So we decided cop doesn't get a report 25% of the time. Before any deaths, cop has a 31.247% chance of getting an innocent. After an innocent dies: 25.715%. After the miller dies: 34.285%. I don't think that ~26% of the time qualifies as unlikely. Sure, more often than not you get a guilty, but 26% isn't that uncommon. If you think 26% is uncommon (and absurd) then we just have a difference in our definition of "highly unlikely" that can't be resolved. Either way, it's good to have somebody who doesn't get REAL MAD when he argues.
deletedover 13 years
Also, completely disagree with your "40% town win is mafia sided, so 40% innocent means it's unlikely". I would say that something is statistically significant when you get below 16.67% or above 83.33% (i.e., outside of the middle 66% of what should happen). I would say that 1/4 is _about_ the threshhold of where I would say something starts being unlikely.
deletedover 13 years
But why even consider the cop being hooked/drunk if you're looking for reports anyways? Once you look at the chances as isolated cop report, you can validify the likelihood of a report flipping innocent/guilty. The chances of cop being drunk, hooked should be
If you care about hooker: 10% (12.5% * .8) of the time Drunk will get to Framer; causing no report. Did more math, refer to setup for examples. If you multiply those average framer percentages by .9, you get roughly the values I gave. It's perfectly reasonable estimation.
I admittedly don't care about those situations because once cop gets a report I want to evaluate how likely it is.
deletedover 13 years
Kail, so at base: cop can get 4 innos, 4 guilties. Framer can target: cop, miller, three blues, and drunk. 66% of the time, framer adds another guilty. On average, 4.66 guilties and 3.33 innocents.
41.625% of getting innocent. 58.375% of getting guilty
~16.75% more likely to get innocent rather than guilty. Not that significant. That's the general situation. It's not that much unlikely.
Scythain
