Did the U.S. Destroy Iran's Missile Force? The Government Cannot Agree.

Two weeks after Operation Epic Fury formally concluded, a sharp dispute has broken into public view between the administration’s declared assessment of the campaign and what classified U.S. intelligence agencies are telling lawmakers about the same campaign. The question is specific and countable: how much of Iran’s ballistic missile capability survived the 66-day U.S.–Israeli air campaign? The administration says the answer is roughly 20 percent. U.S. intelligence assessments described to journalists say the answer is closer to 70 percent. A sitting U.S. senator says both numbers cannot be true.

This is an Open Questions piece. It does not allege wrongdoing. It records what each side has said, what evidence is in the public domain, what named analysts have argued, and what remains genuinely unknown.

What Happened

On May 12, 2026, the New York Times published a report by national-security correspondents Adam Entous, Maggie Haberman, and Jonathan Swan drawing on classified U.S. intelligence assessments. The core finding, as the paper described it: Iran has maintained “operational access” to approximately 30 of its 33 ballistic missile sites along the Strait of Hormuz, has retained roughly 70 percent of its prewar inventory of mobile launchers and missile stockpile, and has preserved about 90 percent of its underground storage and launch facilities.

The report landed against a public-facing backdrop that looked entirely different. President Trump had described Iran as “decimated” in public statements following the conclusion of Operation Epic Fury. Defense Secretary Pete Hegseth, testifying before the Senate Appropriations Defense Subcommittee during the week of May 12, told senators the campaign had “shattered” Iran’s missile and drone industrial base.

Two days after the Times report, on May 14, U.S. Central Command Commander Adm. Brad Cooper testified before the Senate Armed Services Committee. When senators asked about the Times’s 70-percent-retained figure, Cooper said that number was “not accurate” — but when pressed for specifics, he declined to provide them, citing the classified nature of battle damage assessments. He did say the campaign had “damaged or destroyed more than 85% of Iran’s ballistic missile, drone, and naval industrial base” across more than 1,450 strikes, and that Iran’s capability had been “significantly degraded.”

General Dan Caine, Chairman of the Joint Chiefs, was also asked about the Times report in open session. He declined to confirm or deny its findings, stating it would be “inappropriate for me to comment in this forum” on classified battle damage assessments.

Sen. Chris Murphy, Democrat of Connecticut and a member of the Senate Appropriations Defense Subcommittee, who has received classified briefings on the campaign’s results, said publicly during the hearing week: “That’s not what has been testified to us in private briefings” — a direct assertion that the administration’s public characterization diverges from what intelligence officials convey to Congress behind closed doors.

What Officials Say

President Trump described Iran as “decimated” in public statements following Operation Epic Fury’s conclusion, and asserted that roughly 80 percent of Iran’s missile capability had been destroyed.

Defense Secretary Pete Hegseth told the Senate Appropriations Defense Subcommittee that the campaign had “shattered” Iran’s missile and drone industrial base. The Pentagon’s press secretary characterized the Times report as “disgraceful” and suggested the paper was acting as “public relations agents for the Iranian regime.”

Adm. Brad Cooper, commander of CENTCOM, provided the most detailed on-the-record response in public testimony before the Senate Armed Services Committee on May 14. He said: “The Iranian ability to stop commerce has been dramatically degraded through the straits, but their voice is very loud, and those threats are clearly heard by the merchant industry and the insurance industry.” Cooper described the campaign as striking more than 1,450 targets, damaging or destroying “more than 85% of Iran’s ballistic missile, drone, and naval industrial base.” He said the Times’s 70-percent-retained figure was “not accurate,” but declined to say what the accurate figure is on grounds that the data is classified.

Gen. Dan Caine, Chairman of the Joint Chiefs, said: “All of our battle damage assessment matters are classified and it would be inappropriate for me to comment in this forum on that.” He neither confirmed nor denied the intelligence picture the Times described.

Sen. Chris Murphy, speaking with access to classified briefings: “That’s not what has been testified to us in private briefings.” Trump responded to the Times coverage by accusing reporter David Sanger of writing things that are “sort of treasonous.”

Evidence Available

Several facts are in the public record and are not in dispute.

The New York Times report of May 12 is a real published account, bylined by three of the paper’s most senior national-security correspondents, and it cites classified U.S. intelligence assessments as its source. The paper stands by its reporting. Its existence — and the administration’s sharp response — is itself a documented event.

CENTCOM Commander Cooper’s testimony on May 14 is also documented in full. His phrase “not accurate” is specific and on the record. His refusal to provide an alternative figure is equally on the record. The combination — a specific denial paired with a refusal to supply the underlying data — creates a structural asymmetry in the public debate: the administration can challenge the intelligence community’s numbers in public but cannot rebut them with the actual classified numbers without declassifying them.

The Washington Post reported on May 7 — five days before the Times piece — that CIA assessments concluded Iran possesses sufficient resources and operational infrastructure to sustain its hold on Strait of Hormuz shipping routes for months beyond the ceasefire period, based on remaining stockpile depth and distributed launcher positions. That report preceded and is independent of the Times’s May 12 account, but its conclusion — that Iran’s remaining capability is substantial — is directionally consistent.

Sen. Murphy’s public statement exists on the record. Murphy chairs a subcommittee with authority to receive classified executive-branch briefings on defense matters; his statement that private testimony diverges from public claims is attributable, named, and on the record.

What does not exist in the public domain: the underlying intelligence documents, the specific satellite imagery used to assess Strait of Hormuz site operability, the confidence ratings assigned by the producing agency, or any declassified version of the battle damage assessment.

What Critics Question

Named analysts with defense and intelligence backgrounds have publicly engaged the gap between the administration’s framing and the intelligence picture the Times described.

Brett McGurk, a CNN analyst who served as a senior National Security Council and State Department official under four consecutive presidents, characterized the situation in a May 13 CNN analysis as “a great strait stalemate” — a phrase emphasizing Iran’s continuing ability to threaten commercial shipping through the Strait even after the campaign. McGurk argued that Iran retains Shahed drone capability deployable “from anywhere in the Hormuz mountains,” meaning the threat is not confined to fixed missile sites. His analysis implies that the administration’s framing of a decisive degradation understates the distributed nature of what Iran retained.

Alex Plitsas, a CNN national security analyst, Atlantic Council senior fellow, and former Department of Defense official, addressed the question directly on CNN’s AC360 on May 13. Plitsas discussed the gap between the “decimated” language used by administration officials and what classified assessments indicate about retained stockpiles and operational site access. He framed the discrepancy as substantive rather than a matter of degree, noting that the specific figures in the Times report — 30 of 33 sites, 70 percent of stockpile — are granular enough to represent a genuine intelligence finding rather than a general estimate.

These are on-the-record analytical judgments by named experts, not findings of fact. Neither Plitsas nor McGurk had access to the underlying classified documents; they are assessing what the public record indicates. Their positions are notable because both have institutional backgrounds that make them familiar with how classified battle damage assessments are typically produced and communicated.

Competing Theories

Three explanatory frames are in circulation by named sources. None can be resolved on available public evidence.

The first, consistent with the administration’s position, is that the Times’s intelligence sourcing reflects either an inaccurate read of the underlying data or a selective leak by officials who oppose the administration’s Iran policy. Under this account, Cooper’s “not accurate” denial reflects a genuine factual rebuttal based on a fuller picture of battle damage than the Times’s sources had access to. The administration’s “85% of industrial base damaged or destroyed” and the “70% of stockpile retained” figures may also be measuring different things — industrial capacity versus deployed inventory — such that both could be partially correct without contradiction.

The second, consistent with Sen. Murphy’s statement and with the Times report, is that the administration is making public claims that exceed what its own intelligence agencies have verified. Under this account, the “not accurate” denial is a communications posture rather than a factual rebuttal, and the administration’s reluctance to declassify and publish the actual battle damage assessment reflects awareness that the underlying data does not support the strongest version of its public statements. This is Murphy’s direct implication.

The third, articulated implicitly by McGurk and Plitsas, is that the administration and the intelligence community may be assessing different dimensions of Iran’s capability. A 65-day campaign that destroyed most of Iran’s fixed missile production infrastructure — factories, assembly lines, hardened storage depots — could legitimately be described as having “shattered” the industrial base while leaving a substantial deployed inventory of completed missiles and mobile launchers intact. The two statistics — industrial damage and retained stockpile — would then describe different things rather than contradict each other, though the gap between “decimated” and “70% retained” in public-facing language would still be misleading.

Each of these frames is attributed to named sources or logically implied by the on-the-record positions of named officials. Each is consistent with some, but not all, of the publicly known facts. Given the classified nature of the underlying battle damage assessments, none can be definitively confirmed or ruled out.

What Remains Unknown

A list, with no inferences:

The specific confidence level and producing agency of the intelligence assessment the Times cited on May 12.
Whether Cooper’s “not accurate” denial reflects a methodological difference in what counts as “operational access” to a missile site, a genuinely different intelligence picture, or a political-communications posture.
How Cooper’s “85% of industrial base damaged or destroyed” claim relates to — and whether it contradicts — the Times’s “70% of prewar mobile launcher and stockpile inventory retained” figure. The two statistics may measure different things; they may also be incompatible. No official has explained the relationship publicly.
Whether the private congressional testimony that Murphy says diverges from the administration’s public statements aligns with the specific figures the Times published, or describes a different dimension of the campaign’s results.
Whether the Pentagon’s “disgraceful” characterization of the Times report reflects a substantive intelligence rebuttal — one the Pentagon could make publicly if it chose to — or a political-communications posture designed to discredit unwelcome reporting without engaging its substance.
Whether Iran’s retained capability constitutes a genuine operational threat to Strait of Hormuz shipping under current conditions, or whether the 30-of-33 site figure reflects accessibility without reflecting operational readiness.
Whether any declassification of the battle damage assessment is planned, and whether Congress’s classified oversight process will surface additional contradictions or resolutions.

The Bottom Line

What is confirmed: on May 12, the New York Times, citing classified U.S. intelligence assessments, reported that Iran retains operational access to 30 of 33 Strait of Hormuz missile sites, approximately 70 percent of its prewar mobile launcher and missile stockpile, and approximately 90 percent of its underground storage and launch facilities. On May 14, CENTCOM Commander Adm. Brad Cooper told the Senate Armed Services Committee that the 70-percent figure is “not accurate” but declined to provide the accurate figure on classification grounds. Gen. Caine declined to confirm or deny the underlying intelligence picture. Sen. Murphy — with access to classified briefings — stated publicly that the administration’s characterization does not match what intelligence officials convey to senators in private. Trump described Iran as “decimated” and accused a Times reporter of writing things that are “sort of treasonous.”

What is not confirmed: which account more accurately reflects the underlying classified battle damage assessments. The structural constraint is real. The administration can dispute the Times’s sourcing in public, but the information needed to definitively resolve the dispute — the actual classified battle damage assessment, with confidence levels — cannot be evaluated by the public, by most of Congress, or by any outside analyst. Cooper’s denial is on the record. The intelligence picture the Times described is also on the record. A sitting senator with clearances says the two are in tension. The underlying data that would settle the dispute remains classified. That asymmetry — not any imputed bad faith on either side — is why the question remains open.

What happens next: the Senate Intelligence Committee has the jurisdiction and the classified access to probe the contradiction Murphy described. Whether it pursues that inquiry, and whether its findings enter the public record, is not yet known. The IRGC’s operational posture in the Strait — and whether Iran’s retained capability is eventually exercised — may itself provide a real-world test of what the competing assessments actually imply.