Mine BlockSports media field notes
ReportingBroadcastPodcastAnalyticsEthics

How to Read NBA Viewership Ratings

Updated

Headlines about NBA television ratings can sound definitive: a game drew millions, a series rose or fell, or a stream reached a new audience. Each number answers a different question. Before drawing a conclusion from a report by Sports Media Watch or another outlet, identify the event, the audience measure and the viewing included.

This guide explains how to read those reports. It is independent of Sports Media Watch and the NBA; it is not a schedule, a live ratings feed or a claim about the current direction of the league’s audience.

Start with the event and the measurement

Record the date, teams, competition stage and outlet carrying the game. A regular-season game, a conference final and a championship game have different stakes and potential audiences. Even within a series, a deciding game may attract a different audience from an early one. A report about one broadcast cannot establish the trend for the whole season.

Next, look for the unit: a rating, an average audience, a peak or reach. Check whether the number covers households or people, all viewers or a specified age group, and which country or market is measured. Write those details next to the figure instead of relying on a headline alone. If the report omits them, leave the comparison unresolved.

Rating, share and audience are different

A rating expresses viewing as a percentage of a defined population. A household rating and a rating for adults in a specified age range use different populations and are not interchangeable. The percentage is an estimate produced by the measurement system, not a count of every person who watched.

Share uses a different denominator: it describes a program’s portion of the relevant audience watching television at the time. A strong share can coexist with a smaller viewer count than another event if the total number watching TV differs. Always read the source’s definition, including the audience group and viewing window. Nielsen’s explanation of TV audience measurement describes ratings, panels and the distinction between live and time-shifted windows.

Average audience is generally a count of viewers in an average minute of the measured program. It does not mean those same people watched every minute, and it is not the total number who sampled any part of the game. A peak describes a shorter high point; its time interval matters. Reach counts people exposed within a stated window and according to a stated threshold. It will answer a different question from average audience.

Worked example: average, peak and share

Imagine a hypothetical game that averages 2 million viewers, reaches 3 million during a specified peak interval, and earns a 10% share. These invented numbers are only a reading exercise. The average summarizes the measured telecast, the peak describes a particular interval, and the share expresses a proportion of a defined audience. None of the three alone tells you how many different people watched any part of the game.

To compare the imaginary game with another broadcast, find the same measures for a comparable event and check their definitions. If the second report gives only reach, a different territory or an unspecified peak interval, those figures do not establish a like-for-like difference. An early estimate may later be revised; retain the source date and whether a number is preliminary or final.

Live is a viewing window, not a metric

“Live” says when viewing occurred; “average audience” says how it was summarized. A report may give a live average, while another includes live viewing plus playback on the same day or over several days. Neither wording alone reveals every device or distribution channel included. Check the footnote or methodology before treating the two as equivalent.

A broadcast can be available through more than one route. A source may report one network, a combined television and streaming figure, or a platform’s own number. If it does not specify whether a service is included, do not assume that it is either included or excluded. Sports rights and distribution arrangements change, so consult the league or broadcaster for the relevant game and territory rather than relying on an old network list.

TV and streaming numbers need matching definitions

A streaming service may report average audience, concurrent viewers, unique viewers, accounts or video views. These terms should not be treated as synonyms. For example, starts or social video views can count repeated plays or brief clips, while a television average-minute audience describes viewing during an average minute. Nielsen’s comparison of TV and digital metrics explains why those measures answer different questions.

Adding figures is especially risky. A person might watch on more than one device, or a combined figure may already include a platform named separately. Unless the publisher explains overlap and deduplication, a sum can double-count viewing. A combined audience estimate is useful only when its scope and method are clear enough for the question being asked.

Make a fair comparison

For a year-over-year comparison, match the competition stage and, where possible, game number, day, time, geography and distribution. Note the number of games included in a series average: a longer series contains more broadcasts and can have a different mix of early and deciding games. A single game’s average should not be set against a season’s total reach.

Check whether figures use the same demographic and viewing window. Measurement methods, eligible devices and distribution may change between seasons. Such changes do not make a comparison useless, but they can limit what it proves. Do not attribute an increase or decline to a player, rule change or platform without separate evidence; scheduling, competition, matchup and measurement can all matter.

Read a table without overclaiming

Tables can compress important context into abbreviations or footnotes. Look for the column heading before copying a number: a household rating, a persons estimate and a demographic rating might appear side by side. Check the units too. A value expressed in thousands is not directly comparable to one expressed in millions until it is converted, and rounding can make small differences look more precise than the underlying figures justify.

For a series, identify whether the table covers every game or just selected national telecasts. A simple mean of the game averages describes the average across those listed games; it is not the number of unique people who watched at least one. If a source describes an audience record, keep the exact qualifier: a record for one network, game number, age group or measurement period is narrower than a record for the whole league.

Keep a short note of the original link and access date with your comparison. That makes a later correction easier to trace and lets another reader verify the definitions. When estimates conflict, compare their dates, sources, coverage and preliminary status before choosing one; an unexplained discrepancy is a reason to qualify the result, not to invent a cause.

A practical reading checklist

  1. Find the original report: capture its publication date, event and source of the estimate.
  2. Label the number: rating, share, average audience, peak or reach, with the population and geography.
  3. Check the window: live, same-day or another playback period; identify television and streaming coverage where stated.
  4. Match the comparison: use comparable games and definitions, and note changes in distribution or methodology.
  5. State the limit: distinguish an observed difference from an explanation, and identify missing information.

If an article links to a league release, network announcement or measurement explanation, read that material alongside its table. When the definitions cannot be matched, the most useful conclusion may be that the available figures cannot support a clean comparison. For more on the reporting context, explore our sports journalism coverage.