The median catalogue production credits 8 performers, and cast size correlates only 0.22 with episode count.
The relationship is not just weak, it is not even monotonic. Whatever sets the size of a cast here, the length of the finished show is not it.
Across the 66 catalogue records that carry a supporting cast list, the median production credits 8 performers in total. The largest is Love of Silom at 21. Cast size barely moves with episode count (r = 0.22) or with total runtime (r = 0.34), so a longer show is not reliably a more populated one.
Every title record on this site has four separate performer fields: actor for billed leads, supporting_cast, guest_cast and voice_actor. The count used throughout this article is the sum of all four, per record, with no performer counted twice inside a single record. I checked for that: no record repeats an ID across its own four fields.
The exclusion that matters most is our own. Of the 170 records, 55 have an empty supporting_cast field and 106 have an empty guest_cast field. Those two groups nest exactly: every record missing a supporting list is also missing a guest list. That is a coverage gap in our data, not evidence of small productions, so all 55 are excluded from the averages below. Treating them as casts of two would drag the median down by pure bookkeeping.
Of the 55 excluded, 26 are print source records (novels, web novels, manga, manhwa, manhua) where a cast field has no meaning. The other 29 are screen records where we simply have not filled the field in yet: 13 TV series, 8 web series, 5 animation records, 2 specials and 1 film. Seven records list no performers at all in any field.
| Measure | Performers |
|---|---|
| Smallest | 3 |
| Lower quartile | 6 |
| Median | 8 |
| Upper quartile | 10.75 |
| Largest | 21 |
| Mean | 8.6 |
| Records counted | 66 |
| Performer credits counted | 565 |
The twelve largest casts in the catalogue are below. Eleven of the twelve run between 8 and 16 episodes, which is the ordinary band for this catalogue. The one 24 episode entry, Fourever You Part 2, sits at the bottom of this table with 12 credited performers despite carrying 22.8 hours of runtime, roughly 70 percent more than Love of Silom, which tops the table.
| Title | Country | Leads | Supporting | Guest | Total | Eps |
|---|---|---|---|---|---|---|
| Love of Silom | Thailand | 2 | 16 | 3 | 21 | 12 |
| The Heart Killers | Thailand | 4 | 10 | 5 | 19 | 12 |
| Sammy’s Children’s Day | China, Hong Kong | 2 | 2 | 14 | 18 | 9 |
| We Are | Thailand | 6 | 9 | 0 | 15 | 16 |
| Payback | Thailand | 6 | 6 | 2 | 14 | 10 |
| ChermChey | Thailand | 2 | 3 | 9 | 14 | 10 |
| 23.5 the Series | Thailand | 2 | 12 | 0 | 14 | 12 |
| Pluto the Series | Thailand | 2 | 11 | 0 | 13 | 12 |
| WEIRDO-101 | Thailand | 2 | 11 | 0 | 13 | 10 |
| Wishing Upon the Shooting Stars | Taiwan | 4 | 8 | 0 | 12 | 12 |
| Flower Boy | Thailand | 2 | 8 | 2 | 12 | 8 |
| Fourever You Part 2 | Thailand | 6 | 5 | 1 | 12 | 24 |
Those twelve reach their totals by different routes. Love of Silom, 23.5 the Series, Pluto the Series and WEIRDO-101 each pair two billed leads with a deep supporting bench and no guest credits. Sammy’s Children’s Day does the opposite: 14 of its 18 credits are guest performers. Both shapes produce a similar headline number and describe completely different production structures.
The obvious hypothesis was that more episodes means more roles. It does not survive the arithmetic. Across the 64 credited records with an episode count, the correlation between total cast and episode count is 0.22 by Pearson and 0.21 by Spearman. Across the 59 records where a total runtime can be derived, cast against runtime is 0.34 by Pearson and 0.26 by Spearman. Both are weak. Neither explains more than about a ninth of the variation.
Grouping instead of correlating gives the same answer, and adds a wrinkle. Records of 1 to 8 episodes have a median cast of 7. Records of 9 to 12 episodes have a median of 9. Records of 13 to 24 episodes have a median of 8, which is lower than the middle band. The relationship is not just weak, it is not even monotonic.
Episode duration behaves the same way: 0.36 across 58 records. Whatever sets the size of a cast here, the length of the finished show is not it.
Of the 66 credited records, 50 bill exactly two leads. The rest split into 6 records with three leads, 6 with four, 3 with six and one with eight, The Crow Club. Widen the lens to all 144 screen records and 67 bill exactly two. The lead field is close to a constant.
The supporting field is where the range lives: 1 to 16 performers. Across the credited set, 168 billed leads carry 349 supporting performers, a ratio of 2.08 supporting per lead. Per record, the median ratio is also 2.0, with a lower quartile of 1.31 and an upper quartile of 3.0. The extremes run from 0.12 to 8.0.
| Country | Records | Median cast | Range | Supporting per lead |
|---|---|---|---|---|
| Thailand | 39 | 9 | 3 to 21 | 2.21 |
| South Korea | 9 | 7 | 3 to 8 | 1.36 |
| Japan | 7 | 7 | 3 to 10 | 2.50 |
| China | 5 | 6 | 4 to 18 | 1.70 |
| Taiwan | 4 | 9 | 5 to 12 | 1.92 |
| Hong Kong | 1 | 11 | 11 | 2.67 |
| Philippines | 1 | 6 | 6 | 2.00 |
By format, the pattern is flatter still. Among credited records, TV series run to a median of 9 and web series to a median of 6.5. No animation record survives into the credited set, because animation records here carry voice_actor entries and no supporting list, so the format cannot be compared on this measure at all.
Sample: all 170 title records in the blglflix.com catalogue as of 15 September 2026. Fields read: actor, supporting_cast, guest_cast, voice_actor, episodes, episode_duration, movie_length, content_type_text, country_text and couples. Total runtime was derived as episode count multiplied by episode duration in minutes, falling back to movie length where a record has no episodes. Correlations are Pearson and Spearman, computed directly from those values.
The first and largest limitation is the coverage gap already stated: 55 of 170 records have no supporting cast list and 106 have no guest cast list. Only 15 records in the whole catalogue carry any guest credits, 48 credits between them, and 12 of those 15 records are Thai. Any statement here about guest performers describes 15 records, not a catalogue.
Two things looked like findings and did not hold up. First, cast size appeared to correlate with the number of named performer pairings on a record at r = 0.42, which would have been the strongest relationship in this article. It collapses under a country control: all 27 non Thai credited records carry zero pairings, so the correlation is measuring the Thailand and non Thailand difference and nothing else. Within Thai records alone it drops to 0.38, and the five Thai records with no pairing are a sample too small to settle it.
Second, the GMMTV records in the credited set have a median cast of 10 against 7 for the other 52. That gap is real in the data but it cannot be separated from country: all 14 of those records are Thai, and Thai records already carry a median of 9 in the table above. It is reported here as an observation, not a studio effect.
One further caution. These fields reflect how much of a cast list we have entered, which depends on the material available when the record was made. The 2026 records dominate the credited set at 44 of 66, so the figures here lean recent.
Across the 66 catalogue records that carry a supporting cast list, the median total is 8 credited performers. The lower quartile is 6, the upper quartile is 10.75, and the full range runs from 3 to 21.
Love of Silom, with 21 credited performers: 2 billed leads, 16 supporting and 3 guest. The Heart Killers follows at 19 and Sammy’s Children’s Day at 18.
Not reliably. Total cast correlates 0.22 with episode count and 0.34 with total runtime. Records of 13 to 24 episodes actually have a lower median cast (8) than records of 9 to 12 episodes (9).
Two. Of the 66 credited records, 50 bill exactly two leads, and across all 144 screen records 67 do. Supporting cast is where the variation sits, from 1 to 16 performers.
Total cast is the sum of four fields on each title record: actor, supporting_cast, guest_cast and voice_actor, with no performer counted twice inside a record. Runtime was derived as episode count multiplied by episode duration, falling back to movie length. All 170 records were read on 15 September 2026.
Those 55 have an empty supporting_cast field. That is a gap in our own data entry, not evidence of a small production, so including them would lower the median for bookkeeping reasons. 26 of the 55 are print source records where a cast field has no meaning.
No. It reports median cast size and the supporting to lead ratio by country as structural measurements. Sample sizes outside Thailand are small, four of the seven country rows have fewer than eight records, and no quality claim is made or implied.
Two things. Cast size appeared to correlate 0.42 with the number of named performer pairings, but all 27 non Thai credited records carry zero pairings, so that correlation was measuring country. The GMMTV median of 10 against 7 elsewhere likewise cannot be separated from country, since all 14 of those records are Thai.
Metadata and images in this article come from TMDB and official sources, and remain the property of their copyright holders. This product uses the TMDB API but is not endorsed or certified by TMDB. BLGLFLIX is an independent site with no affiliation to the productions covered. Full copyright, correction and takedown policy: Disclaimer.
Discussion 0 comments
Join other fans talking about Cast Size in BL and GL, and What It Tracks. Be kind, no spoilers without tags, no spam.
No comments yet.
Share the first reaction below.