Roadmap / 0.1
The base game: four layers
Version 0.1, 5 Nov 2026. tag
Written at this version and not revised since. Later versions may change these numbers; see the roadmap.
v0.1, commit 1fb740f. File docs/RESULTS_v0.1_can_us.md.
Written at this version and not revised since. Later versions may change these numbers; see the roadmap.
These results come from 2,000 Monte Carlo runs per variant (36,000 runs in total), seed 20261105. Every run assumes the war has already started; the model gives the distribution of outcomes if it does, not the probability that it happens.
Reproduce with:
python -m forceonforce.cli run scenarios/can_us/*.yaml -n 2000
Backtests (must pass before results are read)
| Case | Model median (p10–p90) | Observed |
|---|---|---|
| Iraq 2003 invasion, US KIA | 157 (93–278) | ~139 (DoD, all causes to 1 May 2003) |
| Iraq 2003–11 occupation, US hostile KIA | 3,545 (2,879–4,325) | ~3,500 |
The occupation backtest is in-sample: the casualty rate was tuned to it. Out-of-sample validation is the first item on the roadmap.
Results
| Scenario | Variant | Kinetic days | End month | US KIA median (p10–p90) | Canadian deaths | US fiscal $bn | US GDP loss % | CAN GDP loss % | Troops needed / available | Outcomes % (consolidated / US withdraws / Canada capitulates / coercion succeeds / coercion abandoned / unresolved) |
|---|---|---|---|---|---|---|---|---|---|---|
| alberta_detachment | forward_defence · europe_backs_canada | 12 | 23 | 151 (59–403) | 5,408 | 48 | 1.7 | 16.5 | 37k / 375k | 17 / 47 / 32 / 0 / 0 / 4 |
| alberta_detachment | forward_defence · isolated | 12 | 21 | 119 (53–252) | 5,177 | 35 | 1.0 | 16.0 | 30k / 371k | 20 / 33 / 44 / 0 / 0 / 4 |
| alberta_detachment | preserve_force · europe_backs_canada | 11 | 22 | 164 (64–515) | 2,383 | 63 | 1.7 | 17.0 | 51k / 373k | 15 / 49 / 32 / 0 / 0 / 4 |
| alberta_detachment | preserve_force · isolated | 11 | 22 | 134 (57–333) | 2,258 | 55 | 1.0 | 16.6 | 43k / 371k | 19 / 36 / 42 / 0 / 0 / 4 |
| arctic_seizure | europe_backs_canada | 11 | 30 | 6 (3–17) | 349 | 2 | 1.4 | 12.1 | 1k / 207k | 23 / 37 / 29 / 0 / 0 / 11 |
| arctic_seizure | isolated | 11 | 29 | 5 (2–10) | 336 | 1 | 0.8 | 11.6 | 1k / 207k | 30 / 26 / 35 / 0 / 0 / 9 |
| economic_coercion | canada_cuts_oil · europe_backs_canada | 0 | 60 | 0 (0–0) | 0 | 0 | 2.4 | 14.3 | 0k / 284k | 0 / 0 / 0 / 16 / 23 / 61 |
| economic_coercion | canada_cuts_oil · isolated | 0 | 60 | 0 (0–0) | 0 | 0 | 1.7 | 14.4 | 0k / 284k | 0 / 0 / 0 / 20 / 19 / 60 |
| economic_coercion | oil_keeps_flowing · europe_backs_canada | 0 | 60 | 0 (0–0) | 0 | 0 | 1.8 | 14.2 | 0k / 284k | 0 / 0 / 0 / 16 / 19 / 66 |
| economic_coercion | oil_keeps_flowing · isolated | 0 | 60 | 0 (0–0) | 0 | 0 | 1.2 | 14.3 | 0k / 284k | 0 / 0 / 0 / 21 / 13 / 66 |
| full_annexation | manufactured_crisis_rally · forward_defence · europe_backs_canada | 27 | 40 | 2,609 (441–8,737) | 15,825 | 841 | 2.4 | 21.1 | 580k / 373k | 7 / 20 / 63 / 0 / 0 / 9 |
| full_annexation | manufactured_crisis_rally · forward_defence · isolated | 27 | 32 | 1,740 (390–6,097) | 12,308 | 664 | 1.8 | 20.6 | 498k / 371k | 10 / 12 / 71 / 0 / 0 / 7 |
| full_annexation | manufactured_crisis_rally · preserve_force · europe_backs_canada | 27 | 37 | 3,408 (638–10,679) | 16,123 | 799 | 2.4 | 21.6 | 713k / 373k | 3 / 25 / 63 / 0 / 0 / 9 |
| full_annexation | manufactured_crisis_rally · preserve_force · isolated | 27 | 31 | 2,567 (522–8,062) | 12,621 | 663 | 1.8 | 21.3 | 634k / 372k | 4 / 15 / 75 / 0 / 0 / 6 |
| full_annexation | no_casus_belli · forward_defence · europe_backs_canada | 27 | 15 | 1,069 (293–3,693) | 9,086 | 316 | 2.4 | 21.0 | 576k / 372k | 4 / 67 / 29 / 0 / 0 / 0 |
| full_annexation | no_casus_belli · forward_defence · isolated | 27 | 15 | 830 (275–2,751) | 8,158 | 316 | 1.8 | 20.6 | 493k / 372k | 4 / 58 / 37 / 0 / 0 / 0 |
| full_annexation | no_casus_belli · preserve_force · europe_backs_canada | 27 | 16 | 1,485 (380–4,674) | 7,654 | 327 | 2.4 | 21.5 | 712k / 372k | 2 / 67 / 30 / 0 / 0 / 1 |
| full_annexation | no_casus_belli · preserve_force · isolated | 27 | 16 | 1,248 (342–3,883) | 6,756 | 329 | 1.8 | 21.2 | 628k / 371k | 2 / 59 / 39 / 0 / 0 / 0 |
What the model says, and how much to trust each finding
- The conventional war is short and does not decide the outcome. It takes about 4 weeks to reach the national objectives. Canada's organized defence breaks well before that. This is robust: it holds at any plausible force ratio.
- The troop arithmetic doesn't close. Holding all of Canada at Quinlivan densities needs a median of roughly 490k–710k troops, against a sustainable US ceiling of about 370k without a draft. Control stays at 50–70% and the resistance feeds on the gap. This is robust: it is basic arithmetic on population and force structure.
- US public support at the outset is the single biggest driver. With no casus belli (polling at 2% support for force, rising to around 18% with a rally), the modal outcome is US withdrawal within about 15 months (58–67% of runs). With a manufactured crisis and a rally, the war lasts about 3 years, US deaths roughly double, and the modal outcome becomes Canadian capitulation (63–75%). Ranking runs by initial support moves the withdrawal rate from 64% in the bottom quarter to 8% in the top quarter. Armed-resistance propensity barely moves outcomes; it mostly moves casualties. The direction of this finding is robust. The capitulation share rests on the least-calibrated parameter in the model.
- Consolidated annexation is rare in every variant. It occurs in 2–10% of full-annexation runs and 15–30% of limited-aims runs. Insurgencies outlast decade-long horizons when the occupier is short of troops.
- Canada's strategy choice is a trade-off. Preserving the force (disengaging early and dispersing) cuts Canadian deaths by about 15% in the short full-scale war, makes no difference in the long (rally) war, and cuts them by about 55% in Alberta. It raises US deaths by roughly 30–50% and slightly lowers the chance of consolidation. It does not change who wins the conventional phase.
- European backing adds 8–14 points to US withdrawal and lowers Canadian capitulation. Most of the effect runs through sanctions and through the risk of a wider war if allied troops deploy.
- Limited aims are militarily cheap but not politically free. The Arctic seizure costs about 5 US KIA but still ends in US withdrawal or Canadian capitulation about two-thirds of the time. The economic damage of being at war with Canada is the main cost.
- Economic coercion alone usually leaves things unresolved after 5 years (60–66%). Canada takes about 14% annualized GDP loss; the US takes 1–2.4%. Cutting oil exports raises US pain but changes outcomes only modestly. The success rate (16–21%) is pinned by a parameter set from Pape's sanctions-failure finding, not fitted.
Parameters these results are most sensitive to (fix first)
politics.attacker_initial_support, political.capitulation_base_hazard, resistance.armed_propensity, and the economic coefficients. See docs/ROADMAP.md.