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
Ireland played 30 rounds from 1996 under the scripted method, with scripted events, 30 times. Its capability index moved from 11.93 to a median of 11.83 (−0.09), with 90 per cent of simulations between 11.82 and 11.85 [experiment_objectives].
Starting position
In 1996 the nation ranks 29 on the index against the frozen field, holds 1 of nine domains at Tier 1 and has a tier rank of 19. Its base income is 426 credits a year, calibrated so that its 2026 position is affordable with headroom for investment.
What was decided
A simulation spent a median of 2,344 credits on moves. The five groups that took most were Innovation & Commercialisation (build, 236 credits, 10.0 per cent); Social Infrastructure (build, 181 credits, 7.7 per cent); Financial Stability & Crisis Management (build, 142 credits, 6.0 per cent); Fiscal Capacity & Public Finance (build, 142 credits, 6.0 per cent); Monetary & Currency Capability (build, 140 credits, 6.0 per cent) [experiment_decisions_by_group]. By mode: build 77 per cent, partner 23 per cent [experiment_decisions_by_group]. Concentration of spending (HHI) had a median of 0.073, and the most repeated action appeared in 33 per cent of rounds.
The leading domain by period: rounds 1 to 5, Strategic Infrastructure; rounds 6 to 10, Financial Strength; rounds 11 to 15, Financial Strength; rounds 16 to 20, Production & Innovation; rounds 21 to 25, Trade & Investment; rounds 26 to 30, Strategic Infrastructure [experiment_spend_by_period].
What the world did
A median simulation saw 3 shocks and 1 upside events. The most damaging per firing were S02 (30 firings in 30 simulations, −33.3 points each); S01 (30 firings in 30 simulations, −32.9 points each); S10 (30 firings in 30 simulations, −28.9 points each) [experiment_shocks]. Upside events fired 30 times across all simulations [experiment_shocks].
The groups that lost most to shocks, per simulation: Fiscal Capacity & Public Finance −2.78; Investment Attraction & Projection −2.75; Export Capability & Sophistication −2.66; Economic Leverage & Dependence −2.64 [experiment_shock_damage_by_group].
The edges that carried most change through the graph, per simulation: Transport Infrastructure to Trade Facilitation & Logistics (+1.32 points); Fiscal Capacity & Public Finance to Social Infrastructure (+0.52 points); Fiscal Capacity & Public Finance to Force Generation & Mobilisation (+0.31 points) [experiment_propagation_by_edge].
Where it ended
At 2026 the median rank is 29 (5th to 95th percentile 29 to 29), the median count of Tier 1 domains 4 (4 to 4) and the median tier rank 5; the resilient score, the mean score sum over simulations, is 842 [experiment_objectives].
Tier 1 probability changed most in: human capital (0 to 100 per cent of simulations); information and influence (100 to 0 per cent of simulations); production and innovation (0 to 100 per cent of simulations); strategic infrastructure (0 to 100 per cent of simulations); trade and investment (0 to 100 per cent of simulations) [experiment_tiers_by_round].
Why
Without shocks the index ends +1.023 higher on average, so the shocks cost about 1.023 index points [experiment_probe_attribution].
Holding instead of following this method changes the index by −1.683: 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 +0.002 and +0.018, for example 2017 build Investment Attraction & Projection (26 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 +0.525 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
tier_jump (warn): largest gain in one domain within five years: 7 tiers (30 simulations above 2); tiers are reported, not an objective.
saturation_high (warn): 100% of simulations end with a group at 20 that did not start there (median 1 groups).
retired_words (warn): e4-irl/report.md: 'simulations' 2 time(s), first near line 39; use 'simulations'.
The simulation furthest from the median (seed 30) ends +0.02 from it, with 3 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 11.93 the model ends at 11.83, where the 2026 ratings put it at 12.08. Group scores miss by 2.07 points on average, against 1.32 for assuming nothing changed (skill −0.57); the change in each domain has the right sign in 56 per cent of domains. Furthest below history: trade and investment (−1.82); furthest above: national security (+1.81) [experiment_against_history]. The starting position is reconstructed, so part of the error is the reconstruction's.