The Godfather, Part 3 [Concluded]

[TinCow]

Excellent post, GH. I agree with everything you said. I noted Sasaki's play style very early in this game and have been trying to learn from it. It does indeed seem to be the most effective. I do see value in write-up analysis, and have yet to encounter a game in which it is totally useless. In games where the write-ups are written by the mafia, the style of the writing can be very telling. Though I failed to follow up on it, it's worth noting that I was accurate in my initial profiling of the person/people that created the first write-up. In games where the write-up is done by the host, there always seems to be a hint or a clue somewhere, even if it's just character continuity. The key, IMO, is to use this as a simple piece of evidence in the overall picture.

First, take a look at the game and figure out what kind of a game it is. It should be pretty easy to tell generally how much useful information will be in the write-ups. Games with clues will have obvious clues. Games with mafia-created write-ups will be well-known beforehand. Games with many detailed role descriptions written by the host will give clues as to which specific person killed which target. Games with none of these may have useful information, but it will be harder to find. Second, take a look at the evidence that is being produced from the write-ups. Strong arguments and weak arguments are easily recognizable simply by how much sense they make and how consistent they are with the situation. Finally, add these two factors together to determine the overall weight of the write-up evidence. Strong arguments in games that have high promise for useful write-ups should be given more weight. As the usefulness of the write-ups and the strength of the argument decreases, so should the weight given to any write-up analysis. In any case, even at maximum strength, I would never consider write-up analysis at anything more than 49% of the reason to vote for someone (unless there's literally nothing else to work with). Often 25% to 33% seems a more appropriate figure. Write-ups are best used as reasons to FoS people and put pressure on them. They should not be used exclusively to lynch.

On a personal level, I've found myself delving towards the write-up analysis simply because I can do it and because I'm still learning the 'trade' of mafia detection in other areas. I don't even count Capo as one of my games, because I really had no idea what I was doing and no coherent plan of any kind. Capo was my mafia 'tutorial' and after it was over, I was ready to play the real game. Of my next 4 games, I was mafia in 3 of them. As such, I feel like my efforts as a townie really began with Ephesus and Chicago and continued on into this game. It's easy for me to tell when I don't know what I'm doing, and I've found my own flaws are in my difficulty in IDing strange behavior. In the first two games, I thought this was because I simply did not know peoples' normal play style due to a simple lack of experience. In this game, Sasaki gave me an epiphany with his explanation that we needed to look for inconsistencies in posts. This is what I did when I built my case against shlin, and I was pleased with the product, even if the result was lyching an innocent. I will continue to do write-up analysis when I deem it worthwhile according to the above formula, but I am going to put more effort into Sasaki's proven (and common sense) methodology.

I think my biggest crutch is my posting style. I write in a very deliberate manner that people seem to listen to. While this is great if I'm mafia, it can be a huge hindrance when I'm a townie. It's no good for people to listen to me and vote based on my reasoning if my reasoning is wrong, and it often is. I actively need other people to point out flaws in my analysis, as that is the best way to make sure that any error I make does not damage the town's efforts. Even more importantly, I need to be careful that I do not mount major attacks designed to get a target lynched unless my evidence is very, very strong. Above all, I need to (and try to) remain flexible in my thinking. There is no shame in backing off of a case one has mounted if the counter-arguments are stronger than your own. The key is to get the end result right, not to out-argue the other players. There will always be more good ideas produced by others than you will produce yourself. These need to be identified and embraced.

Above all, I think the town needs to remember that it a team effort working on a large puzzle. No single theory is best, though some are stronger than others. Town efforts seem to work when the totality of the system comes together to produce a mosaic of information from which the various analysis styles can work in harmony. If done properly, the mafia will pop up in the sights of more than one of these styles of analysis. When this convergence occurs, the probability of a good lynch skyrockets. For this reason, townie cooperation and participation is the key to victory, even amongst the dead. The more people who are legitimately trying to analyze the game and locate the mafia, the more likely the town will win. It is easier for the mafia to win when there are only a couple dominant voices because it is easy to assume that role and intentionally lead the town astray. It is far harder for the mafia to win when there are a multitude of strong town voices, none of which disappear after death.