Sport score is the metric used to measure the outcome of sports competitions. These can be team or individual-based.

Points-based scoring is common in many team sports such as football and hockey. Teams earn points for each goal they score. However, there are some sports that have a more complex set of scoring mechanisms.

In racquet sports, players/teams can win points for either hitting the ball or shuttlecock out of bounds or if the opponent misses their own target. In most cases, these types of sports are played against an opponent individually or in teams of two, and they have no time limit.

Game balance is an important topic in sport, since it determines whether a team has a good chance of winning or losing a game. This can be measured by a simple estimate: the fraction of events won by one team relative to that won by the other.

We investigate these issues using data on scoring events from every league game in four sports. We show that these events describe the observed evolution of tempo and balance in games quite well.

Across all four sports, a Poisson process with a sports-specific rate occurs at each second on the game clock (Figures 2, 3). This pattern is remarkably stable over the course of gameplay.

We combine these insights within a model of gameplay and show that this model accurately reproduces the observed tempo and balance patterns in scoring events, and makes accurate predictions of game outcomes, when only a few scoring events have occurred. Cursory comparisons suggest that this model achieves accuracy comparable to or better than several commercial odds-makers, despite knowing nothing about teams, players, or strategies.