If Call of Duty kills were random, what would the distribution of kills look like?
There is probably a nice analytical solution to this, but I kind of just felt like knowing. The important file is CODsim.py. It defines the Player and Sim classes, which are the structure of the simulation. The runner.py file just calls it for a couple different values of players and saves charts of the results as images to the sampleruns directory. There are basically no comments, sorry. I think the names are good.
The sampleruns directory contains pretty pictures to look at. The runner.py file demonstrates what you need to do to use the code.