Some Thoughts On Blood On The Clocktower
In so many words, it's my favorite board game of all time.
If you have talked to me for any real extent of time during the last year or so, you know that I have fallen madly in love with Blood on the Clocktower, a social deduction game published by the Pandemonium Institute. In so few words, it is a brilliant exemplar of exactly what social deduction so often struggles to be; if you take nothing else from this essay, it is that you should find a local storyteller, or hop onto a discord server, and play a game as soon as possible. It might not change your life, but it will give you an appreciation for social deduction that nothing else can.
But what makes it quite so good? There is no single thing, and that’s what makes it quite so excellent. A few years back, Shut Up & Sit Down’s review of Clocktower was remarkably controversial because they were so effusively positive about a game which was, functionally, a reskinned version of Werewolf or Mafia. And in some sense, that’s true. It is very much a game which comes directly from that legacy, but the thousand tiny improvements are what make the game excellent.
Anyway. Here are my thoughts about Blood on the Clocktower, as effusively positive as they are of course going to be.
Perhaps the most immediately obvious improvement that Clocktower makes over traditional Mafia or Werewolf is misinformation. A very common pitch for Clocktower is, quite simply, “Mafia, but the narrator can sometimes lie to you,” and the reason it’s such a good pitch is because it’s a really good way to solve several traditional problems with social deduction games. If your information unequivocally frames a player as evil, but every ounce of your social instinct is telling you that they are good; well, maybe your information is just wrong.
More casual reviews of the game tend to end the discussion of drunkenness and poisoning there. If the game is unsolvable, they say, it lends the social side of the game an unprecedented depth in the social deduction space. However, it opens Clocktower up to a common criticism from less experienced players (or those who have never played the game) — if information cannot be trusted, how can you make any meaningful deduction? After all, that’s literally half of the name of the genre. The answer is that, while misinformation is potent, it is always finite.
Take Trouble Brewing, the simplest script, which has exactly four sources of misinformation, all of which have their distinct tells.
The Drunk. Perhaps the most iconic role in Clocktower, the Drunk is an outsider who thinks they are a townsfolk, but simply isn’t. The Drunk is the subtlest source of misinformation in Trouble Brewing, only generally detectable by a small subset of roles. However, it has a very public tell— it is an outsider which doesn’t know it. If the publicly-known outsider count doesn’t quite line up, then there is a Drunk out there somewhere, and if you can track down where it is, that is very strong information.
The Recluse. The Recluse is another outsider which is a potent source of misinformation— any ability which tries to generate information off of you might register you as evil, or a minion, or the Demon. This misinformation is likely the most obvious source of misinformation out of the four (since the Recluse will almost certainly come out publicly by the end of the game), but the misinformation it provides is both highly specific and highly bluffable. An Undertaker saw you as the Poisoner? Well of course they did, I’m just the Recluse registering as a minion!
The Poisoner. One of the nastiest minions in all of Clocktower, the Poisoner chooses a player each night to receive misinformation and to have their ability misfire. While this trail of poison is about as difficult to locate as the Drunk, if not harder, you know that at most, one player can be poisoned each night. Further, the Poisoner can be avoided with smart social plays and clever bluffs— even if you are the Undertaker, what minion is going to poison the Saint, whose ability loses them the game if they are executed?
The Spy. As the name implies, the Spy has the ability to perfectly blend in with the rest of the town. Whereas the Recluse is a good player who may register as evil, the Spy registers as good, despite their malicious intentions. When first encountered by new players, the Spy’s ability (which also includes seeing the Grimoire, which contains all information about the game) sounds absolutely devastating. But as you become more acquainted with the game, the tells of a Spy game become more obvious. Have the kills this game been in a near-perfect order? Then maybe there’s a Spy in this game, and that player whose information points to a widely-distrusted player as good might just be pinging off of them.
While these four possibilities are all significant, and often impossible to completely solve around, it is still possible to make sophisticated and mechanically-based deductions while still keeping each of the four of them in mind.
More sophisticated scripts than Trouble Brewing have far, far more than this. Often, figuring out just who is receiving misinformation, and figuring out exactly why, is the key to solving the game.
Another excellent change that Clocktower makes is how voting and execution works. As a lifelong voting theory nerd: from an objective, mathematical standpoint, Clocktower’s voting system is a ridiculous farce. For those unaware, to hold a vote in Clocktower, the Storyteller stands in the middle of the circle of players, and spins around slowly. If your hand is up when they point at you, you have voted on that execution; if at least half the living players voted, that player is “about to die”, and can only be saved if more players vote on a different execution that day, though a tie removes both players from consideration for execution.
What makes it so terrible as a general system (requiring quick reflexes, needing unusual numbers of votes required to push a vote through, making the system dependent on who you are next to) is exactly what makes it so good for a social deduction game in specific. Did that player vote for that execution because they wanted to be the final vote? Because they wanted to express public support for their death? Because they screwed up the count in their head, and thought there was actually one fewer vote? They might claim it is any one of these, and it’s impossible to determine which one is actually the truth.
Further, restricting nominations such that each living player can only nominate once per day, and each player can only be nominated once per day gives nominations some real weight. There’s a real mechanical cost (albeit a small one) to putting a player up for death, and on the final days of the game, that cost is far higher than it might seem.
In many more traditional social deduction games, votes are extremely simplistic, and tend to be either private (which means an entire element of the game that you can deduce nothing from), or overwhelmingly public (which means that you certainly can make deductions, but votes tend to be so quick that it’s very difficult to). Clocktower’s elegant and bizarre system massively improves on other systems by throwing away any semblance of neutrality.
I’ve written before about how interesting I find bluffing in Clocktower to be in June 9’s piece, so I’ll restate them and expound further upon them here.
The fundamental problem with bluffing in most social deduction games is that good players tend not to have any incentive to lie, while evil players have incentive to do nothing but lie. If you are a good player, and another player examines your allegiance, and accuses you of being evil, then you know for a fact that they are evil since no good player would ever accuse you of such a thing. This sort of structure makes more elaborate bluffs impossible, too: double and triple bluffs all just collapse into weird, evil-sided plays.
Clocktower fixes this in a number of ways. Drunkenness and poisoning provide potent sources of misinformation, so a good player might actually have information that points to you as evil. Characters like the Ravenkeeper, which can learn any player’s character if killed by the Demon, give good players self-evident reasons to lie, requiring them to draw the Demon’s ire in order to gain information. Double bluffs and triple bluffs can abound, too: is the player openly claiming Soldier, the character which cannot be killed by the Demon, telling the truth? Or are they lying so the Demon will not target them in the night? Or are they telling the truth, so that the Demon will think they are lying?
Another strange vector that bluffs tend to emerge from is the bluffs given to the Demon: at the start of the game, the Demon is given three characters that they know are out of play, and thus safe to bluff as. To explain why this allows an unusual structure of bluffs, imagine you’ve just drawn the Undertaker, the character which learns the roles of those that the town chooses to execute. As soon as the first day begins, a player across the circle loudly proclaims that they are the Undertaker, and wants people not to claim to them so that they can prove themselves to be the Undertaker tomorrow.
In many traditional social deduction games, your first choice of action would likely be to loudly proclaim that no, you are the Undertaker, and get in a game-long protracted fight with this other player. But in Clocktower, there are a number of different approaches to this situation. You could do that loud proclamation, but that risks the Demon learning who the Undertaker really is, and if the other player is the Ravenkeeper or Soldier, it might break the bold bluff they were trying this game. You could approach that player privately, to see if they’ll back down when they learn an Undertaker is actually in play: this helps earn early-game trust, but if that player was a minion without a bluff, you may have just given them critical information.
One of my favorite approaches, though, is to simply stay quiet. See what the other Undertaker does. Do they back down after a conversation with another player? Maybe that other player is the Demon, who just fed their minion an out-of-play role. Do they stick with it, even once the game has started to settle down, and most information is public? Perhaps you’ve found a minion, who believed their chosen bluff to be out-of-play, and has just realized that they’re in a tight spot. Do they back down the next day, or maybe later that same day? Maybe that’s just the Demon with a Ravenkeeper bluff, who wanted to draw some suspicion but realized they were taking on too much heat and needed to get out of there. Hell, maybe it’s a Spy, who knows you’re the Undertaker, but wants to undermine your information by making one of you look extremely evil. Deliberate double claims can be a clever evil move to make both of your information impossible to believe.
I’m not trying to say that any of these strategies are the best, or even that any of them are necessarily better than the others. What interests me is that all of these strategies naturally emerge from the systems of Clocktower, and each one of them can be suitable or unsuitable in all sorts of different scenarios.
A fundamental difficulty in designing social deduction games is player counts. The minority evil team clearly needs to grow larger as the game grows larger, or else they’ll be on unfair footing — but how much larger should they be? If you add another evil player every, say, three additional players, what do you when there is only one or two additional players? They presumably have to be good, but that skews the game back towards good’s favor.
Clocktower solves this problem with outsiders — players playing for the good team, but whose abilities harm the good team. These negative abilities range from the Saint, who loses the game for their team if they are executed; to the Tinker, who can simply die at any time; to the Mutant, who will be summarily executed if they even claim to be an outsider. Further, since the basic outsider count is always known, evil abilities can manipulate it to devastate the town: the Baron, for instance, simply replaces two powerful townsfolk with less-useful outsiders.
Outsiders are a clever solution to an annoying problem, but more than that, they still manage to be fun to play. When you play as a townsfolk, your game is all about maximizing your ability. You want to find ways to gain the most information, or prevent the most death, or to prove yourself publicly. When you play as an outsider, though, you are trying to minimize the impact of your ability. You’re the Butler, who can only vote as long as another player that you choose votes? Work around that by choosing a player who has been proven good. You’re the Saint, who loses the game if executed? Hide yourself, claiming something more powerful, hoping the Demon kills you in the night and your ability is neutralized.
Another subtle thing that Clocktower does to improve the social deduction experience is to not only allow, but encourage private conversations between players. If you need to tell a specific player something, you don’t need to shout across half a dozen other players to get them on board. You can just pull them to the side, so nobody else can overhear what you have to say to each other.
On the evil side, this allows far more complex planning and bluffs than you can traditionally pull off in games like Secret Hitler or the Resistance. On the good side, this allows you to form much smaller chains of trust, letting just a select few people in on your secret information. On both sides, it just lets the game feel a bit more relaxed. Clocktower rarely dissolves into a meaningless shouting match, since you can just talk quietly to another player, trying to solve the game or to get a social reading off of them, rather than have to speak over the entire group at once.
This one’s a fairly minor point, but it’s one that holds fairly strong for a social deduction game. Imagine it is your first game of Clocktower: you have no idea what the hell is going on, besides what you can half-remember of the Storyteller’s starting spiel. You draw the Empath token, which learns how many of its living neighbors are evil each night. During that first night, you are told that neither of them are evil.
Not only is this immediately actionable information (like most other first night information that characters will acquire), but it is in fact one better. You can act on this information immediately, forming an early trust chain, just by talking to the people who you are already sitting next to. You don’t have to stand up, you don’t have to shout across the circle, you can just sit back, relaxing and whispering to the people closest to you, who you already have good reason to trust.
Of course, not all characters have a mechanical hook with where people are sat. But all those that do give you a unique way of looking at the people around you.
You’re a Chef, who knows that a pair of evil players is sitting next to each other, somewhere? You can hone in on that pair of players that is constantly whispering to each other, even though their public information doesn’t really seem to make them trust one another.
You’re the Tea Lady, whose neighbors can’t die if they are good? Then maybe you arrange one of their executions, and once they survive, you can trust both of them, and they can likely trust you.
You’re the No Dashii, a demon whose toxic tentacles poison the townsfolk on either side of you? You can lean into their information, giving them reason to trust it and trust you, even as they are drip-fed nonsense and lies.
Every character with some sort of neighbor-related ability gives you one of these hooks to play with, and each of them can be played with in a dozen unique and interesting ways, letting you trust or distrust the people closest to you for any number of different reasons.
One more facet of Clocktower that is oft overlooked by those not as ingrained into the community is its sheer replayability. The three base editions that come in the box are each phenomenal, and can be played dozens, maybe even hundreds of times before they start to get stale. The sheer amount of dynamism and chaos which emerges from every single storyteller-chosen setup makes every game feel fundamentally different, and it’s difficult to become bored with the myriad options at play.
But if you do become bored of them, or simply want to expand into different conceptual territory, the huge swathe of experimental characters and custom-built scripts allow you to explore whole new realms of possibility. Endless recombination and variety allows you to understand the deeper logical systems and mechanical interactions that every individual character provides, and gives you a different perspective into the game as a whole.
The overall point, then, is that Clocktower is a brilliant system for the characters and mechanics it contains; it is more than just the sum of its parts, it is the connections between those parts that really makes it tick.
I’ll open this final portion by saying that I haven’t storytold all that much (or at least not as much as I would like to); around a dozen or two games of in-person Trouble Brewing.
Every one of those games, though, has felt entirely unique. That is in no small part due to the random chance of who draws what role, and due to the different players interpreting their roles in a fascinating range of different ways. But from behind the Grimoire, it feels like every decision those players make, every execution, every kill, every panicked vote and Slayer shot; every one of them feels like a consequence of your decisions, the final moment of a Rube Goldberg machine you set up the moment you chose what characters you put in the bag. It’s a fascinating experience, one where you are at once referee, narrator, dungeon master, and player.
Even on Trouble Brewing, a script with a reasonably limited amount of storyteller intervention, those moments of choice feel interesting and powerful in your hands: do you show the poisoned Ravenkeeper that a certain player is the Demon, or just a minion? Do you let the Mayor bounce a kill onto the spent Slayer, or onto the Fortune Teller who is coming dangerously close to solving the game?
The Clocktower rulebook encourages storytellers to generally help the losing side, and that the evil team starts losing. This is, broadly, true; but finding when exactly the evil team is no longer losing is incredibly difficult to figure out, and making that deduction is a curious kind of puzzle that I’ve never experienced in any other game. The best storytellers have a knife’s-edge sense of balance, capable of solving the puzzles before the players do, and giving them the exact right sort of misinformation to throw a wrench into that machine. And that’s just one more brilliant thing about storytelling: you can feel like you are getting better at it, as you set up games that work, and games that don’t.
Shut Up and Sit Down’s review of Blood on the Clocktower, as mentioned earlier, is one of the most controversial reviews they’ve ever put out because Quinns found it to be his favorite game of all time. Getting my hands on a copy and falling head-first into the astonishingly deep community has put me into the same situation— this reskin of Mafia is quite handily my favorite board game ever made.
Perhaps the one reason above all others that Clocktower succeeds over Mafia, over Werewolf, over Secret Hitler or 2 Rooms and a Boom or the Resistance or Quest or any other social deduction game ever made, is that it throws away all semblance of neutrality, in favor of pure fun. Clocktower is an unfair game, rigged against both teams by the omnipresent and both-sided Storyteller. You might receive only one piece of information all game, and that information might be false, and that false information might be what loses you the game. Or maybe it will be the opposite, and the Storyteller will try their hardest to give you the best information that they can, but you will simply dismiss it out of hand, convincing yourself that you are the Drunk, even if every piece of information was accurate.
But never before has having the game rigged against you felt so dynamic, so interesting, so empowering. And that, in so many words, is why I love Clocktower.