Compare the same kind of number
When comparing teams, use the same time window and definition on both sides. A partial live game and a completed game are different samples. Likewise, a total and a percentage answer different questions: makes describe volume, while makes divided by attempts describes conversion.
Calculate shooting percentages from totals
Suppose one player shoots 1–1 and a teammate shoots 1–9. Averaging their percentages gives 55.6%, but the combined result is 2–10, or 20%. To produce a team shooting percentage, add the makes, add the attempts, and divide those totals. Do not average individual percentages without accounting for attempts.
For a viewer, this matters because a perfect percentage on a very small number of attempts can look more impressive than a larger body of efficient shooting. Always keep the denominator in view.
Separate a score difference from an explanation
A nine-point lead tells you the current margin. It does not establish which possession, player or tactical choice caused it. Our quick-read summary is a compact description of the displayed score. It is not a play-by-play reconstruction or a claim about momentum.
Allow for incomplete rows
Our comparison view adds the player statistics returned by the provider. A missing player row, a late correction or a team-only adjustment can create a mismatch with an official team total. If player points do not reconcile with the displayed score, treat the comparison as incomplete rather than forcing the numbers to agree.
The same caution applies to rebounds and turnovers, where team-level recording can matter. The scoreboard’s game total and the sum of player rows are distinct data views.
Keep the display readable
A rotating broadcast should hold one comparison long enough to read both teams and the category label. Frequent scene changes can create visual movement while reducing understanding. Pause the studio when examining a particular table, and use the team names and game clock to ensure the comparison still refers to the intended matchup.
What a fair takeaway sounds like
“The returned player rows show more rebounds for one team” is narrower and better supported than “that team controlled the entire game.” The first describes the data. The second needs more context. Good live analysis keeps that distinction visible.
Explore the player box-score guide or review our methodology.