Batch cycle-003. Model 0.2.0, graph 0.2.0, parameters 0.1.1. 30 simulations, seeds 1 to 30; probes on 10. Every figure is a distribution over simulations of a structural model with seed parameter values; it shows what the model implies under these settings, not what will happen. Other nations are frozen at their 2026 ratings. No reconstructed historical data is used in this experiment.
Headline
United Kingdom played 30 rounds from 1996 under the scripted method, with scripted events, 30 times. Its capability index moved from 17.25 to a median of 17.52 (+0.27), with 90 per cent of simulations between 17.52 and 17.53 [experiment_objectives].
Starting position
In 1996 the nation ranks 2 on the index against the frozen field, holds 8 of nine domains at Tier 1 and has a tier rank of 2. Its base income is 2,095 credits a year, calibrated so that its 2026 position is affordable with headroom for investment.
What was decided
A simulation spent a median of 9,661 credits on moves. The five groups that took most were Advanced Education (build, 750 credits, 7.8 per cent); Land Power (build, 544 credits, 5.6 per cent); Intelligence & Assessment (build, 544 credits, 5.6 per cent); Joint Command, Projection & Sustainment (build, 544 credits, 5.6 per cent); Air Power (build, 544 credits, 5.6 per cent) [experiment_decisions_by_group]. By mode: build 91 per cent, partner 9 per cent [experiment_decisions_by_group]. Concentration of spending (HHI) had a median of 0.047, and the most repeated action appeared in 60 per cent of rounds.
The leading domain by period: rounds 1 to 5, National Security; rounds 6 to 10, National Security; rounds 11 to 15, National Security; rounds 16 to 20, Human Capital; rounds 21 to 25, Trade & Investment; rounds 26 to 30, Critical Technology [experiment_spend_by_period].
What the world did
A median simulation saw 5 shocks and 1 upside events. The most damaging per firing were S10 (30 firings in 30 simulations, −21.6 points each); S02 (60 firings in 30 simulations, −17.4 points each); S01 (30 firings in 30 simulations, −16.5 points each) [experiment_shocks]. Upside events fired 30 times across all simulations [experiment_shocks].
The groups that lost most to shocks, per simulation: Labour & Employment −2.58; Investment Attraction & Projection −2.57; Fiscal Capacity & Public Finance −2.51; Trade Facilitation & Logistics −2.37 [experiment_shock_damage_by_group].
The edges that carried most change through the graph, per simulation: Foundational Education to Advanced Education (+0.79 points); Fiscal Capacity & Public Finance to Social Infrastructure (+0.42 points); Fiscal Capacity & Public Finance to Force Generation & Mobilisation (+0.34 points) [experiment_propagation_by_edge].
Where it ended
At 2026 the median rank is 2 (5th to 95th percentile 2 to 2), the median count of Tier 1 domains 8 (8 to 8) and the median tier rank 2; the resilient score, the mean score sum over simulations, is 1,279 [experiment_objectives].
Tier 1 probability changed most in: critical technology (0 to 100 per cent of simulations); production and innovation (100 to 0 per cent of simulations) [experiment_tiers_by_round].
Why
Without shocks the index ends +0.825 higher on average, so the shocks cost about 0.825 index points [experiment_probe_attribution].
Holding instead of following this method changes the index by −1.435: that is the method's worth against doing nothing [experiment_probe_attribution].
Removing any single one of the five largest actions changes the index by between −0.012 and 0.000, for example 2023 build National Resilience & Crisis Management (44 credits) [experiment_probe_attribution]. The largest actions by credits come late, when income is highest, and are rarely the most valuable.
Removing the largest shock of a simulation raises the index by +0.243 on average [experiment_probe_attribution].
The replay probe reproduces the simulations exactly (index change 0.000000), so the attributions are clean [experiment_probe_attribution].
Surprises
saturation_high (warn): 100% of simulations end with a group at 20 that did not start there (median 4 groups).
retired_words (warn): e4-gbr/report.md: 'simulations' 1 time(s), first near line 39; use 'simulations'.
The simulation furthest from the median (seed 19) ends +0.01 from it, with 5 shocks and 1 upside events [experiment_outlier_simulations].
Comparison
This is the baseline (hold) experiment, or there is none for this nation in the batch.
Against history: from a reconstructed index of 17.25 the model ends at 17.52, where the 2026 ratings put it at 16.45. Group scores miss by 1.76 points on average, against 1.22 for assuming nothing changed (skill −0.44); the change in each domain has the right sign in 33 per cent of domains. Furthest below history: production and innovation (−0.42); furthest above: national security (+2.88) [experiment_against_history]. The starting position is reconstructed, so part of the error is the reconstruction's.