Answer first: SportSG’s 2024 Sports Facilities National Census contains 16,597 rows in the current data.gov.sg download. Our reproducible analysis found 3,835 distinct non-blank venue names and 10,808 distinct venue-facility pairs, so a row is not the same thing as a bookable public venue.
This guide is written for a resident, coach or community organiser trying to shortlist venues for a specific sport. The task is to use the official facility census as a discovery layer, then verify access, ownership, booking and duplicate facility records before travelling. It separates official rules from calculations, planning assumptions and outcomes that still require a live readback.
Before acting, write the case-specific date, person, property, account, venue or product at the top of the working note. A source can be authoritative and still be applied to the wrong facts. The controls below are designed to expose that mismatch early.
What the official sources establish
The file is a facility-level census
The data.gov.sg page describes facilities collected through the 2024 national census and shows five fields. This is stated in the primary official source.
What to do: Use each row as a facility observation, not a unique venue. For the reader, that means the next action should be tied to a named record and a date, not to a remembered headline.
The download has 16,597 rows
The current CSV hash and row count were recorded at analysis time; two rows had blank venue names and four had blank facility types. This is stated in the primary official source.
What to do: Retain the snapshot date because the dataset can be revised. The distinction matters because a broadly correct rule can still produce the wrong decision when applied to the wrong route or date.
Venue names repeat heavily
The analysis found 3,835 distinct non-blank venue names and 12,760 additional rows using an already-seen name. This is stated in the primary official source.
What to do: Group carefully without deleting separate courts, pools or gyms. Keep the source beside the decision it supports so that a later reviewer can see both the rule and the case-specific input.
Even venue-facility pairs repeat
There were 10,808 distinct venue-facility pairs and 5,787 repeated rows beyond those pairs. This is stated in the primary official source.
What to do: Treat duplicates as a data-quality prompt, not proof of extra capacity. If the underlying fact changes, reopen this step instead of allowing an old conclusion to travel forward unnoticed.
Facility type counts show discovery patterns
Swimming Pool appeared 3,532 times, Fitness Corner 2,872, Gymnasium 2,569 and Badminton Court 1,177. This is stated in the primary official source.
What to do: Use counts to understand the file, not to promise public access or availability. A blank or disputed input is a stop condition. It is not permission to insert a convenient assumption.
Booking is a separate system
ActiveSG’s facilities page is the live route for participating public venues and programmes. This is stated in the supporting official guidance.
What to do: Verify operator, access, hours and booking before travelling. For the reader, that means the next action should be tied to a named record and a date, not to a remembered headline.
Decision table
| Question | Reader action | Authority |
|---|---|---|
| The file is a facility-level census | Use each row as a facility observation, not a unique venue. | primary official source |
| The download has 16,597 rows | Retain the snapshot date because the dataset can be revised. | primary official source |
| Venue names repeat heavily | Group carefully without deleting separate courts, pools or gyms. | primary official source |
| Even venue-facility pairs repeat | Treat duplicates as a data-quality prompt, not proof of extra capacity. | primary official source |
| Facility type counts show discovery patterns | Use counts to understand the file, not to promise public access or availability. | primary official source |
| Booking is a separate system | Verify operator, access, hours and booking before travelling. | supporting official guidance |
Work down the table in sequence. Do not close an item with a search snippet, an undated screenshot or another person’s outcome. For material money, eligibility, safety or legal points, save the current authority page or formal readback and note its date.
Two tools that add practical value
A reproducible count of rows, distinct venue names, facility types, missing-name records and repeated venue-facility pairs
Build this as a compact table with columns for the reader’s actual input, the dated evidence, the rule it activates, the responsible person and the next irreversible step. The table must preserve alternatives rather than collapse them into one total. Mark estimates and pending confirmations visibly, because an elegant calculation based on the wrong route is still wrong.
A discovery-to-booking decision tree that prevents a dataset coordinate or facility label from being mistaken for guaranteed public access
Use this as a separate challenge to the first analysis. Test the strongest contrary scenario, the missing document and the event most likely to make the answer stale. Keep the check practical: it should change a date, amount, route, booking, household rule or go/no-go decision. If it cannot affect the decision, remove it instead of padding the file.
Worked example
A coach filtering for badminton sees 1,177 rows, but does not tell players that 1,177 courts are available. The coach groups by venue, checks whether repeated rows represent separate facilities, narrows by location and then opens the operator or ActiveSG page. A condominium or school facility in the census may not be open for casual public booking.
The example is labelled as an illustration. Replace its circumstances and figures with current evidence, and preserve the branch that was rejected so the reasoning can be reconstructed later.
Action checklist
- Use each row as a facility observation, not a unique venue. Retain the evidence supporting this point: The data.gov.sg page describes facilities collected through the 2024 national census and shows five fields.
- Retain the snapshot date because the dataset can be revised. Retain the evidence supporting this point: The current CSV hash and row count were recorded at analysis time; two rows had blank venue names and four had blank facility types.
- Group carefully without deleting separate courts, pools or gyms. Retain the evidence supporting this point: The analysis found 3,835 distinct non-blank venue names and 12,760 additional rows using an already-seen name.
- Treat duplicates as a data-quality prompt, not proof of extra capacity. Retain the evidence supporting this point: There were 10,808 distinct venue-facility pairs and 5,787 repeated rows beyond those pairs.
- Use counts to understand the file, not to promise public access or availability. Retain the evidence supporting this point: Swimming Pool appeared 3,532 times, Fitness Corner 2,872, Gymnasium 2,569 and Badminton Court 1,177.
- Verify operator, access, hours and booking before travelling. Retain the evidence supporting this point: ActiveSG’s facilities page is the live route for participating public venues and programmes.
Assign an owner and due date to every open item. If the earliest irreversible step arrives before the critical evidence, pause. That is especially important before paying, signing, applying, travelling, changing payroll or relying on a health or safety plan.
Limits and final readback
The analysis is a dated computation over the downloaded file, not a field inspection. Names, coordinates, ownership and access can be incomplete or duplicated. The live operator controls use.
At the point of action, reopen both the primary official source and the supporting official guidance. Check publication or update dates, confirm that the quoted rule still appears, and use the authority’s live service or named operator where a case-specific result is required.



