Making Sense of Iran's Deterrence Failures

An early appraisal of the missteps that got Iran to this point.

The joint U.S.-Israeli strikes that began on February 28 (Operation Epic Fury and Operation Roaring Lion) have killed Iranian Supreme Leader Ali Khamenei, decapitated much of Iran’s senior military and intelligence leadership, and plunged the country into the worst crisis it has faced since at least the Iran-Iraq War (if not the founding of the Islamic Republic itself).¹ Strikes have hit targets across at least 24 of Iran’s 31 provinces. Iran is retaliating, launching missiles at Israel, at U.S. bases across the Gulf, and at civilian airports in the UAE and Kuwait, while threatening to close the Strait of Hormuz. Three American service members are already dead.

An F/A-18E Super Hornet, attached to Strike Fighter Squadron (VFA) 14, prepares to make an arrested landing on the flight deck of Nimitz-class aircraft carrier USS Abraham Lincoln (CVN 72) in support of Operation Epic Fury, March. 1, 2026. (U.S. Navy photo)

It is far too early to know how this ends and that’s not the purpose of this post (I am writing on Sunday night from D.C.). What I wanted to comment on was Iran’s catastrophic approach to deterrence over the last year or so. Iran’s deterrence strategy failed comprehensively: it failed last year with the 12 Day War, and it failed again 48 hours ago. The causes, as I see it, are a series of compounding errors. I’m writing this not as a final word, of course, but as something of a way to get some thoughts down on why Iran has found itself in this predicament.

Missile Missteps

Iran’s ballistic missile arsenal, the largest in the Middle East, was an important pillar of its deterrence strategy. Before the June 2025 war, Iran possessed an estimated 2,500 to 3,000 missiles and was ambitiously seeking to expand that number—perhaps as far as 8,000. The theory was sound on paper: mass salvos could saturate Israeli and American missile defenses, imposing unacceptable costs on any attacker. (Good old deterrence-by-punishment.)

In practice, thiscollapsed against the reality of integrated air and missile defense. Iran’s two direct strikes on Israel in 2024—True Promise I in April and True Promise II in October—laid bare the severe limitations of its missile force. Interception rates were extraordinarily high. The strikes did minimal damage. Crucially, these operations revealed Iranian missile capabilities and tactics to the U.S.-Israeli intelligence and defense architecture without imposing meaningful costs. You might go as far as to argue that Iran was effectively conducting live-fire training exercises for its adversaries’ missile defenses.

Then came the 12-Day War. Iran launched around 500 missiles at Israel; Israeli forces reported that only 31 landed in populated areas. Meanwhile, Israel destroyed hundreds of missiles in preemptive strikes and took out approximately 200 of Iran’s estimated 400 transporter-erector-launchers. The Israeli Air Force operated with near-total freedom in Iranian airspace, exploiting air defense gaps created by the October 2024 strikes that had destroyed Iran’s S-300 systems. (Uzi Rubin’s assessment for the Begin-Sadat Center of Operation Rising Lion contains a useful tabulation.)

The missile arsenal was supposed to be Iran’s insurance policy. Instead, its premature and ineffective employment across 2024-2025 revealed its weaknesses, degraded its stocks, and failed to deter further escalation. Iran spent a lot for little return.

Crucially, this time around it appears that the IRGC did “learn” a little. During Operation Rising Lion, the IAF successfully disrupted Iranian C2 in ways that delayed retaliation. In recognition that this would likely manifest again, Iran has fired, somewhat spasmodically as I feared last year, at countries across the region—presumably in a bid to impose costs sufficient to prompt their pleas to the United States to stand down (a perverse attempt to catalyze deescalation by proxy). I see no evidence in the first 48 hours of this conflict that this approach will work; instead, Iranian strikes have had the effect of inflaming popular anger against Tehran (while anger persists against Israel and the United States).²

After the June 2025 ceasefire, Iran embarked on a frantic effort to reconstitute. By early 2026, intelligence assessments suggested Tehran had rebuilt its missile stockpile to roughly 2,000-2,500 missiles, though there’s obvious uncertainty around exact stocks. Underground “missile cities” had partially protected production infrastructure. Dozens of missiles were being produced monthly.

But reconstitution without a changed strategic concept is just restocking a losing hand. Iran was rebuilding the same types of capabilities that had already failed against the same adversaries. More critically, the reconstitution effort itself became a justification for the very strikes it was meant to deter. Israeli intelligence tracked the rebuild meticulously, conveyed assessments to Washington that Iran was rapidly approaching pre-war missile levels, and appears to have effectively argued that the window for action was closing.

The reconstitution trap works as follows: rebuild fast enough to worry your adversaries, but not fast enough to actually deter them. Iran fell squarely into it.

n.b.: In the coming days, we could see the spasmodic nature of Iranian targeting worsen in ways that cause real chaos and terror regionally. This won’t be evidence of tactical success, but probably an extension of attempts to seek intra-war deterrence effects by imposing as much pain as possible (including obviously against U.S. forces.)

The Nuclear Threshold

Perhaps Iran’s most consequential error was its approach to nuclear weapons. I wrote about this last year amid the 12-Day War, but it’s worth revisiting. For years, Tehran pursued a “threshold” strategy: developing the technical capability to build a nuclear force if it chose to do so rather quickly without actually crossing the line. By early 2025, Iran got particularly close: it could produce enough weapons-grade uranium for a single weapon³ in under a week. It possessed enough highly enriched uranium, if further enriched, for roughly nine or ten weapons. But it never built one.

The theory behind the threshold strategy had a certain appeal: maintain latent capability as a deterrent, avoid the diplomatic costs of overt weaponization, preserve the Supreme Leader’s fatwa against nuclear weapons, keep the door open to negotiations, and explore if you might be able to manifest non-weaponized nuclear deterrence effects of some kind. The problem is that a threshold capability deters only if your adversary believes you will cross that threshold if attacked—and then still only if they believe the consequences of your crossing it are worse than the consequences of striking you before you do.

Iran achieved neither condition. The U.S. intelligence community repeatedly assessed through 2025 that Iran had not decided to build a weapon. Khamenei’s fatwa was cited by Iranian officials as proof of restraint. But this very restraint, which was intended to signal responsibility, was read by adversaries as a signal that Iran could be struck before it weaponized. Iran had, in effect, constructed the worst possible nuclear posture: close enough to a bomb to justify preventive attack, yet unwilling to cross the threshold that might have actually deterred one.

Contrast this with North Korea’s approach, which I’ve spent quite a bit of time thinking about. Pyongyang raced past the threshold, tested weapons, developed delivery systems, and presented the world with a fait accompli of what you might term minimally viable nuclear deterrence by 2017. Whatever the costs of this approach—and they were substantial!—nobody is bombing North Korea today. Instead of following the North Korean model to its logical conclusion, they stopped in the worst possible middle ground. Pyongyang and Tehran will present two parables for the next proliferator; it seems pretty clear which approach has more appeal.

A Diplomatic Bridge to Nowhere

The final error in Iran’s deterrence approach appears to have been diplomatic—and fundamentally, I think, misreading the Trump administration’s willingness to negotiate in good faith. In the weeks before the February 28 strikes, Iran was engaged in Oman-mediated indirect talks with the United States. Oman’s foreign minister announced that Iran had agreed to degrade its stockpiles of nuclear material to “the lowest level possible.” A second round of talks had been scheduled for Geneva. The Wall Street Journal’s Laurence Norman has a good thread and story on X about the very last bit of the diplomatic tick-tock:

Washington’s demands were maximalist, including a permanent end to all uranium enrichment and strict limits on ballistic missiles. These were terms no Iranian government could accept based on everything we’ve heard to date. Iran appears to have calculated that demonstrating willingness to negotiate would buy time and create diplomatic space. Instead, the parallel track of massive U.S. military buildup—the largest since the 2003 Iraq invasion, including two carrier groups—made clear that diplomacy was running on a separate clock from military planning.

Iran appears to have mistook the form of negotiations for protection against attack. It used the negotiations to signal moderation rather than to rapidly close a deal that might have preempted the military option (even if it later reneged on clearly unacceptable terms).

---

More can be said on Tehran’s missteps. I won’t get into the deterrent component that used to be Iran’s network of regional proxies: the so-called “Axis of Resistance.” Together, Hezbollah in Lebanon, Hamas in Gaza, militias in Iraq, the Houthis in Yemen, and Iranian-aligned forces in Syria constituted a potent deterrent threat for a while: attack Iran and the entire Middle East erupts. Israel has gradually seen to these proxies since October 2023, but more broadly, Tehran long treated proxy deterrence as a substitute for, rather than a complement to, direct deterrence capabilities. When the proxies were degraded, Iran had no fallback it turned out (because of what I discussed above).

You might argue that the errors I spotlight above might have been survivable in isolation; states make missteps strategically, but can still end with something resembling workable deterrence effects against their aggressors all the time. In cumulative, these failures appear to have cost Iran dearly.

If you missed the first footnote, let me take this opportunity to just underscore that nothing here should be read as an endorsement of the U.S.-Israeli strikes. I do not fundamentally believe this war is wise, legal, or likely to achieve stated objectives (…which are actually not particularly well stated outside of regime change). In any case, the killing of Khamenei does not equal regime change, and I’m looking to Iran country experts to contextualize the fluid internal political situation in the coming days. The strikes have exactly the kind of regional conflagration that Iran did appear to threaten in the past—with missiles flying at Gulf airports, oil markets in turmoil, and the Strait of Hormuz under threat. The long-term consequences for nonproliferation, regional stability, and U.S. credibility are likely to be severe. (On nonproliferation specifically, I already charted a grim prognosis for the Financial Times last year; you can read that here.)

1. This post is not about my normative views concerning the justifiability of this conflict, but if you’re wondering, I strongly abhor this war of choice—especially given its wanton illegality under U.S. and international law.

2. I do have a concern that the coming days will see this get much worse. See Iranian FM Araghchi here, effectively indicating that the IRGC is apparently freelancing with the supreme leader dead now.

3. Remember: a single weapon a deterrent does not make (and why “breakout” muddies the conversation).