TOWN:
- The practical win condition of town is to simplify the game into a state where it is impossible for the mafia team to be able to prevent themselves from all being eliminated. Nothing else truly matters for town except for reaching this state.
- This functionally means that it is not always correct to kill who you think is mafia during the day, which is increasingly more likely to be correct the earlier in the game it is.
- There is no practical downside to eliminating town early on in the game, especially if you're eliminating players that are more likely to be automatic eliminations later on in the game and/or major distractions. The first mafia death is the highest potential swing in momentum in mafia, but it has exponentially more value the later on in the game it is.
- It is far more important to remove obstacles for town from coming to the correct solve later in the game than to eliminate mafia. This gives the mafia fewer opportunities to manipulate the town into eliminating themselves later in the game after all the players that either had correct reads or were holding town together are no longer alive.
- The mafia rely on 'resources' in order to secure enough miselims to win the game. Whether it's players that are always going to be at risk of being removed by default, personality clashes that will become a major distraction, or really anything that indicates town will look to push on other town or will actively not work with each other. Mafia have to lean on their resources to further their win condition, so removing them will give them fewer options to work with.
- Removing these 'resources' by resolving the conflicts via removing them from the game or convincing them in some other way is the primary way to further your win condition. If you have scumreads that also does this, then you can effectively double dip on it.
- Cop roles are most often effective for clearing slots that are good default eliminations or targets that are likely going to be targets several days in the future. Vigilante roles are most often useful on targets that are likely to be eliminated the next day or the slots that are causing the most problems during the day. Trying to snipe mafia reads are rarely useful in a practical sense.
- Not using your limited abilities until well into the midgame is often correct, especially if you have an ability that doesn't lose value over time (like traffic analyst). Information swings are far more potent once the early game is over.
- Eliminating mafia day one is often a disadvantage to town, rather than an advantage. Any scumteam worth their salt will neutralize the numbers advantage unless there's a severe skill gap or town get incredibly lucky, town very commonly will get more complacent since they are winning, and the mafia will very often get more resources to work with than they otherwise would since town is very likely to get a lot more confident in incorrect reads.
MAFIA
- You only need a baseline of being able to think through plausible ways of looking at the game from a perspective of somebody who doesn't know anyone's alignment to be an acceptable mafia player. Being able to emulate how you sound and act as town is often enough to be a great mafia player if you know how to play around what town is doing.
- Your key to success is utilizing town's lack of information to push them towards believing that the wrong ideas and reads are correct.
- You rarely need to be elaborate with whatever you're doing. The simplest plans are often more than enough, you simply need to do the bare minimum and stick your neck out only when you either need to or when there's a distinct advantage to it.
- Don't utilize your biggest resources unless you either have to in order to win or you will gain a lot of ground with it. If there's players that are causing a lot of problems for town, they're far better to you alive than as a corpse, or if there's a massive tunnel war between two players happening.
- Individual reads on you rarely are much of a threat. You can be townread by virtually nobody for the entire game and still be fine, as long as you always have people below you on the list of people to have removed from the game.
- Generally, as long as you're able to avoid the town preventing each other from eliminating their incorrect scumreads as well as the game solving itself the moment one of you is removed from the game, you should reach the endgame with good odds of winning the game.
GENERAL THEORY AND HOW IT RELATES TO MY APPROACH TO MAFIA
- Most of the town games I consider my strongest ones are those games where I'm able to create enough of a swing in the game to where the rest of the town is able to solve the game on their own rather than just me having strong reads.
- A lot of the town stomps I see are simply through either town identifying the scumteam as their targets initially, not giving the scumteam enough resources to be able to work through the game (and therefore having no way to prevent themselves from being inevitably killed), or simply not allowing the scumteam to utilize their resources to get a foothold in the game (which is quite rare).
- The scumteam is often far more in tune with the general direction the game is going than town is, which is a significant reason why I primarily look to figure out what the mafia are likely doing.
- My general approach to most games is to attempt to look at the game and find the most important parts that are likely to cause either a response by mafia or require them to adjust what they're trying to do. Mapping what each shift does and the general approach scum would take that makes the most sense with that particular set of data points allows me to get a rough guess of what mafia is relying on in that particular gamestate. Eliminating the player that most accurately fits that model gives me the ability to test whether I'm on the right track or not, with a town flip often meaning I have incorrect reads on what the gamestate actually is.
- 90% of games, the only reads I'm particularly interested in being correct on are the wagonees. The majority of the rest of my scumreads are players I am least likely to gamestate read into being town later on in the game.
- My working theory on the pinnacle of town play is the ability to remove the mafia's ability to win the game very early on in the game regardless of your own reads or death.
- Read strength is an overrated skill. I have plenty of games where my reads are above average or better, but most of the games I consider my strongest are games where my reads are below average to average.
- Playing as mafia really is not hard, especially if there's ego in the game. Ego from town is the single easiest thing to exploit as mafia in the game.
I'll probably have more thoughts later, but this is a general gist of how I approach the game, what I believe are fundamental truths about mafia, and why I obsess over the gamestate in all my games.