Martian Massacre
By Claire Tang, Carl Liu, Xatrik Bera
Artist Statement
A research base has been infiltrated by aliens! Among the humans in the crew, there is an alien that is capable of infecting humans and turning them into its allies. Martian Massacre is a social deduction game inspired by the classic sci-fi horror movie The Thing (1982) where your trust will constantly be tested and the decisions you make will determine if humans will survive or go down with aliens. Humans will work together to identify the original alien before the infection spreads too far while the alien team attempts to spread and eliminate humans in the base. In addition to discussing with the others on the base, you must manage limited resources and use your abilities wisely
Martian Massacre is designed for anyone who enjoys bluffing and strategy. You must adapt to new information and teams as they shift and act carefully. Join the research station and see if you can survive until the end!
…At the table everyone can hear you scream
Concept Map
Concept map created on Canva.
Initial Decisions about Formal Elements
Objective: Humans and aliens have different objectives. Humans want to identify and eliminate the original alien before the aliens take over by infecting everyone. Aliens want to eliminate all the humans.
Players: This game is designed for 6-14 players and one moderator, starting with 1-3 aliens depending on the size of the group. We want a moderately large group for exciting discussions and reasonable balance between the two teams based on our design of the game.
Procedures: The game uses day and night cycles similar to many social deduction games such as Mafia. In the daytime, players have the ability to discuss and vote people out. In the nighttime, the Alien has the ability to infect another player who joins the alien team.
Conflict: Each person has a hidden identity and missing information. Each player chooses how to communicate to accomplish their objective.
Testing and Iteration History
Our game has evolved substantially through testing iterations. We initially started with a simpler game directly built off Mafia, where the game was simply humans against a team of Aliens spreading through a nightly infection process. We realized that this design did not allow for the human team to gain any real information thus was not enjoyable to play since it would not feel like the human team would have any real strategy, thus adding a “scientist” role that would be able to test a player every round.
Playtest 1: 10 players in class on 7/2
This game played much like a traditional, unmodified Mafia, which generally has the concern from the humans of being difficult or boring to play since most of the time for a human is spent having no useful information and nothing to do. Accusations in this version were based on no real information with humans just ultimately deciding that action needed to be taken and making very random guesses. It also seemed that the game was very biased towards the Aliens, since they were essentially a spreading “mafia” role that was inserted into a basic game with already very little information. In this playtest, the Aliens just slowly won by spreading every round while the humans were simply guessing and waiting. The outcome of the game was pretty random since it just depended on if the humans could correctly guess the Alien before they gained the majority.
Iteration 1: Adding Action Cards
We found our first draft of the game to be plain and too close to just a Mafia mod, except with the spread of aliens, making the alien team especially powerful. This motivated the implementation of action cards which would give all players something to actually do and more information and complexity in circulation each round. We drafted some ideas of separate actions for humans and aliens that could help them accomplish their objective.
We designed sample cards with the use of AI. Attached is a picture of cards that we printed physically for our playtests of actions. (The card ‘Sleep’ was renamed ‘Stasis’)
Playtest 2: 7 players in class on July 9
Our first playtest put the full ruleset in front of seven players who had not seen the game before.
The parts of the design we were most unsure about turned out to be the parts that landed. Players consistently named the abilities and the fluid team membership as the most interesting things in the game. The abilities gave the table something concrete to argue about instead of pure vibe reading, which is the usual failure mode of Mafia with new players, and several playtesters said it made the game feel less like guessing. The token economy also produced real spending decisions. That told us the core loop was worth keeping and that our problems were in the framing around it, not the pieces themselves.
The win condition broke completely. Because becoming an alien counted as a win, no human had any reason to resist conversion. Sitting still and waiting to be turned was the dominant strategy, everyone ended up on the winning side, the game had no fail state and tension. This was the highest priority fix coming out of the session.
The second problem was that point values on the ability cards didn’t match the values in the rule sheet, so the moderator had to make rules live and playtesters read the inconsistency. The moderator script was also under specified, it didn’t say clearly when the players open or close their eyes, or whether a given instruction was directed at one alien, the original alien, or every alien. None of this is a design insight, but it cost us table time and made it harder to evaluate the mechanics we actually came to test.
We made four changes in response. We cut the researcher role, which was not doing enough work to justify the extra night phase step it required. We added a hard round limit so that the game has a clock and a real losing condition, which gives the humans a reason to act instead of stall. We restructured the win condition so that conversion carries a cost instead of being a free upgrade. And we made the cards the single source of truth for all numbers and regenerated the rule sheet from them, then rewrote the moderator script as something a person outside the team could read aloud cold.
Iteration 2: Slowing Down Aliens
An easy fix to the imbalance was to make infection cost tokens, thus drastically slowing down the Alien team. Feedback from the teaching staff indicated that it seemed desirable as a human to join the Alien team since the game seemed to heavily favor the Aliens. We decided to make infection an action card rather than a default nightly action. We hoped that this edit would give humans an incentive to actually stay as a human and work together to find the Alien by giving them a good shot to win by playing strategically and using deduction skills.
Playtest 3&4: 8 players out of class on July 11 (Recording included for playtest 4):
We played two rounds of the game with a group outside of class. With the update of having infect as an action, the game did feel smoother and not immediately biased towards the Alien team. The game played out to the end and did not seem overly short or unfair. We had initially set the cost of the infection to 2 tokens, which proved to be a bad choice since it is more advantageous to infect than to murder, which was feedback we received from the Alien in this round. We additionally received feedback from the humans that the blood test was too cheap thus too powerful. We also received an idea to change the function of the sabotage action, which we thought was interesting thus implemented as suggested by our playtester. The players all gave feedback about enjoying the actions in the game, particularly the murder action. We adjusted details about the costs of actions and implementation of them based on direct feedback of this play and played another round with the same group.
The second round with this group, which is the recording attached, went smoother as the group was already familiar with the setup. There were no obvious problems or disadvantages with the gameplay during this test. There were lively discussions again and some more concrete feedback about certain actions that felt either too weak or too powerful. Specific feedback included decreasing the power of the sleep card since if the tokens were doubled in the day, that would make the card extremely powerful, adding more repair cards total to the deck since we only had one in circulation in the current version, limiting the Aliens to only play up to one of each action per night (ie can’t do two infections even if they have the tokens for it), and making the blood test results private. We also additionally got a suggestion for the token count to be made private if possible.
Notably, humans won in both rounds of this game, suggesting that our edits had at least somewhat helped balance out both sides although it may not be perfect still.
Iteration 3: Adjusting Action Cards
This playtest group gave fairly specific feedback about actions based on how powerful/useful they perceived them to be during their play, as well as specific feedback about parts of the rules or the descriptions of actions were still confusing. We adjusted in direct response to their feedback between the two plays, and incorporated some of their final thoughts at the end of both plays into our final version as well.
Future Steps
Watching the game playout a few times was very helpful to our development process. The most difficult and unique part of the game to design were the action cards, including what they were, how much they should cost, and what the ratio of each card should be. We consistently got feedback from our players that this was a very enjoyable aspect of the game, but it was difficult as designers to finalize the variables listed above about the action cards even after our playtests since it would likely need more iterations for us to see how strategies can form and how players tend to use these cards on each team. Trying different values of those variables is our biggest tool in balancing the game based on just this structure, and we ideally would want there to be fair and engaging strategies that can form based on our design. We do not think we got enough iterations of feedback to necessarily arrive at this perfectly, and that would likely be a good future direction to develop a fully functional and enjoyable game. We would also preferably implement the idea of having hidden token counts in the future, which may be easier done through a digital component.
Final Decisions about Formal Elements and Values
Objective: Humans and aliens have different objectives. Humans want to identify and eliminate the original alien before the aliens take over by infecting everyone. Aliens want to eliminate all the humans.
Players: This game is designed for about 8 players and one moderator, starting with one alien among the group.
Procedures: The game uses day and night cycles similar to many social deduction games such as Mafia. In the daytime, players draw an action card such as Blood Test, Stasis, or Strangle, and use their tokens to perform them. Players also have the ability to discuss and vote people out each day. In the nighttime, Aliens can choose to perform 3 actions with the use of tokens accumulated: Murder, Infect, or Sabotage.
Resources: We introduce a token system with action cards. Players can earn tokens throughout the game and use them to perform action cards that they draw.
Conflict: Each person has a hidden identity and missing information. Each human decides how they want to use their resources and who to trust and each alien decides how to manipulate others to stay unsuspected and expand their team.
Values stayed the same.
Final Components of Game
Rules:
Setting
You are in a base on Mars, there are humans, and there is an alien.
Human: Work together to identify and eliminate all hidden aliens before the station falls.
Alien: Pretend to be human, gather tokens, sabotage the station, and infect the remaining crew.
Players & Setup
Best with ~8 players, plus one Moderator who doesn’t play
Deal roles secretly (cards, slips of paper, or an app): one Alien, everyone else is Human (Crew).
No one reveals their role at setup. Only the Moderator knows everything.
Roles
Alien: Blend in with humans and eliminate Humans. Aliens gain tokens each day and may play Alien ability cards. Only the original Alien may play Infect cards during the night. The alien is given 2 extra tokens at the beginning of the game
Human: Work together to identify and eliminate the original Alien. Humans gain tokens each day and may play Human ability cards to investigate, protect the station, and eliminate suspected aliens.
Win Conditions
Each player’s win conditions will be based on the role that they are at the end of the game.
Humans win when the original alien is dead.
Aliens win when all humans are dead.
Gameplay:
The moderator will pass out cards to assign roles. If your card says “Alien”, you are the original Alien. Otherwise, you are a human.
Day:
Each player gets 1 token at the start of each day.
Each player is dealt one card at the beginning of the day, then each person plays as many cards as their tokens permit, clockwise starting from the moderators left.
Human Cards:
Eat food: gain an extra 1 token
Stasis: gain 2 tokens but lose the ability to vote in the next round
Blood Test (costs 2 tokens): test one person to check if they are an alien
Repair (costs 2 tokens): fix a sabotage
Strangle (costs 4 tokens): kill one person
Votes are cast by awake players after cards are played, votes may be skipped and players in stasis may not vote.
Night:
Humans close their eyes
Alien cards are laid out on the table and are bought by tokens without discarding the card.
At night the aliens are awakened and may play one of each action card as their tokens permit.
Alien Cards:
Murder (costs 2 tokens): kill a player.
Infect (costs 3 tokens): convert one human into an alien beginning the following night can only be done by the original alien.
Sabotage Greenhouse (costs 5 tokens): infect one more human per night. Everyone gets one less token.
Events:
A die is cast at the start of every day:
If a 1-4 is rolled the day is normal. If a 5 is rolled a supply drop takes place, and if a 6 is rolled a dust storm takes place.
Supply drop: everyone gets their tokens doubled.
Dust storm: No tokens are dealt at the beginning of the game.
Components List:
Identity cards:
10 Human Cards
3 Alien Cards
Human Action Cards:
12 Food cards
6 Stasis cards
6 Murder cards
4 Repair cards
Alien Action Cards:
1 Murder Card
1 Sabotage Card
1 Infect Card
Event Cards:
1 Sandstorm Card
1 Supply Drop Card
Dice:
1 Six-Sided Die
Tokens:
120 Energy Tokens
Video of Final Playtest
video still uploading – this will be updated soon
5 timestamps:
1. First use of the murder card 5:40
Players responded really well to the murder card and found needlessly killing each other very funny, but I think the cost of the Murder card balanced out its randomness.
2. Players argue over who to bloodtest 6:20
One player tries to prove his innocence by asking for a blood test on himself, but another player agrees too eagerly and is blood tested instead. This moment shows how limited information and social dynamics and conversation influence decisions in the game, which make the dynamics exciting and not simply rule driven.
3. The second Alien is revealed through his hesitance to vote for a revealed Alien 6:50. A converted alien hesitates to vote against the revealed alien which unintentionally revealed an alliance. The players are able to use deduction to find the aliens.
4. Players try to use token costs to narrow down the alien suspects 9:50. The players try to use resources as information to narrow down who the alien could be, suggesting to us that the resources themselves are actually part of a possible deduction process and that humans are actively forming strategies as they continue to play the game more.
5. A player deduces and kills the original alien because he thinks there are too many aliens to conduct an unbiased vote 10:20, I think this shows the value of the Murder card. Players have begun to adapt their strategies as the game progresses, showing how our game has multiple ways to win as the players make decisions based on the current state of the game.
