STRATEGIC FORECAST & ESTIMATE
[SUBJECT / SITUATION - FORECAST TITLE - ENGAGEMENT REF]
The Strategic Forecast & Estimate is the long-horizon, multi-driver, calibrated forward assessment of how a situation, system, or strategic question will evolve over a defined future period. It is the flagship of the Strategic Forecasting & Early Warning node and answers a different question from its neighbours: not “what is the environment now” (the Regional Study) and not the sharp, bounded single-question judgment of the Intelligence Estimate (near/mid-term, one question, consumed here as an input or commissioned as a deeper look at one forecast line), but how a strategic situation is most likely to develop over the longer term, expressed as a set of calibrated probabilistic forecasts - explicit numeric point probabilities and tight ranges layered on the ICD 203 estimative bands, anchored to a reference class and base rates (the outside view), time-phased across the horizon, and tracked against resolvable questions and signposts so the forecast can be scored and updated. Its load-bearing analytic device is calibrated estimative probability (Good Judgment / Superforecasting discipline on top of the ICD 203 floor), and its collection logic is forecasting-question → driver → indicator/signpost → source. It does not generate the full set of distinct, internally-consistent alternative futures of the Alternative Futures Analysis (this product carries a calibrated central forecast plus a light alternative-outcomes treatment; the full 2×2 / cone-of-plausibility scenario set is deferred there); the wide-aperture weak-signal sweep for emerging and over-the-horizon issues of the Emerging Threats Report; the standing signpost-monitoring cadence of the Indicators & Warning (I&W) Program (retained service - this forecast defines the signposts; the ongoing watch against them is delivered there); or the live scoring/tracking surface of the Probabilistic Forecasting Dashboard (system). Where those needs surface, raise them as RFIs in §15 and escalate. This product delivers calibrated probabilistic forecasts; it is not a prediction, a guarantee, or a statement of fact about the future, and it is not legal, investment, security, or travel advice or a risk rating.
Document Control
| Field | Value |
|---|---|
| Report Reference | [REF-YYYY-###] |
| Date of Report | [YYYY-MM-DD] |
| Forecast As-Of Date | [YYYY-MM-DD - the date through which evidence is assessed] |
| Forecast Horizon | [e.g., 12–24 months / 2–5 years - and any time-phasing (near / mid / far)] |
| Classification / Handling | [CONFIDENTIAL - CLIENT EYES ONLY / TLP:AMBER] |
| Client | [CLIENT NAME] |
| Requesting Party | [CONTACT - ENGAGEMENT REF] |
| Forecast Subject / Question Set | [The strategic situation or question set forecast - e.g., “The trajectory of [state/sector/conflict/market] through [horizon]“] |
| Engagement Purpose | [Decision the forecast informs - e.g., strategic planning, posture-setting, investment thesis, capability/positioning, early-warning baseline] |
| Scope / Depth | [Long-horizon multi-driver calibrated forecast - see §2 scope] |
| Prepared By | [ANALYST NAME / ID] |
| Reviewed By | [REVIEWER NAME / ID] |
| Approving Officer | [APPROVER NAME / ID] |
| Version | [1.0] |
| Distribution | [NAMED RECIPIENTS] |
Handling & Legal Caveat
Handling: [Classification/TLP]. Disseminate only to the named authorized recipients. Reproduction or onward sharing prohibited without originator approval. Where the forecast references named individuals, it may contain personal data - store and transmit per the client data-processing agreement and applicable data-protection frameworks (GDPR, CCPA, and local equivalents).
Nature of this product (READ FIRST): This is a calibrated probabilistic forecast of how a situation may develop over a defined horizon, prepared to inform the client’s strategic planning - not a prediction, not a guarantee, not a certainty, and not a statement of fact about the future. Forecast probabilities are the firm’s independent, policy-neutral analytic judgments, calibrated to the stated scales; by design a well-calibrated forecast will be wrong a predictable fraction of the time, and probabilities are an aid to planning under uncertainty, not measurements. Likelihood and analytic confidence are stated separately, and plausible alternative outcomes are explicitly weighed. This is not legal advice, investment advice, security or travel advice, a risk rating, or a credit opinion. It is one analytic input; the client’s legal, security, financial, and in-country advisers should be relied on for operational and transactional determinations.
Analytic independence & policy neutrality (ICD 203): Forecasts here are the firm’s independent analytic judgment, reached on the evidence and free of policy advocacy, a client-preferred conclusion, or any government’s or party’s line. Reporting is distinguished from analyst judgment throughout; uncertainty drivers and change indicators are stated explicitly; alternative explanations and outcomes are tested, not suppressed; and the forecast is not adjusted to suit a preferred answer.
Calibration & accountability: Forecasts are framed as resolvable questions with resolution dates and criteria (§4, §8) so they can be scored after the fact (e.g., Brier-tracked). Numeric probabilities are anchored to a reference class and base rates (the outside view, §7) before case-specific adjustment (the inside view). The intent is honest calibration, not false precision - ranges widen where the evidence is thin.
Sourcing & legality: Findings derive from open, licensed, and lawfully accessed sources - public record, official statistics, reputable media, academic and institutional reporting, recognised forecasting/conflict/economic datasets used within licence, and (where commissioned) discreet in-country human sourcing - current as of the as-of date and graded (Annex A). No classified, proprietary-without-licence, or unlawfully obtained material is used; copyright and database-licence terms are respected. Official statistics and state-controlled media are treated as potentially self-interested and corroborated against independent sources.
Perishability: This is a point-in-time forecast. New events and information shift probabilities; the forecast decays as the horizon advances. Probabilities are time-sensitive - re-verify against current reporting and re-score before any consequential or time-critical decision, and commission an Indicators & Warning Program for standing signpost monitoring and forecast updating.
Reliance: Reliance is limited to the named client for the stated purpose; no third-party reliance without originator consent.
Forecast Snapshot
| Field | Value |
|---|---|
| Forecast Subject / Question Set | [The strategic situation or question set] |
| Geographic / Subject Bounds | [Country / region / sector / actor / system - the bounded target of the forecast] |
| Forecast Horizon | [Period covered, with any time-phasing] |
| Forecast As-Of Date | [YYYY-MM-DD] |
| Base-Line Product Used (if any) | [Reference to Regional Study or Intelligence Estimate used as input, or note if none] |
| Headline Forecast | [The single most decision-relevant calibrated forecast - point probability + ICD 203 band - see §8] |
| Principal Alternative Outcome | [The most plausible alternative, with its probability - see §9] |
| Dominant Driver | [The single driver that most shapes the forecast - see §6] |
| Pivotal Uncertainty / Threshold | [The uncertainty that most widens the range of outcomes - see §6/§9] |
| Highest-Priority Signposts | [The 1–3 signposts whose movement would most change the forecast - see §10] |
| Coverage Confidence | [HIGH / MODERATE / LOW - access, source diversity, currency - see §17] |
| Overall Forecast Confidence | [HIGH / MODERATE / LOW - evidence base, base-rate availability, model uncertainty - see §17] |
Table of Contents
- BLUF
- Executive Summary & Scope
- Key Judgments
- Forecasting Questions & Priority Intelligence Requirements (PIRs)
- Forecast Framing & Methodology
- Strategic Drivers & Trends
- Reference Class & Base Rates (The Outside View)
- The Forecast (Calibrated Outcome Set)
- Alternative Outcomes, Wildcards & Discontinuities
- Signposts & Warning Indicators
- Key Findings Summary
- Analysis of Competing Hypotheses (ACH)
- Key Assumptions Check (KAC)
- Premortem / Red-Team Check
- Collection Gaps & RFIs
- Assessment & Recommendations
- Annex A - Sources & Methodology
- Annex B - Appendices
(Page numbers populate on export to Word/PDF.)
1. BLUF
2–3 sentences. Lead with the headline forecast - the most likely outcome with its calibrated probability (point % + ICD 203 band) over the horizon - then the dominant driver or pivotal uncertainty shaping it, and the most decision-relevant implication for the client, with the recommended posture or next step. Written so the decision-maker can act on this line alone.
[BLUF]
2. Executive Summary & Scope
Triggering requirement and engagement purpose; why this situation is being forecast and for what client decision-horizon. Scope in/out stated explicitly - the subject/question set, the forecast horizon and time-phasing, the drivers in scope, and what is deferred to sibling/downstream products (full alternative-futures scenarios → Alternative Futures Analysis; weak-signal horizon scan → Horizon Scan; standing signpost monitoring → I&W Program; live tracking surface → Probabilistic Forecasting Dashboard; the bounded single-question estimate → Intelligence Estimate; the environmental baseline → Regional Study). Narrative synthesis of the headline forecast, the principal alternative, and the dominant drivers and uncertainties, to the ICD 203 floor - reporting separated from analytic judgment, the calibration basis named, alternatives weighed. State the headline forecast and its probability here.
[EXECUTIVE SUMMARY & SCOPE]
3. Key Judgments
The analytic bottom line - the headline forecasts, ordered by decision-relevance. Each judgment carries a calibrated likelihood (point probability + ICD 203 band) and analytic confidence as separate columns (never combined - ICD 203; likelihood is about the event, confidence is about the evidence base behind the probability) and a change-indicator stating what would shift the probability. The numeric probabilities are developed and reconciled in the forecast register (§8); keep the two consistent.
| # | Key Judgment (Forecast) | Likelihood | Analytic Confidence | Change Indicator (what would shift the probability) |
|---|---|---|---|---|
| KJ-1 | [e.g., [Outcome X] occurs within [horizon]] | [e.g., 65% - likely] | [HIGH/MOD/LOW] | [ ] |
| KJ-2 | [ ] | [e.g., 30% - unlikely] | [ ] | [ ] |
| KJ-3 | [ ] | [ ] | [ ] | [ ] |
| KJ-4 | [ ] | [ ] | [ ] | [ ] |
| KJ-5 | [e.g., The dominant driver of the forecast is [driver]; the pivotal uncertainty is [threshold event]] | [n/a - analytic statement] | [ ] | [ ] |
4. Forecasting Questions & Priority Intelligence Requirements (PIRs)
The collection-management spine for a forecast: each forecast is framed as a resolvable question (clear resolution criteria and a resolution date, so it can be scored) and tied to the drivers and indicators that inform it - forecasting-question → driver → indicator/signpost → source. State each question, its current probability, the key evidence, and analytic confidence. Summarize in the matrix.
- FQ-1 - [Primary forecast question]: [State as a resolvable question with explicit resolution criteria and date - e.g., “Will [event] occur by [date]?” / “Which of [outcomes A/B/C] obtains by [date]?“] [Current probability / key evidence / confidence]
- FQ-2 - [Second forecast question]: [ ]
- FQ-3 - [Third forecast question]: [ ]
- PIR-A - Driver intelligence: What is the current state and trajectory of the key drivers (§6), and what indicators reveal change? [ ]
- PIR-B - Base-rate intelligence: What is the appropriate reference class and historical base rate for each forecast question (§7)? [ ]
- PIR-C - Signpost intelligence: Which observable signposts most diagnostically confirm or disconfirm each forecast path (§10)? [ ]
- [Add engagement-specific forecasting questions / PIRs.]
| Question / PIR | Driver(s) | Current Probability / Answer | Confidence | Key Gap |
|---|---|---|---|---|
| FQ-1 | [ ] | [ ] | [H/M/L] | [ ] |
| FQ-2 | [ ] | [ ] | [H/M/L] | [ ] |
| FQ-3 | [ ] | [ ] | [H/M/L] | [ ] |
| PIR-A | [ ] | [ ] | [H/M/L] | [ ] |
| PIR-B | [ ] | [ ] | [H/M/L] | [ ] |
| PIR-C | [ ] | [ ] | [H/M/L] | [ ] |
5. Forecast Framing & Methodology
The load-bearing methodology section for a calibrated-forecast product - state the rules before any probability appears so the forecast is transparent and scorable. Define, in order:
- Forecast questions & resolution criteria. Each forecast framed as a resolvable question with an explicit resolution date and unambiguous resolution criteria, so it can be scored (e.g., Brier-tracked) after the fact.
- Horizon & time-phasing. The forecast horizon and any phasing (near / mid / far), and how probabilities are allowed to evolve across phases.
- Calibrated estimative probability. Numeric point probabilities and tight ranges layered on the ICD 203 seven bands (finer than the bands alone); likelihood (about the event) kept separate from analytic confidence (about the evidence base). Ranges widen where evidence is thin - calibration over false precision.
- Outside view → inside view. Anchor each forecast to a reference class and base rate (§7, the outside view) first, then adjust for case-specific factors (the inside view); state the adjustment and its rationale to avoid base-rate neglect.
- Driver & signpost logic. The forecast is built on identified drivers (§6) and tracked by signposts (§10); state how driver movement maps to probability change.
- Techniques applied. The structured techniques used (reference-class forecasting, ACH §12, Key Assumptions Check §13, premortem §14, and - where commissioned - quantitative or model-based methods); note any model and its uncertainty.
- Relationship to other products. This consumes the Regional Study baseline and/or Intelligence Estimate estimates; carries a light alternative-outcomes treatment but defers the full scenario set to Alternative Futures Analysis; defines signposts for the I&W Program; feeds the Probabilistic Forecasting Dashboard.
- Principal coverage constraints. Access, base-rate availability, reflexivity (the forecast or client action altering the outcome), and model uncertainty.
| Framing Element | Content |
|---|---|
| Forecast subject / question set | [ ] |
| Forecast horizon & time-phasing | [Near / mid / far - and the dominant phase] |
| Calibration approach | [Point % + range on ICD 203 bands; reference-class-anchored; confidence kept separate] |
| Outside-view reference basis | [The reference class(es) and base-rate source(s) used - see §7] |
| Techniques applied | [Reference-class forecasting / ACH / KAC / premortem / model - note model uncertainty] |
| Relationship to other products | [Consumes Regional Study/Intelligence Estimate / defers scenarios to Alternative Futures Analysis / defines signposts for I&W Program / feeds Probabilistic Forecasting Dashboard] |
| Principal coverage constraints | [Access, base-rate availability, reflexivity, model uncertainty] |
Calibrated likelihood scale (the load-bearing scale of this product) - numeric probability mapped to ICD 203 bands:
| ICD 203 band | Probability range | Typical point usage |
|---|---|---|
| almost no chance / remote | 01–05% | [ ] |
| very unlikely | 05–20% | [ ] |
| unlikely | 20–45% | [ ] |
| roughly even chance | 45–55% | [ ] |
| likely | 55–80% | [ ] |
| very likely | 80–95% | [ ] |
| almost certain | 95–99% | [ ] |
State a point probability and a range (e.g., “65%, range 55–75%”); never assert 0% or 100%. Keep likelihood separate from analytic confidence - a 60% forecast can be held at HIGH or LOW confidence depending on the evidence base.
6. Strategic Drivers & Trends
The forces that shape how the situation will develop: identify the key drivers (structural and proximate), their current state and trajectory, the direction and rate of change, and how they interact (reinforcing, offsetting, conditional). Distinguish slow-moving structural drivers from fast-moving proximate ones, and identify which drivers are the most decision-relevant and the most uncertain. This is the analytic foundation on which the forecast probabilities (§8) rest.
| Driver | Type | Current State | Trajectory & Rate of Change | Interaction with Other Drivers | Forecast Relevance | Uncertainty | Confidence | Source Grade |
|---|---|---|---|---|---|---|---|---|
| [ ] | [Structural / Proximate] | [ ] | [↑/→/↓ + rate] | [Reinforces / Offsets / Conditional on …] | [Dominant / Significant / Marginal] | [High / Med / Low] | [H/M/L] | [A–F/1–6] |
Driver synthesis: [The dominant driver(s), the most uncertain driver, and how the drivers combine to produce the forecast - the bridge from drivers to the §8 probabilities.]
7. Reference Class & Base Rates (The Outside View)
The outside view, applied before case-specific reasoning to avoid base-rate neglect and over-confidence. For each forecast question, define the reference class (the set of comparable past cases or situations), state the historical base rate of the outcome within that class, and note the quality and limits of the comparison. The base rate is the anchor; §8 then states the case-specific adjustment (the inside view) and its rationale.
| Forecast Question | Reference Class (comparable cases) | Base Rate of Outcome | Quality / Limits of Comparison | Anchor Probability (pre-adjustment) | Confidence | Source Grade |
|---|---|---|---|---|---|---|
| FQ-1 | [ ] | [e.g., outcome occurred in X of N comparable cases ≈ NN%] | [ ] | [ ] | [H/M/L] | [ ] |
| FQ-2 | [ ] | [ ] | [ ] | [ ] | [H/M/L] | [ ] |
| FQ-3 | [ ] | [ ] | [ ] | [ ] | [H/M/L] | [ ] |
Outside-view note: [Where a credible reference class exists vs. where the situation is genuinely novel (low base-rate confidence), and how that bounds the forecast.]
8. The Forecast (Calibrated Outcome Set)
The analytic spine of the product - the calibrated forecasts. For each forecast question, give the point probability and range, the ICD 203 band cross-walk, the analytic confidence (separate from likelihood), the base-rate anchor and the case-specific adjustment from it, the dominant supporting drivers, the time-phasing across the horizon, and the resolution criteria/date. Where a question has mutually exclusive outcomes, the probabilities across them should be coherent (sum to ~100% across an exhaustive set). Narrate the reasoning beneath the register so the probability is auditable, not asserted.
| Forecast Question / Outcome | Point Probability | Range | ICD 203 Band | Analytic Confidence | Base Rate → Adjustment | Dominant Drivers (§6) | Time-Phasing | Resolution Criteria & Date |
|---|---|---|---|---|---|---|---|---|
| FQ-1 - [outcome] | [e.g., 65%] | [55–75%] | [likely] | [H/M/L] | [base NN% → +/- adj because …] | [ ] | [near/mid/far] | [resolves [criteria] by [date]] |
| FQ-2 - [outcome] | [ ] | [ ] | [ ] | [H/M/L] | [ ] | [ ] | [ ] | [ ] |
| FQ-3 - [outcome] | [ ] | [ ] | [ ] | [H/M/L] | [ ] | [ ] | [ ] | [ ] |
Forecast reasoning: [The case for the central forecast(s) - how the drivers and base rates combine, why the probability is set where it is, what the calibrated range expresses, and how the probabilities phase over the horizon. State the headline forecast used in the BLUF and Snapshot.]
9. Alternative Outcomes, Wildcards & Discontinuities
The plausible alternatives to the central forecast, weighed honestly - the principal alternative outcome (the most likely way the central forecast is wrong) with its probability, and the low-probability / high-impact wildcards and discontinuities (shocks, black-swan-adjacent events, non-linear breaks) that would invalidate the forecast. This is a light alternative-outcomes treatment to bound the central forecast and flag tail risk; the full set of distinct, internally-consistent alternative futures is deferred to the Alternative Futures Analysis - flag where that is warranted.
| Alternative / Wildcard | Type | Probability | Impact if it Occurs | Leading Indicators | Confidence |
|---|---|---|---|---|---|
| Principal alternative | [Alternative outcome] | [ ] | [ ] | [ ] | [H/M/L] |
| Wildcard / discontinuity | [Low-prob / high-impact shock] | [e.g., <5%] | [Crit/High] | [ ] | [H/M/L] |
Alternatives note: [The most likely way the central forecast fails, the tail risks worth planning against, and whether a full Alternative Futures Analysis is recommended.]
10. Signposts & Warning Indicators
The observable signposts that confirm or disconfirm each forecast path - specific, monitorable signs whose appearance would raise or lower the probability of an outcome. Each signpost is tied to the forecast question/driver it bears on, the direction and magnitude of probability change it would signal, a lead time, and a disposition. These are the deliverable that seeds the Indicators & Warning (I&W) Program and the Probabilistic Forecasting Dashboard.
| # | Signpost (observable) | Forecast Question / Driver | Probability Change Signalled | Lead Time | Current Status | Disposition |
|---|---|---|---|---|---|---|
| 1 | [ ] | [ ] | [e.g., FQ-1 65%→80% if observed] | [ ] | [Not present / Emerging / Present] | [Watch / Notify client / Re-forecast / Route to I&W Program] |
Direction definitions: a confirming signpost raises the probability of the central forecast; a disconfirming signpost lowers it. Prioritise diagnostic signposts (those that discriminate strongly between outcomes) over merely consistent ones.
11. Key Findings Summary
Consolidated register of the load-bearing findings behind the forecast, each graded by evidentiary status and confidence so the reader can see what is established vs. assessed vs. contested. Materiality flags the findings that most drive the forecast probabilities.
| # | Finding | Status | Confidence | Materiality |
|---|---|---|---|---|
| 1 | [ ] | [Established / Assessed / Contested] | [H/M/L] | [ ] |
12. Analysis of Competing Hypotheses (ACH)
Apply to the central forecast question - array the competing outcomes as hypotheses, marshal the diagnostic evidence for/against each, and identify the most consistent outcome and the evidence that would overturn it. ACH disciplines the forecast against confirmation bias and anchoring; do not pad the apparatus where the evidence is one-sided - say so and close.
| Evidence / Indicator | H1: [Outcome A] | H2: [Outcome B] | H3: [Outcome C] |
|---|---|---|---|
| [ ] | [C/I/N] | [C/I/N] | [C/I/N] |
(C = consistent · I = inconsistent · N = neutral.) Most consistent outcome: [ ] - [rationale + the diagnostic evidence that would overturn it].
13. Key Assumptions Check (KAC)
The assumptions underpinning the forecast - about the reference class, the durability of current drivers, the continuity of current arrangements, the absence of exogenous shock, and the reliability of the data - each with its basis, confidence, and the impact on the forecast if it proves wrong. Linchpin assumptions (those that, if wrong, materially move the probabilities) are flagged.
| # | Assumption | Basis | Confidence | Impact on Forecast if Wrong | Linchpin? |
|---|---|---|---|---|---|
| 1 | [e.g., The chosen reference class is genuinely comparable] | [ ] | [H/M/L] | [ ] | [Y/N] |
| 2 | [e.g., The dominant driver continues its current trajectory] | [ ] | [ ] | [ ] | [ ] |
| 3 | [e.g., No major exogenous shock intervenes within the horizon] | [ ] | [ ] | [ ] | [ ] |
14. Premortem / Red-Team Check
A structured premortem to surface the ways the forecast could fail. Imagine it is the resolution date and the central forecast was badly wrong: what most plausibly happened, which assumption broke, which driver was mis-read, and what bias (anchoring, confirmation, base-rate neglect, recency, optimism) most likely contributed. Capture the corrective - the signpost or check that would have caught it - and feed it back to §10.
| # | Failure Mode (forecast wrong because…) | Assumption/Driver That Broke | Likely Bias | Corrective / Early Catch (→ §10) |
|---|---|---|---|---|
| 1 | [ ] | [ ] | [Anchoring / Confirmation / Base-rate neglect / Recency / Optimism] | [ ] |
Premortem note: [The most credible failure mode and the single check most worth installing against it.]
15. Collection Gaps & RFIs
Where coverage is thin or contested, the impact on the forecast, the collection that would close the gap, and where to escalate. Forecasting gaps are typically base-rate/reference-class availability, driver-state visibility, signpost observability, or a question needing a deeper bounded estimate.
| Gap | Forecast Question / Driver | Impact on Forecast | Recommended Collection | Escalation Target | Priority |
|---|---|---|---|---|---|
| [ ] | [ ] | [ ] | [ ] | [Intelligence Estimate / Scenario Analysis / Horizon Scan / I&W Program / In-country HUMINT] | [H/M/L] |
16. Assessment & Recommendations
16.1 Forecast Assessment
Overall assessment - the headline forecast and its probability, the principal alternative, the dominant drivers and pivotal uncertainties, the calibration basis and its limits, and the implications for the client’s stated purpose and decision horizon. State the figures used in the BLUF and Snapshot, and the confidence/coverage caveats that bound them. Be explicit about the irreducible uncertainty - what cannot be forecast and must be hedged or monitored instead.
[ASSESSMENT]
16.2 Recommendations
- For decision-makers / principals: [What the forecast means for the client’s stated purpose; the planning assumption the forecast supports; the no-regret moves robust across outcomes; the conditions under which the forecast should be revisited.]
- For planners / risk owners: [The forecasts to plan against, the hedges against the principal alternative and tail risks, and the signposts to put under watch.]
- For analysts (next intelligence steps): [Which questions warrant a deeper bounded estimate (Intelligence Estimate) or full scenario set ( Alternative Futures Analysis), which signposts to operationalise into an I&W Program, and the recommended re-forecast cadence.]
- Escalations / downstream products: [Items routed to the Intelligence Estimate, Alternative Futures Analysis, Horizon Scan, I&W Program, or Probabilistic Forecasting Dashboard.]
- Conditions for reliance / refresh: [The perishability window and the signposts/events (per §10) that should trigger a re-forecast before reliance on these probabilities.]
17. Annex A - Sources & Methodology
Collection methods and scope; the source register graded with the Admiralty two-axis code; the reference scales (below); statement of the likelihood-vs-confidence separation and the calibrated-probability method (numeric point + range on ICD 203 bands, reference-class-anchored, resolvable/scorable questions); the forecasting techniques applied (reference-class forecasting, ACH, KAC, premortem, and any model with its uncertainty); coverage and currency limitations (access, base-rate availability, reflexivity, model uncertainty); and the treatment of official statistics and state-controlled media as potentially self-interested and corroborated against independent sources.
Source reliability (Admiralty, A–F): A Completely reliable · B Usually reliable · C Fairly reliable · D Not usually reliable · E Unreliable · F Reliability cannot be judged.
Information credibility (Admiralty, 1–6): 1 Confirmed by other sources · 2 Probably true · 3 Possibly true · 4 Doubtful · 5 Improbable · 6 Truth cannot be judged. (Each sourced datum carries a two-character grade, e.g., B2.)
Estimative probability / likelihood (ICD 203): almost no chance / remote (01–05%) · very unlikely (05–20%) · unlikely (20–45%) · roughly even chance (45–55%) · likely (55–80%) · very likely (80–95%) · almost certain (95–99%).
Calibrated estimative probability (the load-bearing method of this product): numeric point probabilities and tight ranges are stated on top of the ICD 203 bands (finer than the seven bands alone), anchored to a reference class and base rate (the outside view) before case-specific adjustment (the inside view). Forecasts are framed as resolvable questions with resolution criteria and dates so they can be scored (e.g., Brier score) and the firm’s calibration tracked over time. Never assert 0% or 100%; ranges widen where evidence is thin. This is the Good Judgment / Superforecasting discipline layered on the ICD 203 floor - calibration over false precision.
Risk scoring (where a discrete risk is quantified - e.g., the impact of a §9 wildcard or discontinuity): Likelihood (1–5) × Impact (1–5) = 1–25; key: 1–5 Low · 6–10 Moderate · 11–15 Elevated · 16–20 High · 21–25 Critical. (Conditional - the forecasting spine is calibrated probability, above; this scale applies only where a discrete tail risk is scored for impact.)
Analytic confidence (evidence base - kept separate from likelihood): HIGH (multiple independent reliable sources, strong base-rate support, no significant contradiction) · MODERATE (some corroboration, gaps, partial base-rate support, minor unresolved inconsistency) · LOW (single/uncorroborated source, weak or absent reference class, significant gaps, plausible alternatives open). Never combine a likelihood term and a confidence level in the same sentence - a 60% forecast may be held at HIGH or LOW confidence.
Coverage confidence (product-level): HIGH (strong access, diverse independent sources, credible reference class and base rates, low reflexivity/model distortion) · MODERATE (adequate coverage with gaps, partial base-rate support, some reliance on official/contested sources) · LOW (significant access/base-rate limits, novel situation with weak comparison class, high reflexivity or model uncertainty, currency degraded).
Analytic-framework note: The forecast is built on driver/trend analysis and reference-class base rates, expressed as calibrated probabilities against resolvable questions, and tracked by diagnostic signposts; the collection logic is forecasting-question → driver → indicator/signpost → source. Probabilities are independent and policy-neutral per ICD 203 and are not adjusted to suit a client-preferred answer.
Searched sources (record source type, provider/title, as-of date, coverage scope, limitations, and any self-interest/reliability caveat): [TABLE]
18. Annex B - Appendices
- Appendix A - Forecast Question Register: every resolvable forecast question with resolution criteria, resolution date, current probability, and range - the basis for later scoring.
- Appendix B - Driver & Trend Register: the full §6 driver table with current state, trajectory, interactions, and confidence.
- Appendix C - Reference Class & Base-Rate Workbook: the comparison cases, base-rate derivations, and outside-view-to-inside-view adjustments behind §7–§8.
- Appendix D - Signpost & Warning Matrix: the §10 signposts with current status, probability-change signalled, lead time, and monitoring notes (the I&W hand-off, → I&W Program).
- Appendix E - ACH / KAC / Premortem Workpapers: the structured-technique workings behind §12–§14.
- Appendix F - Calibration & Track Record: prior forecasts on this subject, their resolutions, and the firm’s running calibration/Brier record where available.
- Appendix G - Time-Phasing Chart: the forecast probabilities plotted across the horizon phases.
- Appendix H - Full Source Register: every source, Admiralty grade, access date, reference, and any coverage/limitation note.
- Appendix I - Glossary & Abbreviations.
- Appendix J - Revision History.
END OF REPORT.
Verification disclaimer: This strategic forecast is a point-in-time, calibrated probabilistic assessment based on open, licensed, and lawfully accessed sources current as of the as-of date; it is not a prediction, a guarantee, or a statement of fact about the future, and it is not legal, investment, security, or travel advice or a risk rating. Forecast probabilities are the firm’s independent, policy-neutral analytic judgments calibrated to the stated scales and anchored to reference-class base rates; a well-calibrated forecast is expected to be wrong a predictable fraction of the time, and probabilities are an aid to planning under uncertainty, not measurements. Forecasts decay as the horizon advances and as new information arrives - probabilities are time-sensitive and should be re-scored against current reporting and signposts before any consequential or time-critical decision. Official statistics and state-controlled sources were treated as potentially self-interested and corroborated where possible. No classified, unlawfully obtained, or out-of-licence material was used in the production of this report.
Document control footer: [REF-YYYY-### · Version · Classification/TLP · Prepared/Reviewed/Approved · Distribution].
Model wiring
Generated from cell frontmatter at publish time.