Make this research yours. Add it to a free WorldbyFlow workbench to run follow-ups, ask questions, and re-check it as events move.

What the literature establishes

The published wargame literature converges qualitatively — across at least two independently designed and adjudicated exercise series (CSIS's 2023 wargame and CNAS's 2022/2023 wargames) — on the finding that specific preferred long-range strike munitions, most concretely LRASM, would be exhausted within roughly the first week of a Taiwan Strait contingency; it does not converge on a single quantitative 'rounds consumed' figure spanning strike missiles and air-defense interceptors together, and dedicated wargame-derived PAC-3/THAAD interceptor consumption rates are largely absent from the open published record, with the interceptor figures now circulating instead traced substantially to real-world 2025-2026 Middle East expenditure data rather than Taiwan wargame play.

Adjacent only

Asked: How many long-range strike and air-defense interceptor rounds do published wargames estimate a Taiwan Strait contingency would consume?

Studied: Individual wargames (CSIS 2023, CSIS/House Select Committee 2024, CNAS 2022/2023) each separately estimated three-week or first-week expenditure for specific named strike-munition types (LRASM, JASSM, JASSM-ER, Harpoon, Tomahawk) under their own baseline-inventory and scenario assumptions; none of the located primary sources published a matching wargame-derived consumption table specifically for air-defense interceptors (PAC-3, THAAD, SM-6, SM-3) at the same level of granularity, and the interceptor-depletion claims most prominent in current commentary draw on real-world 2025-2026 Middle East consumption data rather than Taiwan wargame outputs.

The strike-munitions half of the question (LRASM/JASSM/Tomahawk) is directly studied by name across multiple wargames with published figures; the interceptor half (PAC-3/THAAD) is adjacent only — secondary sources attribute an interceptor-exhaustion finding to CSIS's 2023 wargame, but the primary wargame reports available in this search emphasize offensive strike munitions, and the specific PAC-3/THAAD numbers circulating in 2025-2026 discourse are substantially anchored to real-world Iran-war and Red Sea expenditure rather than a Taiwan-scenario wargame table.

A moderate-sized but methodologically narrow corpus: a handful of named, unclassified wargame series run by two or three think tanks and one service-school press, each reporting its own game-specific outputs rather than a shared, standardized reporting format, making cross-study quantitative comparison difficult.

volume: Roughly four to six named, distinct wargame or tabletop-exercise series directly relevant to Taiwan munitions expenditure (CSIS 2023 invasion wargame with 24 iterations; CSIS 2025 blockade wargame with 26 iterations; House Select Committee/CSIS 2024 tabletop; CNAS 2022 Dangerous Straits; CNAS 2023 House Select Committee TTX; Air University 2023-2024 series), plus a smaller number of secondary industrial-base and interceptor-inventory analyses (Heritage TIDALWAVE, CSIS December 2025 interceptor report) that draw on but do not themselves run Taiwan-specific wargames. · window: 2022 (CNAS Dangerous Straits) through 2026 (CSIS's ongoing interceptor-inventory and blockade-wargame analysis). · designs: Manual, human-adjudicated tabletop and operational wargames with expert players (retired officers, think-tank staff, occasionally sitting legislators), not computational or statistical models; outputs are reported as game-specific totals or ranges across iterations rather than as data amenable to conventional statistical convergence testing. · funding: predominantly industry

Maturity: An active, policy-relevant but methodologically fragmented literature: individual wargame series are well-documented and transparent about their own assumptions, but there is no standardized cross-study reporting convention, no shared baseline-inventory dataset, and only one identified formal attempt (Air University) to replicate a prior study's scenario outcome — and that attempt targeted overall outcome, not the munitions-consumption tables themselves.

Searched: CSIS primary wargame reports and follow-on analyses, House Select Committee on the CCP press materials and simulation-results documents, CNAS wargame reports and congressional testimony, RAND Taiwan-related publications, Air University Press volume, Heritage Foundation TIDALWAVE model materials, and defense-trade press coverage (Naval News, Defense One, Air & Space Forces Magazine, CIMSEC, War on the Rocks).

The question as posed asks for a consumption rate — rounds per week or per contingency — that published wargames could in principle estimate directly. What the literature actually produced is narrower and more heterogeneous: CSIS's 2023 24-iteration wargame reported specific three-week expenditure totals for offensive strike munitions (JASSM, LRASM, Harpoon, Tomahawk) and a qualitative first-week exhaustion finding for LRASM specifically, while a distinct 2024 House Select Committee/CSIS tabletop simulation used a separate 2024-baseline inventory table and reported its own usage figures for LRASM, JASSM, JASSM-ER, Tomahawk, and Taiwan's anti-ship cruise missiles. These are two different exercises with different baseline years, inventory assumptions, and outputs, and public commentary has repeatedly blurred the two.

On the air-defense interceptor side specifically — PAC-3, THAAD, SM-6, SM-3 — the open wargame literature is thin. CSIS's 2023 report is cited by secondary sources (Defense One, Heritage) as concluding the U.S. would run out of PAC-3 MSE, THAAD, and SM-6/SM-3 stocks within the first week, but the primary wargame reports available in the open record emphasize offensive strike-munition expenditure (LRASM, JASSM, Tomahawk) far more granularly than defensive interceptor rounds; a dedicated, wargame-derived PAC-3/THAAD consumption-rate table comparable to the LRASM/JASSM tables was not found in the searched primary sources. The interceptor figures now circulating most heavily in 2025-2026 commentary — the Heritage Foundation's TIDALWAVE model and CSIS's own 2025 interceptor-inventory analysis — draw substantially on real-world consumption data from the 2025-2026 Middle East air-defense campaign (Red Sea Houthi engagements and the Iran war), not from Taiwan-specific wargame play, and CSIS itself frames its interceptor-depletion projections partly through that empirical lens rather than through the 2023 or 2024 Taiwan tabletop outputs alone.

Independent replication exists at the qualitative-outcome level but not at the quantitative-munitions level. Air University's Taiwan Deterrence Warfighting Advantage Research group ran a separate series of wargames in 2023-2024 explicitly intended to test whether CSIS's outcome could be replicated under different force structures; the published volume focuses on outcome replication (whether Taiwan falls, what force posture changes the outcome) rather than reproducing or cross-checking CSIS's specific munitions-count tables. CNAS ran its own 2022 "Dangerous Straits" wargame and a separate 2023 tabletop for the House Select Committee, both reaching a qualitatively convergent finding — that preferred long-range munitions would be exhausted early — using independent game design and adjudication from CSIS, which supports a convergent qualitative finding (rapid exhaustion of preferred long-range PGMs) but not a convergent quantitative figure, since CNAS's public reporting does not appear to publish a directly comparable round-count table.

The literature's own methodological caution further limits how these figures should be read: a U.S.-China Perception Monitor-style critique (reflected in defense-community commentary such as War on the Rocks' caution against false precision in wargame outputs) explicitly warns that wargames are not predictive models and that citing specific loss or expenditure statistics from them risks manufacturing a false sense of quantitative precision the underlying method cannot support. That caution applies with particular force to this question, since the object being asked about — a single reliable "rounds consumed" figure spanning strike missiles and air-defense interceptors — is not a figure any single published wargame set out to establish as a validated point estimate; each study reports its own game-specific totals under its own assumptions, and the totals vary by baseline year, force posture assumed, and whether the game covered an invasion or a blockade scenario.

Each finding graded by the convergence of the record behind it. One study is not a literature — convergence requires independent teams on independent data, named.

Genuine splits in the literature, with both sides’ studies named. A disagreement is a finding — never averaged into a middle the literature does not contain.

The named studies this review rests on, each with its replication record checked. A retraction lead is a keyword match to check — never a verdict.

Published wargames converge, across at least two organizationally independent series (CSIS and CNAS), on a qualitative finding that the United States would exhaust its preferred long-range strike munitions — most concretely LRASM — within roughly the first week of a Taiwan Strait contingency, with named point-estimate figures such as approximately 450 LRASMs and approximately 4,000 JASSMs expended over three weeks reported by the single most-cited CSIS wargame. The literature does not, however, converge on a single 'rounds consumed' figure spanning both strike missiles and air-defense interceptors together, and a dedicated wargame-derived PAC-3/THAAD consumption-rate estimate comparable to the strike-munitions tables was not found in the searched primary sources; the interceptor figures circulating most prominently in 2025-2026 discourse instead trace substantially to real-world Middle East expenditure data layered onto, rather than derived from, Taiwan-scenario wargame play.

Does CSIS's own 2023 wargame report contain an unpublished or appendix-level PAC-3/THAAD/SM-6 consumption table that did not surface in the sources located for this review, and if so, how does it compare to the strike-munitions figures that are publicly emphasized?

How much of the similarity between the CSIS 2023 wargame and the 2024 House Select Committee tabletop's findings reflects genuinely independent scenario dynamics versus shared adjudication assumptions carried across both CSIS-facilitated exercises?

Has any classified DoD-internal wargame (referenced obliquely in press accounts such as Defense One's July 2021 'It Failed Miserably' reporting) produced interceptor-consumption estimates that would, if declassified, either corroborate or contradict the open-source LRASM/JASSM-centric findings reviewed here?

The claims behind this analysis, each with its verification status — including what is contested, unverified, or could not be established.

September 3, 2026

September 3, 2026

September 3, 2026