Our methodology
SovRank scores each territory on a 0–1 000 scale of strategic sovereignty. Raw observations are normalised to 0–1, combined within dimensions and then across dimensions using softmax weighting, summed to a base score, and adjusted by circuit-breaker penalties before a floor at zero. Everything below is open: every indicator, weight, formula, data source, and aggregation rule that feeds the final number.
How Scores Are Computed
A raw observation passes through six stages — normalisation, two layers of aggregation, summation, and circuit-breaker deductions — to produce a final score between 0 and 1 000. Each stage is described in detail below.
- 1Raw Data
Per-indicator observations from accredited sources
accredited, publicly documented datasets
- 2Indicator Scores
Normalised to 0–1 via one of six strategies
linear · inverted · log · tier · banded · boolean
- 3
- 4
- 5Base Score
Sum of module points
module maxima sum to 1 000
- 6
Scoring Methods
Each raw observation is converted to a 0–1 score using one of six normalisation strategies. The strategy determines how the raw value maps to the score — linearly, with diminishing returns, in discrete tiers, or as a simple yes/no check. A seventh strategy, percentile, is available as a global mode override.
Linear
The engine maps the raw value linearly between a worst-case threshold (scores 0) and a best-case ideal (scores 1). Used for indicators where more is better at a constant rate — e.g. gross national savings as a share of GDP.
(raw − worst) / (best − worst)Worked example: raw value of 60 normalises to 0.6.
Inverted
The engine maps the raw value linearly between a best-case threshold (scores 1) and a worst-case ceiling (scores 0). Used for indicators where less is better — e.g. grid losses, energy import dependence.
1 − (raw − best) / (worst − best)Worked example: raw value of 30 normalises to 0.7.
Logarithmic
Higher raw values produce higher scores, but with diminishing returns. Logarithmic compression prevents extreme outliers (e.g. military budgets, patent counts) from compressing everyone else into a narrow band. Used for heavy-tailed absolute indicators.
(ln(raw) − ln(floor)) / (ln(ceiling) − ln(floor))Worked example: raw value of 100 normalises to 0.667.
Banded
The raw value is compared against ascending thresholds. The highest threshold it meets or exceeds sets the score. Used for indicators that fall naturally into discrete ranges.
highest satisfied threshold → mapped valueTiered
The raw value is a discrete category (a tier number). Each tier maps to a fixed score. Used for qualitative assessments — e.g. drone capability levels, strategic reserve adequacy.
lookup by tier key → mapped valueYes / No
A binary check: 1 if the territory has the attribute, 0 if it does not. Used for membership flags and presence/absence indicators — e.g. NATO membership, nuclear weapons, orbital launch capability.
1 if truthy, 0 if falsyHow Scores Combine
After each raw value is normalised to a 0–1 score, the engine combines scores in two steps: first within each dimension (grouping related indicators), then across dimensions to produce a module score. Module scores are summed to form the base score, then circuit-breaker penalties are deducted and the result is clamped at zero.
Within-dimension: Weighted Average
d = Σ(normalisedi × weighti) / Σ(weighti)Missing data is penalised: the indicator's weight counts in the denominator but contributes 0 to the numerator.
Each indicator in a dimension contributes proportionally to its weight. Higher-weight indicators have more influence on the dimension score. This is the default for all dimensions unless the module declares otherwise.
Used by 15 dimensions across all modules.
Within-dimension: Max (Substitutable)
d = MAX(normalisedi × weighti)Example: a territory in NATO carries a weight of 1.0. Even without CSTO (weight 0.65) or other alliances, the defence alliances dimension scores 1.0 × 1.0 = 1.0. One strong source is enough.
The indicator with the highest weighted product sets the dimension score; all others contribute nothing. Each indicator's weight acts as its inherent strength ceiling. This models substitutability: a territory needs only one strong source, not all of them.
Across-dimensions: Softmax
wi = exp(T · di) / Σj exp(T · dj)module_score = max_points × Σi wi · diThe temperature T controls how much the strongest dimension dominates. Higher T = more substitutable. Lower T = more complementary.
Dimensions are combined using softmax weighting with a per-module temperature T. A high temperature lets the strongest dimension dominate the module score; a low temperature forces all dimensions to contribute more equally.
Graph:combined module score versus the strongest dimension's score, with a second dimension fixed at 0.5. High T (red, T=2.0) hugs the diagonal — the strongest dimension dominates. Low T (green, T=0.7) sags toward the average.
Substitutable — the strongest dimension dominates
Moderate — the strongest dimension has meaningful influence
Moderate — the strongest dimension has meaningful influence
Complementary — all dimensions contribute roughly equally
Complementary — all dimensions contribute roughly equally
Example (Defense, T = 2.0): spending 0.8, headcount 0.5, deterrent 0.1, alliances 0.3 → softmax weights: Spending=0.46 Headcount=0.25 Deterrent=0.11 Alliances=0.17 → combined 0.559
Final score assembly
base_score = Σ(module points)final_score = max(0, base_score − Σ circuit_breaker_penalties)Module maxima are aspirational credit, not cohort-derived. A territory that hits every indicator's ideal threshold can reach 1 000 in principle.
Two of these penalties scale with modifier columns (conflict intensity, sanctions breadth), so the effective deduction can exceed the base values shown above. The final score is clamped at zero.
A territory earns 540 base points and triggers one -100 circuit breaker: 540 - 100 = 440.
Another territory earns 120 base points and triggers a -300 circuit breaker: max(0, 120 - 300) = 0. The floor prevents negative scores.
Indicators by Module
How well a territory can defend itself. A territory is secure if it is strong in any one of: military spending, troop numbers, nuclear weapons, defence alliances (NATO, US allies, CSTO, and others), drone capability, or the protection of a guarantor power for smaller territories under treaty. Higher scores mean greater security.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| NATO Membership | Defence alliances | ×1 | Yes / No | 2026 | |
| CSTO Membership | Defence alliances | ×0.65 | Yes / No | 2026 | |
| Major US Ally (non-NATO) | Defence alliances | ×0.85 | Yes / No | 2026 | |
| Five Power Defence Arrangements | Defence alliances | ×0.5 | Yes / No | 2026 | |
| EU Mutual Defence Clause (Art. 42.7) | Defence alliances | ×0.7 | Yes / No | 2026 | |
| ECOWAS Mutual Defense Protocol | Defence alliances | ×0.5 | Yes / No | 2026 | |
| SADC Mutual Defense Pact | Defence alliances | ×0.4 | Yes / No | 2026 | |
| France Defense Treaty (Bilateral) | Defence alliances | ×0.6 | Yes / No | 2026 | |
| Rio Pact (TIAR) Membership | Defence alliances | ×0.55 | Yes / No | 2026 | |
| GCC Defense Agreement (Peninsula Shield) | Defence alliances | ×0.5 | Yes / No | 2026 | |
| Regional Security System (RSS) Membership | Defence alliances | ×0.35 | Yes / No | 2026 | |
| Arab League Joint Defence Treaty | Defence alliances | ×0.35 | Yes / No | 2026 | |
| Defensive Drone Capability (Gen 1) | Drone capability | ×1 | Tiered | SIPRI Arms Transfers2026 assessment | 2026 |
| Defensive Drone Capability (Gen 2) | Drone capability | ×1 | Tiered | OSINT + SIPRI Arms Transfers2026 assessment | 2026 |
| Military Expenditure (Absolute) | Military spending | ×1.5 | Logarithmic | 2025 | |
| Active Military Personnel | Troop numbers | ×1 | Logarithmic | 2025 | |
| Nuclear Weapons | Nuclear weapons | ×1.5 | Yes / No | 2026 |
How advanced a territory's technology and AI capabilities are — including research, digital infrastructure, talent, and the AI industry. Higher scores mean greater competitiveness in the AI era.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| Secure Internet Servers | Compute and AI infrastructure | ×0.6 | Logarithmic | 2024 | |
| Semiconductor Fabrication Capacity | Compute and AI infrastructure | ×1.5 | Logarithmic | 2024 | |
| Independent Orbital Launch Capability | Compute and AI infrastructure | ×0.6 | Yes / No | 2026 | |
| Frontier Compute Access | Compute and AI infrastructure | ×1.3 | Logarithmic | 2026 | |
| AI Infrastructure | Compute and AI infrastructure | ×1.5 | Linear | 2024 | |
| Patent Applications (Resident) | Talent and research | ×0.8 | Logarithmic | 2021 | |
| Research & Development Expenditure | Talent and research | ×1.2 | Linear | 2024 | |
| Highly-Skilled Talent Flow (Brain Drain) | Talent and research | ×0.8 | Linear | 2024 | |
| AI Talent | Talent and research | ×1.5 | Linear | 2024 | |
| AI Research | Talent and research | ×1 | Linear | 2024 | |
| AI Operating Environment | AI ecosystem | ×0.4 | Linear | 2024 | |
| AI Development | AI ecosystem | ×1 | Linear | 2024 | |
| AI Government Strategy | AI ecosystem | ×0.6 | Linear | 2024 | |
| AI Commercial | AI ecosystem | ×0.8 | Linear | 2024 |
Energy Sovereignty
150 ptsStrongest area dominatesT = 1.5Low-carbon generation, Fuel security, Grid interconnection use strongest-onlyHow reliably a territory can power itself without depending on hostile suppliers. A territory is energy-secure if it has any one of: domestic energy production, nuclear or renewable generation, fuel stockpiles, guaranteed import contracts, grid links to stable neighbours, or data-centre hosting capacity. Higher scores mean greater energy independence.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| Renewable Electricity Share | Low-carbon generation | ×1 | Linear | 2024 | |
| Nuclear Electricity | Low-carbon generation | ×1 | Linear | 2024 | |
| Strategic Petroleum Reserve | Fuel security | ×1 | Tiered | 2026 | |
| Gas Storage Depth | Fuel security | ×0.7 | Tiered | 2026 | |
| Guaranteed Energy Import Treaty | Fuel security | ×0.6 | Yes / No | 2026 | |
| Synchronous Grid Connectivity | Grid interconnection | ×1 | Tiered | 2026 | |
| Strongest Grid Interconnect | Grid interconnection | ×0.8 | Tiered | 2026 | |
| Energy Imports | Domestic energy production | ×0.8 | Inverted | 2023 | |
| Grid Transmission & Distribution Losses | Energy infrastructure | ×1 | Inverted | 2024 | |
| Electricity Consumption per Capita | Energy infrastructure | ×1 | Logarithmic | 2024 | |
| Data-Centre & Energy Capacity | Data-centre capacity | ×0.8 | Logarithmic | 2024 |
The overall strength of the economy — its size, wealth, government finances, debt sustainability, monetary credibility, and resilience to external shocks. A territory scores well if it is strong in any major area. Higher scores mean a larger, more resilient economy.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| Current Account Balance | External resilience | ×1.2 | Linear | 2031 | |
| FX Reserves (Import Cover) | External resilience | ×1.3 | Linear | 2024 | |
| Export Diversification | External resilience | ×1 | Linear | 2024 | |
| Critical Minerals & Rare Earth Control | External resilience | ×0.7 | Logarithmic | 2024 | |
| Food & Caloric Independence | External resilience | ×0.6 | Linear | 2024 | |
| GDP (current US$) | Economic size and wealth | ×2 | Logarithmic | 2024 | |
| GDP per Capita (PPP) | Economic size and wealth | ×2 | Logarithmic | 2024 | |
| Real GDP Growth | Economic size and wealth | ×0.5 | Linear | 2024 | |
| Unemployment | Economic size and wealth | ×1 | Inverted | 2031 | |
| Debt-to-GDP Ratio | Fiscal health | ×1.5 | Banded | 2031 | |
| Primary Fiscal Balance | Fiscal health | ×1 | Banded | 2031 | |
| Inflation | Money and markets | ×1.1 | Inverted | 2024 | |
| Reserve Currency | Money and markets | ×1.2 | Tiered | 2026 | |
| Domestic Market Depth | Money and markets | ×0.8 | Linear | 2024 |
The quality of a territory's institutions and its influence on the world stage — rule of law, political stability, corruption control, cultural influence, and seats in major international forums (UN Security Council, G7, G20, OECD, BRICS+, Commonwealth). A territory needs only one major forum seat to score well on influence. Higher scores mean better governance and greater global influence.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| UN Security Council Permanent Seat | Diplomatic influence | ×1 | Yes / No | 2026 | |
| G7 Membership | Diplomatic influence | ×0.85 | Yes / No | 2026 | |
| G20 Membership | Diplomatic influence | ×0.65 | Yes / No | 2026 | |
| OECD Membership | Diplomatic influence | ×0.5 | Yes / No | 2026 | |
| BRICS+ Membership | Diplomatic influence | ×0.45 | Yes / No | 2026 | |
| Commonwealth Membership | Diplomatic influence | ×0.3 | Yes / No | 2026 | |
| Rule of Law | Institutions | ×1.2 | Banded | 2024 | |
| Political Stability / Absence of Violence | Institutions | ×0.9 | Banded | 2024 | |
| Control of Corruption | Institutions | ×1.1 | Banded | 2024 | |
| Soft Power | Global influence | ×0.8 | Linear | 2024 |
Planned Indicators
These indicators are defined but not yet in the scoring model. They will be activated once accredited data sources are available.
| Indicator | Subtype | Strength | Type | Source | Year |
|---|---|---|---|---|---|
| Net International Investment Position | External resilience | ×1 | Linear | Source not specified | — |
| Gross National Savings | External resilience | ×0.7 | Linear | — | |
| Economic Complexity Index | Economic size and wealth | ×1 | Linear | — | |
| Interest-to-Revenue Ratio | Fiscal health | ×1 | Inverted | Source not specified | — |
| Old-Age Dependency Ratio | Fiscal health | ×0.5 | Inverted | — | |
| Revenue-to-GDP Ratio | Fiscal health | ×2 | Banded | — | |
| Interest Burden | Fiscal health | ×1 | Inverted | Source not specified | — |
| GDP per Capita (development check) | Fiscal health | ×2 | Banded | — | |
| Inflation Volatility | Money and markets | ×1 | Inverted | — |
Modifiers by Group
Circuit Breakers
4 modifiersBoolean flags that subtract points from the base score when triggered. Each carries a fixed penalty defined in the modifier registry.
| Modifier | Type | Effect / Scope | Source | Year |
|---|---|---|---|---|
| Active Sovereign Default | boolean | -300 pts | 2026 | |
| Active War on Home Territory | boolean | -100 pts · scales with conflict_intensity | 2026 | |
| Capital Controls Imposed | boolean | -100 pts | 2026 | |
| Active International Sanctions | boolean | -80 pts · scales with sanctions_breadth | 2026 |
Data Sources
30 modifiersModifier columns that provide raw data to a scored indicator.
| Modifier | Type | Effect / Scope | Source | Year |
|---|---|---|---|---|
| Reserve Currency Status | categorical | Feeds reserve_currency | 2026 | |
| Conflict Intensity | categorical | — | 2026 | |
| Independent Orbital Launch | boolean | Feeds orbital_launch_capability | 2026 | |
| Sanctions Breadth | categorical | — | 2026 | |
| NATO Defense Pact (Scored) | boolean | Feeds nato_membership | 2026 | |
| CSTO Defense Pact (Scored) | boolean | Feeds csto_membership | 2026 | |
| Major US Ally Defense Pact (Scored) | boolean | Feeds us_ally_membership | 2026 | |
| FPDA Membership (Scored) | boolean | Feeds fpda_membership | 2026 | |
| EU Mutual Defence Clause (Scored) | boolean | Feeds eu_mutual_defense | 2026 | |
| ECOWAS Mutual Defense Protocol (Scored) | boolean | Feeds ecowas_mutual_defense | 2026 | |
| SADC Mutual Defense Pact (Scored) | boolean | Feeds sadc_mutual_defense | 2026 | |
| France Defense Treaty (Scored) | boolean | Feeds france_defense_treaty | 2026 | |
| Rio Pact (TIAR) Membership (Scored) | boolean | Feeds rio_pact_membership | 2026 | |
| GCC Defense Agreement (Scored) | boolean | Feeds gcc_defense_pact | 2026 | |
| Regional Security System (RSS) Membership (Scored) | boolean | Feeds rss_membership | 2026 | |
| Arab League Joint Defence Treaty (Scored) | boolean | Feeds arab_league_joint_defense | 2026 | |
| Defensive Drone Capability Gen 1 | categorical | Feeds drone_capability_gen1 | SIPRI Arms Transfers2026 assessment | 2026 |
| Defensive Drone Capability Gen 2 | categorical | Feeds drone_capability_gen2 | OSINT + SIPRI Arms Transfers2026 assessment | 2026 |
| UNSC P5 Permanent Seat (Scored) | boolean | Feeds unsc_p5_seat | 2026 | |
| G7 Membership (Scored) | boolean | Feeds g7_seat | 2026 | |
| G20 Membership (Scored) | boolean | Feeds g20_seat | 2026 | |
| OECD Membership (Scored) | boolean | Feeds oecd_seat | 2026 | |
| BRICS+ Membership (Scored) | boolean | Feeds brics_seat | 2026 | |
| Commonwealth Membership (Scored) | boolean | Feeds commonwealth_seat | 2026 | |
| Strategic Petroleum Reserve (Scored) | categorical | Feeds strategic_petroleum_reserve | 2026 | |
| Gas Storage Depth (Scored) | categorical | Feeds gas_storage_depth | 2026 | |
| Guaranteed Energy Import Treaty (Scored) | boolean | Feeds guaranteed_import_treaty | 2026 | |
| Synchronous Grid Connectivity (Scored) | categorical | Feeds synchronous_grid_tier | 2026 | |
| Strongest Grid Interconnect (Scored) | categorical | Feeds strongest_interconnect | 2026 | |
| Nuclear-Armed State | boolean | Feeds nuclear_weapons | 2026 |
Memberships
21 modifiersBoolean flags that group territories into analytical blocs — e.g. NATO members, EU member states.
| Modifier | Type | Effect / Scope | Source | Year |
|---|---|---|---|---|
| NATO Member | boolean | Bloc nato | 2026 | |
| EU Member | boolean | Bloc euro_area | 2026 | |
| Major US Ally (non-NATO) | boolean | Bloc us_allies | 2026 | |
| OECD Member | boolean | Bloc oecd | 2026 | |
| BRICS+ Member | boolean | Bloc brics_plus | 2026 | |
| OPEC Member | boolean | Bloc opec | 2026 | |
| UN Security Council Permanent Member | boolean | Bloc unsc_p5 | 2026 | |
| UN Founding Member | boolean | Bloc un_founding | 2026 | |
| NATO Founding Member | boolean | Bloc nato_founding | 2026 | |
| EU Founding Member | boolean | Bloc eu_founding | 2026 | |
| G7 Member | boolean | — | 2026 | |
| G20 Member | boolean | Bloc g20 | 2026 | |
| Commonwealth Member | boolean | Bloc commonwealth | 2026 | |
| EU Candidate Country | boolean | Bloc eu_candidate | 2026 | |
| ASEAN Member | boolean | Bloc asean | 2026 | |
| SCO Member | boolean | Bloc sco | 2026 | |
| OAS Member | boolean | Bloc oas | 2026 | |
| African Union Member | boolean | Bloc au | 2026 | |
| GCC Member | boolean | Bloc gcc | 2026 | |
| CSTO Member | boolean | Bloc csto | 2026 | |
| International Criminal Court Member | boolean | Bloc icc | 2026 |
Risk Flags
4 modifiersBoolean risk signals displayed for analytical context. No direct score impact yet.
Strengths
3 modifiersBoolean strength signals displayed for analytical context. No direct score impact yet.
Classifications
5 modifiersUN or institutional territory classification, used for display grouping only.