Some shows have an episode that everyone seems to agree is extraordinary. How rare is that? I looked beyond the IMDb Top 250 at all TV series and miniseries with at least 25,000 series votes in IMDb’s official downloadable datasets. An exceptional episode earns at least 9.7/10 with at least 5,000 episode votes.
Snapshot: 03 October 2026 (UTC). Ratings and vote counts can change.
31 qualifying shows are outside the saved Top 250 cohort, contributing 62 exceptional episodes. The saved list is a historical comparison: I haven’t verified today’s Top 250 membership because IMDb’s live chart blocked access.
This is an episode count, so longer shows have more chances to qualify. It is not a ranking of overall show quality. The labels show each count and its series-vote total; the full searchable table follows. On phones, labels sit above thicker bars that use the available chart width.
All 1,340 eligible shows are included, even those with zero exceptional episodes. Search for a title or genre. On narrow screens, expand a show’s row to see all details and vote cutoffs. “Saved Top 250” refers to historical membership, not today’s chart.
Exceptional share = 100 × exceptional episodes ÷ total episodes. The table’s percentage uses the primary cutoff: rating ≥9.7 with ≥5,000 episode votes. “Total episodes” means every episode linked to the show in IMDb’s snapshot, including specials, unrated entries, and any upcoming placeholders. It is not restricted to episodes that received enough votes.
The rating cutoff stays at 9.7 and the series-vote cutoff stays at 25,000. Only the minimum number of episode votes changes.
| Minimum episode votes | Shows with exceptional episodes | Share of eligible shows | Exceptional episodes |
|---|---|---|---|
| 1,000 | 89 | 6.64% | 250 |
| 5,000 | 78 | 5.82% | 229 |
| 10,000 | 57 | 4.25% | 169 |
Each tab applies its own minimum episode-vote cutoff. Every episode is from an eligible show and has rating ≥9.7. Search for a show or episode title; use the IMDb link to open its episode page.
The existing CSV contains 250 distinct show IDs. Of those, 218 meet the 25,000-series-vote cutoff in this snapshot. This comparison uses official episode ratings for those historical IDs. I haven’t rebuilt the chart by sorting bulk ratings: IMDb’s Top 250 uses its own ranking rules.
| Cohort | Shows | Shows with exceptional episodes | Share | Exceptional episodes |
|---|---|---|---|---|
| All eligible shows | 1340 | 78 | 5.82% | 229 |
| Saved Top 250: all members | 250 | 48 | 19.20% | 168 |
| Saved Top 250: eligible members | 218 | 47 | 21.56% | 167 |
| Eligible shows outside saved Top 250 | 1122 | 31 | 2.76% | 62 |
The cohort rows overlap: eligible saved members and eligible shows outside the saved list together make up the broader universe. The all-member saved cohort also includes 32 shows below the series-vote cutoff.
The universe comes from IMDb’s official
non-commercial datasets: title.basics identifies series
and miniseries, title.episode links episodes to their
parent shows, and title.ratings provides weighted ratings
and vote counts. Series votes are for the show itself; they are not a
sum of episode votes. There are no language, country, animation, or
documentary exclusions.
Existing CSVs serve as validation checks. The stored Top 250 CSV has 15,519 rows; 15,405 match a unique official series/season/episode key. Twenty matched rows change their primary exceptional classification between snapshots. Because the stored CSV lacks episode IDs, matching is provisional and can be affected by renumbering as well as changes in ratings and votes.
Specials and season-zero entries are included. The downloads lack exact episode air dates, so this analysis does not apply an exact release-date filter. A rating record is required. Vote cutoffs reduce the influence of very small samples but do not eliminate fandom effects or organized voting.
Download all eligible shows · Download primary exceptional episodes · Download cohort sensitivity results · Download CSV validation summary
The project source and refresh instructions include dataset hashes and source HTTP metadata for this dated snapshot. The page renders from checked-in data; building the website does not download IMDb’s full datasets.