Datasetpublic

illustricious/Indonesian-High-School-Basketball-Complete-Stats

Demographics told you who they are. This tells you how they play. A structured, multi-table dataset covering high school basketball competition in Indonesia match results with full team box stats, per-player per-game performance, and multi-season player statistics. Delivered as compressed Parquet files, ready for pandas, polars, DuckDB, or the datasets library. This is the follow-up to illustricious/indonesian-highschool-basketball-demographics(player height, weight, gender, and school city… See the full description on the dataset page: https://huggingface.co/datasets/illustricious/Indonesian-High-School-Basketball-Complete-Stats.

sourceHugging Facecc-by-sa-4.0updated 26d agoView on Hugging Face

0likes38downloads

Dataset Card

Indonesian High School Basketball Complete Stats

language:

—id
—en

tags:

—basketball
—sports
—sports-analytics
—indonesia
—high-school
—tabular
—real-world-data

Demographics told you who they are. This tells you how they play.

A structured, multi-table dataset covering high school basketball competition in Indonesia match results with full team box stats, per-player per-game performance, and multi-season player statistics. Delivered as compressed Parquet files, ready for pandas, polars, DuckDB, or the datasets library.

This is the follow-up to `illustricious/indonesian-highschool-basketball-demographics`(player height, weight, gender, and school city only). This dataset goes further — actual quarter-by-quarter scores, made/attempted/percentage splits for 2PT/3PT/FT, rebounds (offensive/defensive), assists, steals, blocks, turnovers, fouls, and points off turnovers, at both the match level and the player level, plus season-aggregated player statistics across multiple years.

⚠️ Real data, not synthetic — but please validate

This dataset was collected from real match and player records, not generated or simulated. Unlike some sports datasets circulating on Kaggle/Hugging Face that are partially or fully synthetic (randomly generated stats dressed up as real competition data), everything here traces back to actual games and actual players.

That said: this was compiled independently, not released by an official body, so treat it as community-collected, not authoritative. A few things to keep in mind before you rely on it for anything serious:

—Some fields may contain parsing artifacts or NULL values where the original source data was inconsistent or incomplete (see Notes & limitations below).
—Cross-check any numbers you plan to publish or cite against another source if possible.
—If you spot an error, please open a discussion on this dataset's page — corrections are welcome.

Dataset structure

`matches.parquet`

One row per game. Team-level (_a / _b) box score for both sides.

Column	Type	Description
`match_id`	INT32	Primary key
`match_time`	TIMESTAMP	Game date/time
`venue_id`	INT32	FK → `venues.venue_id`
`category`	VARCHAR	Competition category
`school_id_a` / `school_id_b`	INT32	FK → `schools.school_id`
`score_a` / `score_b`	SMALLINT	Final score
`q1_a` … `q4_a`, `q1_b` … `q4_b`	SMALLINT	Per-quarter score
`ot_a` / `ot_b`	SMALLINT	Overtime score
`fg_made_` / `fg_att_` / `fg_pct_*`	SMALLINT / SMALLINT / FLOAT	Field goal stats
`fg2_`, `fg3_`	SMALLINT / FLOAT	2PT and 3PT splits
`ft_made_` / `ft_att_` / `ft_pct_*`	SMALLINT / SMALLINT / FLOAT	Free throw stats
`reb_off_` / `reb_def_`	SMALLINT	Rebounds
`assists_`, `steals_`, `blocks_`, `turnovers_`, `fouls_*`	SMALLINT	Standard box score stats
`points_off_turnover_*`	SMALLINT	Points off turnovers

`player_boxscores.parquet`

One row per player, per match — the granular fact table.

Column	Type	Description
`id`	INT32	Primary key
`match_id`	INT32	FK → `matches.match_id`
`school_id`	INT32	FK → `schools.school_id`
`jersey_no`	VARCHAR	Jersey number
`player_id`	INT32	FK → `players.player_id`
`position`	VARCHAR	Player position
`seconds_played`	INT16	Time on court, in seconds (parsed from `MM:SS`)
`points`, `assists`, `rebounds`	SMALLINT	Game totals

`player_statistics.parquet`

One row per player, per season — aggregated performance.

Column	Type	Description
`id`	INT32	Primary key
`player_id`	INT32	FK → `players.player_id`
`year`	SMALLINT	Season year
`gp`	SMALLINT	Games played
`mpg`	FLOAT	Minutes per game (decimal minutes, parsed from `MM:SS`)
`pts`	SMALLINT	Points per season total
`fg_pct`, `three_pt_pct`, `ft_pct`	FLOAT	Shooting percentages
`trb`, `ast`, `blk`, `stl`	SMALLINT	Rebounds, assists, blocks, steals

`players.parquet`

Column	Type	Description
`player_id`	INT32	Primary key
`is_male`	BOOLEAN	`TRUE` for male players, `FALSE` for female
`birth_year_estimate`	SMALLINT	Estimated birth year
`height_cm`, `weight_kg`	SMALLINT	Physical attributes
`school_id`	INT32	FK → `schools.school_id`

`schools.parquet`

Column	Type	Description
`school_id`	INT32	Primary key
`name`, `city`, `province`	VARCHAR	School info

`venues.parquet`

Column	Type	Description
`venue_id`	INT32	Primary key
`name`	VARCHAR	Venue name

Notes & limitations

—Foreign-key id columns were parsed with TRY_CAST during conversion; a small number of rows may contain NULL where the source value didn't parse cleanly.
—seconds_played (per-game, integer seconds) and mpg (per-season average, decimal minutes) use different units by design — one is a raw per-game duration, the other a seasonal average.
—This is real, community-collected data — not synthetic — but it is unofficial and has not been independently audited. Please verify accuracy before using it for anything beyond exploratory analysis or modeling practice.

Suggested use cases

Win/loss prediction, player performance modeling, shooting efficiency analysis, team strength rating, season-over-season player development tracking, and general sports analytics practice.