An introduction into the mechanics of tackle football, and the factors that differentiate it from other sports, is provided in the context of the challenges these provide in terms of predicting game outcomes. Additional complicating factors specific to predicting outcomes of Canadian Football League (CFL) games are identified and discussed. The Bradley-Terry Model and various generalizations are presented as potentially useful for making such predictions, alongside derivations of well-known minorization-maximization (MM) algorithms and descriptions Artificial Neural Net- work (ANN) processes that can be leveraged to estimate such models' parameters. Various candidate models, parameterized by regular-season CFL game outcomes from 2004-2017, are performance-tested against post-season playoff game outcomes to relatively assess their prospective usefulness. i