To ensure fairness and clarity, the structure of individual matches must also be carefully defined. Decisions such as whether matches consist of a single game, a best-of-two, or a longer series significantly impact the reliability of results. The chapter emphasizes that playing both colors—White and Black—minimizes bias and promotes a more accurate evaluation of each AI’s strength. Clear rules regarding move submission, time controls, and expected behavior under the tournament’s API ensure that students understand not only how their AIs should behave, but also how to design them accordingly.

Evaluation and scoring form the backbone of the competitive process. Standard chess scoring—one point for a win, half a point for a draw, and zero for a loss—provides a familiar framework for ranking teams. However, AI competitions introduce additional considerations, as technical failures such as crashes, illegal moves, or timeouts must be treated consistently. The chapter details how these non-standard outcomes are recorded and how tie-breakers, such as Buchholz or head-to-head results, ensure fair and transparent rankings when teams finish with equal scores.