One theory has been thrown around among the more avid watchers of the game who aren't so analytically inclined - the theory that basketball is nothing more than a game of scoring runs. Essentially, an individual game is ultimately decided by the frequency and magnitude of scoring runs each team can mount against the other. The team that wins usually ends up mounting more scoring runs or longer runs overall.

It's a theory that has legs in my humble opinion. I've seen the effect scoring runs can have, especially when the crowd begins to ride upon the home team's momentum - a court packed with cheering fans combined with the psychological impact of a widening scoring differential can work wonders for the home team.

I'm going to investigate this. To do this though, I need to define a scoring run in objective terms. On the surface level, most people would characterize a scoring run as a phenomenon (usually short-lived) where one team begins racking up far more points than their opponent in a runaway fashion. Can we translate that definition into something resembling a mathematical formula? Of course.

A scoring run can be fully defined as the period between two events, the event that initiates the run and the event that concludes the run.

• A scoring run is only initiated when one team scores six points without the other team scoring at all.
• A scoring run concludes when any of the following events takes place:
  • A timeout is called
  • The other team scores twice on consecutive possessions without the "hot" team scoring once
  • The "hot" team goes two consecutive possessions without scoring once
  • The other team scores at least 50% of what the "hot" team is scoring during the run

This is a fairly restrictive definition. But it's a definition I feel very comfortable analyzing. Best of all, it's rigorous enough to the point where there's little doubt about its precision.