P1 Rumors — Team 19

If players roll an even number, they can gossip with another employee and receive a truthful hint about said employee’s relationships. 

Players who do not have a love/hate relationship with each other can still interact in the game to gain information about the broader social network. 

• Moderator and players laughed at the silly rumors about relationships within our group of acquaintances and teammates.  
• People were spreading both true and false information.

• Rumor formats were nonstandard so some rumors communicated more information than others while others were confusing.
• Because there were limited point scoring opportunities, players rushed to make guesses rather than spread rumors.
• It was hard to fit guesses on the board and confusing who had made each guess.
• People were more inclined to “guess” their own relationships than others’ relationships. If all players put their own relationships on the board, the game ends. That was not what we designed the game for.
• There was not enough dynamics with only three players, because each player ended up being connected to every other player (loving one and hating another).

• Instead of having a public board where people take up slots to make guesses, we ask each player to guess the entire network. 
• Rumors should be standardized in the format “X loves/hates Y”. On the back of each rumor card, there are “from” and “to” fields for players to fill out, so that everyone can keep track.
• 

• We should have at least four players in a game, possibly between 4 and 10 players. For n players, each player needs to figure out 2(n-1) relationships. At the start of the game, players have no information about others’ relationships. Even when they receive a rumor, they cannot tell whether the sender shared it to help them or to mislead them. To address this, we introduced an extra rule: once per game, a player may choose not to spread a rumor and instead verify one relationship with the moderator. 
• Since there are only four of us on the team and an additional moderator is required to playtest, we created a simple computer program to serve as the moderator in case there are not enough human players.
• 

• Players were pretty engaged.
• Players naturally developed strategies (e.g., helping the person they loved make good guesses, lying to other players).
• Players made subtle signals to communicate. 

• Players were still unclear what should be passed; some broke the rumor format.
• Individual rumor-passing turns felt slow and unengaging.
• Players didn’t have enough time to process information at the end of each round.

• Discussion should be off while passing rumors. 
• Each player writes their rumor down, then they all pass them at the same time.
• There should be an explicit open discussion period. 
• Players need notecards to write down their thoughts.

• Players quickly understood the rules after initial clarification.
• Most players developed clear strategies by the second round and passed rumors with intent.
• Players did use the confirmation mechanic.
• Players provided positive feedback about the game’s overall enjoyment.

• Because of the simultaneous rumor passing, players even missed rumors passed to them or misunderstood who passed to whom. 
• There was no clear distinction between rounds, leading to players advancing at different paces.
• Point tallying at the end was time-consuming and felt mechanical. 

• The original model of passing rumors in sequence rather than all at once was better for reduced chaos and allowing others to better observe the flow of information.
• Structured discussion periods should be enforced so that more information is shared and there are more social interactions.
• The true information should be revealed relationship by relationship at the end to create a dramatic reveal.

• Players who had previous experience with the game were able to strategize from the beginning.
• The new relationship matrix made it easier for the moderator to check relationships.

• Players who didn’t know each other had a hard time remembering others’ names.
• Players were quiet and introverted, leading to very little discussion or social interaction.
• One player who was neither loved or hated felt isolated from the game.
• A player misunderstood how to use the new matrix, confusing rows and columns.
• The new forwarding rumor mechanic confused players, possibly complicating the dynamics even more.
• Five players allowed for too many possibilities, making the game more difficult.
• Not all players used the confirmation mechanic, which was meant to help them verify who was trustworthy and who was not. 
• Players in the first round didn’t have any outside information to spread, except by revealing their own relationships to the person they love, making it obvious to all players.

• Printed name tags with player color and number are necessary.
• We should limit the game to 4 players plus 1 moderator for now to reduce information overload.
• We are getting rid of the forwarding mechanic; if players want to forward a rumor, they can replicate the content and send it themselves.
• We should use the old list of relationship facts instead of the matrix to minimize effort for players.
• The moderator should pick the relationships to ensure no one is left unrelated to anyone (up until this point, players had decided their relationships).
• We decided to add the office premise more fully immerse players: The boss needs to select a new HR director from among the employees and decides to test their social deduction skills through this game!
• The moderator should leak some information at the beginning of the game so that players in the first round have material to bluff about. We designed a new mechanic where a “lucky employee” is secretly selected and given an extra piece of information, allowing players to pretend to be the lucky employee and spread false information.
• There needs to be more definitive information circulating. We introduced dice rolling: players have a chance to receive a hint in the form of “X loves/hates Y or Z”, but there is also a risk that they must reveal a hint about themselves.

• Players stayed focused and were actively thinking. No player was left isolated.
• Players consistently rolled the dice every time, showing engagement with the new hint mechanic.
• Players actively took notes and often drew graphs when provided with paper.

• Inventing clues was nontrivial and arbitrary; it was hard for the moderator to keep track of the hints given or to avoid redundant information.
• One player thought it was unclear that each player has one confirmation per game.
• There was still little talking among players.
• Players believed there might be too much information (or too many rounds), and the game felt quite slow.
• Players felt like they had no incentive to speak during discussion because they might get one point if they convince the person they love of something, but that person will get two points

• We added a timer to speed up turns.
• The confirmation mechanic was removed, since we already have the hint mechanic through dice rolling.
• We should let the player choose whose hint they want. This makes the moderator’s job easier and allows players to get more useful information.
• For four players, we should limit the number of rounds to four to reduce information overload.
• We should enforce that rumors are read aloud during the discussion period at the end of each round to prompt conversation.
• We need to rebalance the point system to provide stronger incentives for players to share information.

• The redesigned point system worked well. Players had equal motivation to help the person they loved and to confuse the person they hated. 
• The restructured dice rolling mechanic also worked well. Players chose to roll the die in some rounds and not in others, showing careful and thoughtful decision-making.
• The new hint format was useful for the player who rolled. They could obtain valuable information about a specific player if they were willing to take the risk. 
• The rumors-reading was key; the discussion period was full of laughter, bluffing, and accusations.

• It was hard for players to remember who sent each rumor card.

• We need to use the colored version for the backs of the rumor cards to differentiate between players, making it easier for others to observe. 

• Having employees (players) and the boss (moderator) read separate rule sets helped them understand their roles better.
• Players showed strategic thinking by deciding whether to roll for hints and by intentionally sending misinformation.
• Players engaged in social deduction activities, such as being suspicious of the “lucky employee” claim.
• Moderator responsibilities became easier and more streamlined as the game progressed.
• Players actively tracked their points, showing self-management without needing heavy moderator intervention.

• Players often found the beginning of the game overwhelming due to the volume of auditory information.

• We should make rumor cards larger to make them easier to read and manage.
• In the rule sets, we should clarify upfront whether the lucky employee is required to reveal their status or can lie about it as part of strategy.
• Players should be encouraged to announce rolls verbally to make round structure clearer.