Sports score systems vary considerably from sport to sport, and they are often defined by a combination of points, duration, time limits, and scoring events. Nevertheless, a few common patterns can be observed across a wide variety of sports.

For example, hockey games are divided into four scoring periods, and the game clock is used to control the speed of the gameplay. As the game progresses, players must reset their positions at the end of each period.

A similar pattern is found in basketball. The first half of a game is more predictable than the second.

However, the NFL and CFB are less predictable. In the first half of CFB games, the Markov model overestimates the lead-size variation in comparison to the empirical pattern. This suggests that lead-size dynamics can influence the dynamics of gameplay.

One such process is the Poisson process, which describes the timing of scoring events. It predicts that C(n) will be zero for all n that are greater than or equal to 0; for all n that are less than 0 and not greater than 0 the probability of scoring increases.

Another model, called the Bernoulli process, is simpler. It is based on a data-driven distribution of events, which can be modeled without memory. When a team is leading, its probability of scoring increases.

Some sports have specific scoring events, such as the long jump, shot put, and hammer throw. Others, such as rugby, have more complicated procedures.