A team's run differential is determined by subtracting the total number of runs (both earned and unearned) it has allowed from the number of runs it has scored.
The 2016 Chicago Cubs scored 808 runs during the regular season and allowed 556 runs, giving them a run differential of +252. The 2018 Baltimore Orioles, meanwhile, scored 622 runs as a team but surrendered 892 runs. Their run differential was -270.
Why it's useful
Examining a team's run differential can help to identify teams that are overachieving and teams that are underachieving. While there have certainly been clubs that have finished a season with a winning record and a negative run differential -- and vice versa -- those teams are statistical outliers. Looking at a team's run differential early in the season can prove instructive when trying to determine whether a club is capable of either sustaining a "hot" start or capable of rebounding from an early slump.
Like any stat, run differential has its limitations and is far from infallible. The team with the best run differential won't always win the World Series. The 2016 Cubs (+252) did just that, but the 2017 Indians (+254) and 2018 Astros (+263) failed to do so. All three clubs won their respective divisions. Generally speaking, the stat is a good barometer for the overall talent of a given team. It is also closely tied to pythagorean winning percentage -- another metric that aims to provide a truer glimpse of a team's talent than raw winning percentage.