In recent time, a new surge of people have found their way to chess. A crucial part of modern chess is the systems used to rate players. These systems build the foundation for how players are matched against opponents with similar strength. The first rating system that used statistical theory laid the foundation for almost all systems in use today, both in chess and other sports. This was the Elo rating system. As years passed, the system was found to have big issues which led to more complex systems taking its place. This paper investigates the possibility of revitalising the use of a simple rating system. The suggested system in this paper is based on the Elo rating system and uses ordinal logistic regression to model chess game outcomes. The paper uses an iterative simulation study to estimate and simulate new player ratings. The findings from this study indicate that a simple system can still perform well in rating chess players. The findings show our estimated model for calculating expected scores have a good fit on the player ratings which our simulation study produces. It further finds that running an iterative simulation study does not increase the performance of the suggested system heavily.