending the combat as efficiently as possible. So what you actually get is the much-foreshadowed climactic encounter with the elder dragon Droom in which the hunter looses off the Arrow of Dragon Slaying the very second Droom swoops into view, killing him instantly. Droom ends up being all talk and no action; the band of nameless goblins they encountered in the forest played a bigger part in the story and came far closer to defeating the heroes.
There's nothing wrong with using these items like this, and it can feel narratively satisfying - the natural point for the powerful item to be pulled out just as it's most needed. Similarly, a hero calling on their most dangerous powers at the moment of greatest peril is very fitting. However, there's also a danger of such items or abilities transforming a climactic and dramatic occasion into a damp splat. It's particularly important to bear this kind of thing in mind if such one-shot abilities crop up moderately often, since you can end up with players routinely storing them up to win easy victories over powerful monomonsters, while what should be lesser battles against multiple opponents are relatively difficult. That's not bad playing - it's very sensible, tactical playing - but it can still become unsatisfying for everyone.
Ths isn't a big problem, it's just something a GM needs to be aware of. It's potentially better to throw out battles with several potent enemies, or one quite powerful enemy with some moderately powerful allies, so that these abilities can turn the battle around without ending them in the first round.
As usual, a load of rambling with not much in the way of conclusions, just some things to think about. I don't say GMs should avoid using monomonsters or anything like that, it's just that there are quite a few aspects of their interaction with mechanics, narrative and (in particular) the actions of the PCs which it's worth thinking about when planning and running a scenario.