NationStates Dossier Explorer

1. Data provenance — downloaded versus derived

2. World Census Explorer — ranks and bars

Each of the approximately ninety World Census scales (e.g. “Most Efficient Economies”) ranks every nation. The API reports two quantities per scale:

• Score — the entity’s raw value; used in charts and formulas, not shown on the bar.
• Percentile rank (prank) — the official world percentile. A rank of Top 15% indicates that the entity outperforms 85% of nations on that scale.

The bar is a visual encoding of the rank: its length equals 100 − prank, so the highest-ranked entities fill the bar. Scales are grouped into nine colour-coded thematic families; selecting a row opens the official history of that scale as a trend graph. Bars, ranks and trends are all downloaded; the Explorer derives nothing.

3. Social-progressiveness gauge

Definition. The gauge summarises how open the society is, using the civil-liberties and political-rights scores exclusively. Economic indicators are deliberately excluded: a wealthy nation with restricted civil liberties is not classified as progressive.

Formula: progressive = (Civil Rights score + Political Freedom score) / 2, each axis scaled 0–100.

The labels characterise the civil-liberties regime, not economic policy: a market economy may still be classified “Traditionalist” where personal freedoms are restricted.

4. Political compass — coordinate model

The compass places the entity on a two-axis plane whose origin is the world average (0, 0). Each axis spans −100 to +100. Values are clipped only when the government-size adjustment would push them beyond the ±100 boundary.

X — economic axis (Left ↔ Right)

Interpretation. Negative values indicate predominant state ownership and management of the economy; positive values indicate predominantly private enterprise and market allocation.

The axis is driven by a structural measure of who controls production — market actors or the state — rather than by output. Where the API publishes the production-sector split (share of GDP from state-owned industry versus private industry plus black market), it defines the lean; otherwise — for regions, the world and historical months — the lean is derived from the Economic Freedom census scale (ID 48), the game’s own market-versus-state indicator, which is also incorporated in every case. The “Economy” census is not used alone: it measures output, not ownership, and a fully state-run economy may still score “Powerhouse”. Notation: market = private + black market, state = state-owned.

sectorLean = (market − state) / (market + state)  ·  without a split: sectorLean = (Economic Freedom − 50) / 50  (−1 fully state; +1 fully market), where Economic Freedom is the raw signed score (−100 fully state … +100 fully market; rescaled to the 0–100 index used elsewhere as (raw + 100) / 2)

index = (Economic Freedom + Economy) / 2  (both on the 0–100 index; falls back to Economy when scale 48 is unavailable)

X = sectorLean × 92 × 0.9 + (index − 50) × 0.1  — applied when the lean is decisive (|lean| > 0.25); for mixed economies the two components are blended in a 45/55 proportion.

Y — authority axis (Authoritarian ↔ Libertarian)

Interpretation. Negative values denote concentration of power in a small ruling group; positive values denote broad civil liberties and accountable leadership.

Y = ((Civil Rights + Political Freedom) / 2 − 50) × 2, reduced for large governments: Y −= (Government Size − 50) × 0.6, where Government Size is census scale ID 27.

The background is divided at ±33 on each axis into nine zones (Authoritarian Left … Libertarian Right); zone colours correspond to the compass legend.

5. Ideology — classification method

NationStates defines no official ideology. The label shown is the output of a deterministic classification procedure applied to the data described in Section 4:

1. Positioning. The compass point is placed with the formulas in Section 4.
2. Similarity ranking. The point is compared with a reference set of — named ideologies. Each reference ideology stores an anchor on the compass plane together with its expected traditionalism and radicality; the society profile therefore discriminates between ideologies that occupy nearly the same compass position (e.g. Paleoconservatism versus One-Nation Conservatism, or Theocracy versus generic Authoritarianism). The reference set covers the entire plane, including the democratic far left, state socialism, the ecological centre and traditionalist variants.
3. Census validation. Selected ideologies carry a “signature”: a definitional trait that must be present in the data. An Anarcho-Primitivist label, for instance, requires the entity to rank among the most primitive nations (or to combine high eco-friendliness with low technological development). Where the nearest ideology contradicts the census profile it is rejected and the next-nearest compatible ideology is assigned; the dossier explanation records any such rejection.

Complete list of inputs. Every datum the classifier consumes, and the role it plays:

Observed inputs and range calibration. Traditionalism is the average of Social Conservatism and Religiousness. Secularism is deliberately excluded: in NS data it is the exact inverse of Religiousness, so including both would count the same signal twice. Every raw input is normalised to its observed range before use, because most census scales are not 0–100 indices:

Religiousness contributes to individual nations only: for regions and the world the percentile map differs, so those dossiers rely on Social Conservatism alone. Radicality is read from Ideological Radicality and rescaled as above. Economic Freedom contributes to the X coordinate (Section 4) and to the economic characterisation (laissez-faire / mixed-market / planned).

Each dossier’s ideology card includes a written justification. The text states the assigned label and the evidence supporting it: the map zone and coordinates, the underlying production and civil-liberties measures, the society profile compared with the reference expectations, and — where a nearer ideology was not selected — the criterion that excluded it (a failed census signature or a percentile-profile comparison). Near-equal candidates are resolved by the entity’s percentile profile across welfare, ecology, order and society scales; the world dossier and the historical rewind carry scores only and therefore use the map-based comparison alone.

Evidence availability and the primacy of the current label. Not every evidence channel is archived by NationStates. The census history endpoint stores numeric scores only: it carries no record of the game category (the sole authority on the form of the head of state) and no record of the government budget shares used by several census requirements. The current, live classification is therefore the most precise label the data can produce: it alone can apply the category gate for the monarchy labels and the budget-share gates for Theocracy, Clerical Fascism, Neoconservatism, Social Democracy, Technocracy, State Capitalism and the green family. The historical rewind re-classifies each past month through the same pipeline but from archived scores alone, so category- and budget-dependent requirements cannot fire there; rules whose data is absent are skipped rather than treated as contradictory, and those months are labelled by compass position and society profile alone. Historical labels are consequently best read as map-based approximations, while the label in the dossier header — computed from the complete live evidence set — is always the authoritative one.

The table below reproduces the full reference data stored for every ideology: its family and exact anchor coordinates; the expected traditionalism and radicality (0–100) against which the entity is compared; any census requirement — a definitional trait that must actually be present in the data for the label to be retained — and any preferred profile, a set of traits used only to break near-ties on the map. Requirements draw on three evidence kinds: census percentile intensities, government budget shares (e.g. “Spirituality budget share ≥ 20%”) and the game’s own category (the sole authority on the form of the head of state, used for the monarchy labels). A requirement listing several alternatives separated by “or” is satisfied when any one of them holds; a row showing “definitional label only” has no mandatory trait, and its description claims are all measurable through the anchors or the preferred profile. All thresholds use percentile intensity (100 − prank, 0–100); a rule whose data is unavailable is skipped, so score-only contexts (the world dossier and archived months) are never vetoed on rank evidence.

Description policy. Every sentence in the ideology summaries states a fact the data can confirm — claims that NationStates cannot measure (historical programmes, hereditary legitimacy, revolutionary intent) are either re-expressed through their measurable signature or explicitly anchored to the game category.

Example nations. Under each ideology name the table lists a real nation whose census profile this classifier detects as that ideology (“for example”). The example rotates daily (one per ideology, cycling through the verified pool as the days advance) and clicking it opens that nation’s dossier. The pools are maintained server-side with two cooperating sources, all inside the NationStates rate limits: (1) the daily data dump — the same once-a-day download used for the regional trade totals — is mined for candidates whose game category and freedom scores match an ideology’s profile, at zero extra API cost; (2) region rosters of a few large gameplay regions rotated by day, sampled randomly. Every candidate is then verified live by the classifier above before joining a pool, so coverage grows monotonically over time even when no one is using the site. Two scheduled mechanisms cooperate: a daily scan ahead of the visitor-facing rotation, and an hourly “tick” that runs a small, strictly paced verification step (a handful of requests at one per second — far below the NationStates rate limit and invisible to browsing) only while any ideology still lacks an example; the moment all ideologies are covered the tick becomes a no-op and the hunting stops automatically. The page fetches the current pools once per day and caches them locally; a row that still shows “—” simply has no verified example yet, which is common for the rarest and most census-gated ideologies. The labels reflect each nation’s profile at scan time and may evolve as the game does.

The table is generated from the live classifier and therefore cannot diverge from the code. The label is a statistical classification, not an evaluation: real nations blend ideologies, and each dossier’s explanation states which label was selected and why.

6. Economy report — methodology

Inputs. The report consumes the three official word ratings (Economy, Civil Rights, Political Freedom), the population in millions, the tax rate and the Administration share of the budget. Each rating word maps to a fixed numeric modifier taken from the NSEconomy base-value tables.

Model attribution. The report implements the NSEconomy economic model as documented at broomdces.com/nseconomy/calculations.php. The base values (original v0.7.0 NSEconomy) are credited to Commerce Heights. The black-market statistics originate with Sunset, whose ideas Tbone Steak implemented and extended with permission (published 2011). NSEconomy itself is written by Tbone Steak (© 2004–2026), with code contributions from Commerce Heights and Sunset. All modifiers and equations in this section are reproduced verbatim from that reference. The export formula requires totals over the entire region; the way a single-nation dossier obtains them is described under “Net trade” below.

Step 0 — word ratings to numeric modifiers

Representative values are listed below; the complete table is defined in code.

Reading: an “Economy: Good” rating corresponds to production of 7,500 per person-year and a consumer-confidence modifier of 0; “Powerhouse” corresponds to 25,000 and +0.02. Civil Rights and Political Freedom ratings feed their respective modifiers analogously (e.g. Civil Rights “Average” → 0, “Excellent” → +0.03, “Outlawed” → −0.08).

Step 1 — primary multipliers

Worker enthusiasm willingness to supply labour

Base value 1, modified by the civil-rights, political-freedom and tax terms. Formula: 1 + CivilRightsModifier + PoliticalFreedomModifier + TaxModifier. Tax contribution: 0% adds +0.02; each subsequent ten-point bracket decreases the modifier by approximately 0.02 (10% → 0…−0.02, 30% → −0.04…−0.06, 90% → −0.16…−0.18), interpolated within the bracket.

Consumer confidence propensity to spend and invest

Base value 1, raised by a strong economy rating and by civil rights, lowered by a weak economy and by restrictions. Formula: 1 + EconomyModifier + CivilRightsConfidenceModifier. The value participates in three downstream quantities: spending, imports and the black market.

Government efficiency fiscal effectiveness of the state

Base value 1, reduced by political restrictions and divided by the administration-budget curve. The curve equals 1 when Administration spending is exactly 30% of the budget; otherwise, with x = Administration % ÷ 100: curve = 1 − x × sign(x − 0.3) × ∛|x − 0.3| + 0.2x. Formula: 1 + PoliticalFreedomEfficiencyModifier ÷ AdminCurve.

Step 2 — sequence of derived aggregates

Each row of the report's ledger is produced in this order, every later quantity reusing the earlier ones. All of them use the entity’s live census words, tax rate and population read at the moment the dossier is opened.

Output annual production, in dollars

Production × Population × WorkerEnthusiasm × ConsumerConfidence

Consumption private spending net of tax

Output × (1 − TaxRate) — the share of production left in private hands after taxation. At typical tax rates (20–40%) this is the largest domestic component; at confiscatory rates (90%+) it collapses to a residual, because nearly all output is routed through the government instead. This is arithmetic, not an anomaly: the lower the tax, the larger the consumption share.

Government budget total revenue raised

Output × GovEfficiency × (TaxRate + ConsumerConfidence/10 + WorkerEnthusiasm/40) — the tax base constitutes the principal component; consumer confidence and worker enthusiasm contribute secondary components.

Government expenditures revenue applied to services

Budget × GovEfficiency. The difference between budget and expenditures constitutes government waste.

Imports

(1 ÷ ConsumerConfidence) ÷ 8 × (Consumption + GovernmentExpenditures) — the import propensity rises with consumption and falls as confidence rises.

Net trade exports minus imports, referenced to the region

Exports follow Exports = TotalImports × (Output ÷ TotalOutput), where both totals are summed over every nation of the region; NetTrade = Exports − Imports. These two region sums are the only figures in the report sourced from the official daily nations dump (nations.xml.gz) — exactly as the NationStates API documentation prescribes for region-wide aggregates, which are impractical to gather nation-by-nation. The dump is used for nothing else: every other line of the economy report (Output, Consumption, Budget, Imports, GDP, exchange rate, unemployment, black market) is computed from the live API of the entity whose dossier is open, so the dump's at-most-24-hour staleness affects only the region-wide reference totals of exports and imports, never the nation's own figures. A serverless endpoint downloads the dump at most once per day (the documentation asks scripts not to download it more frequently; a scheduled job pre-warms the cache at 04:00 UTC so a visitor almost never waits for the download), computes the totals of every region in a single streamed pass, and caches them so any region of any size resolves in one request. The report stays in a loading state (with a progress indicator, indeterminate while the server works) and is rendered once the totals arrive. When the endpoint is unreachable, the application falls back to requesting the export-related fields from each member nation — one minimal request per nation, paced to respect the API rate limit, without blocking the interface — and the provisional estimate Imports × (ConsumerConfidence − 1) is shown only if even that scan cannot collect usable data.

GDP and GDP per capita

GDP = Consumption + NetTrade + GovernmentExpenditures  ·  GDP/capita = GDP ÷ Population

Government waste revenue raised but not applied

GovernmentWaste = GovernmentBudget − GovernmentExpenditures. It is not displayed as its own ledger row, but it enters the report twice: as the shrinkage and full-waste terms of the black-market estimate (Step 3), and implicitly in GDP, since expenditures (not the budget) are the component that counts.

Exchange rate quoted in both directions

√(GDP/capita × Production ÷ 404,000,000) returns dollars per one unit of national currency — the value of the local unit expressed in dollars (NSEconomy's convention: a weak economy yields a small number, e.g. “1 Sain = $0.01”). The report quotes both directions: 1 local = X $ and 1 $ = 1/X local. The nation’s own currency name is used when available. Production is the per-person constant from the economy-rating table above.

Unemployment

(GDP/capita − 37,500)² × 1.25e-10 − 1.5e-6 × |GDP/capita − 37,500| + 0.03, floored at zero — a U-shaped function minimised near the reference income of 37,500.

Average income private income per inhabitant

GDP/capita × 0.8 — the per-capita product discounted by a fixed 20% depreciation/retention adjustment, following the convention used by NSEconomy. It is a presentation figure: GDP/capita remains the primary output measure, and this row re-expresses it as take-home income.

Step 3 — black market

Underground activity is modelled as the sum of five terms, each driven by the three multipliers and by the size of the state. First define BM Confidence = (GovEfficiency + WorkerEnthusiasm + ConsumerConfidence) ÷ 3 — an index of how well-run, motivated and self-confident the formal economy is. When this index falls below 1, the model treats the economy as one in which informal activity flourishes; when it is above 1, formal institutions suppress it.

1. Consumer fraud and theft grow — (1 ÷ BM Confidence) × 0.5 × Output. When confidence is poor, off-the-books trade expands, but is capped around half of one confidence-unit of output because even a criminal world struggles to absorb more.
2. Shrinkage under good governance — the same multiplier applied against government waste: (1 ÷ BM Confidence) × 0.75 × GovernmentWaste. A low-waste state leaves little public leakage to be recycled into underground profit.
3. Ethics offset — a small subtraction proportional to state effectiveness: (1 ÷ BM Confidence) × 0.75 × GovEfficiency, representing societies whose ethical climate keeps enterprises on the books.
4. Tax evasion and illegal enterprise — Output × 3 × TaxRate ÷ GovEfficiency. This is by far the dominant term at high tax rates: it scales with the entire output, triples for emphasis of the incentive effect, and explodes further as government efficiency drops, because a weak state cannot police evasion. It is the reason a 90%+-tax, low-efficiency state can show a black market larger than its entire official output.
5. Government waste — added in full, since money the state raises but fails to apply to services is, in this model, leakage available to the informal economy.

Black market = growth − shrinkage + evasion + waste (the ethics term being one of the subtractions). The net effect: high tax rates combined with weak institutions multiply the estimate dramatically, while a well-run, low-tax state keeps the shadow economy small — and the result may legitimately exceed the official GDP for confiscatory-tax regimes.

Worked example

Unit convention. Dollar figures use the comparative unit in which NationStates publishes income and GDP, so economies using different national currencies can be compared; this report retains that same unit throughout.

7. Regional dossiers and the comparison module

• Regional dossiers apply the same procedures to the region-level figures published by the API (averages, delegate, founder, officers, census). The compass and ideology of a region therefore describe its average member — representative for homogeneous regions, approximate for mixed ones.
• The comparison module fetches both entities in parallel and presents their compass positions, freedom profiles and key statistics side by side.

8. Spotlight entries (“Nation/Region of the day”)

The selection is deterministic and requires no authentication: the region is taken from a fixed list indexed by the day of the year, and the featured nation is that region’s roster entry at the same index. All users observe the same entries on a given day.