The 30-to-1 Problem: Why Drone Defence Economics Favour the Attacker

Thirty to one is the number that gets quoted. What gets missed is the second-order effect, which is worse.

At thirty to one, the attacker does not need to win engagements. The attacker needs only to cause engagements. Every launch either destroys something worth thirty times its cost, or forces a response that costs more than the launch did. There is no outcome in which the defender comes out ahead on the exchange, which means the attacker's optimal strategy is simply more.

And "more" is a production question, not a tactical one. By 2025, Ukraine and Russia were each producing 100,000+ FPV drones annually. That is the actual weapon: not the individual munition, but the rate at which it can be replaced.

Yemen: six hundred to one, and the reason nobody says it out loud

Ukraine is the well-documented case. Yemen is the more alarming one, and it happened first.

Houthi drones costing $5,000–20,000 forced Saudi Patriot interceptors costing $3,000,000+ per launch.

Run that at the low end and the ratio is in the hundreds to one. Run it at the high end and it is still catastrophic. And unlike the Ukraine figure — where the comparison is a drone against the artillery piece it destroyed — the Yemen figure compares a drone against the interceptor fired at it. The defender pays that cost whether or not the intercept succeeds, and whether or not the drone was carrying anything.

That distinction matters enormously and is almost always collapsed in the retelling. An attacker who understands it does not need to arm every airframe. Empty drones cost the defender exactly as much to engage as armed ones. At that point the attacker is not conducting a strike campaign; the attacker is conducting a procurement attack, and the target is your magazine and your budget line.

Yemen proved the model years before Ukraine made it famous. The operational conclusion was available in 2019 and was largely not acted upon, because the conclusion is expensive and unwelcome: the defender's economics are structurally unsustainable at scale.

Nagorno-Karabakh: forty-four days, and what money could not buy

The 2020 Nagorno-Karabakh conflict is the case that removes the comfortable objection — that this is a problem for under-equipped defenders who did not invest properly.

Armenia had invested. Armenian S-300, Buk and TOR air defence systems were destroyed by TB2 and Harop within the opening days. Armenian losses reached 150+ T-72 tanks and 180+ IFVs, confirmed on video. The war lasted 44 days.

Those are strategic air defence systems, and they were not defeated because they were cheap or old. They were defeated because they were built for a different target set, and because the mid-altitude belt between MANPADS and strategic SAMs — the SHORAD gap — was where the actual threat operated.

The finding is precise: SHORAD gaps are decisive vulnerabilities. You cannot spend your way past this by buying a better strategic SAM, because the strategic SAM is not where the problem is. And a strategic SAM engaging a loitering munition is the exchange-rate problem in its most extreme form.

March 2026: what saturation does to the arithmetic

In March 2026, more than 3,000 drones and missiles were launched against all six GCC states in 36 hours.

I had been building the analytical position since 2020. The campaign validated it, and I take no satisfaction in that, because what it validated is that the exchange rate is not merely unfavourable — it is exploitable on demand.

Here is why saturation is the economic problem rather than a tactical one. Interceptor magazines are finite and slow to replenish; drone production lines are fast and cheap. An attacker who can generate engagements faster than you can replace interceptors does not have to defeat your air defence. The attacker has to outlast it, and the timescale for outlasting it is measured in hours, while the timescale for replenishing it is measured in months.

The campaign changed the conversation more than five years of analysis had. Governments that had treated drone defence as a future concern began treating it as an immediate operational emergency. That shift happened not because the technology changed in March 2026 — it had not — but because the arithmetic finally arrived somewhere with a budget.

Why the standard responses do not fix it

Four responses come up in every discussion of this problem. Each one helps. None of them changes the exchange rate.

"Build a cheaper interceptor." This is the most sensible response and the most limited. Even a dramatically cheaper interceptor is still consumable, still magazine-limited, and still on the wrong side of a production-rate comparison. Cheaper interceptors improve the ratio; they do not invert it, because the attacker's unit cost is falling at least as fast.

"Use directed energy." Cost per shot is genuinely excellent, and this is the most promising direction. It also has a hard set of constraints — power, weather, engagement time per target, and the fact that engagement time per target is exactly the wrong property to have in a saturation event. Directed energy is very good against the single intruder and structurally challenged against three thousand.

"Jam everything." Electronic warfare has no per-engagement cost, which is the right property. It also has an expiry date, and that date has already passed in one important case: fibre-optic guided FPVs emerged specifically to defeat RF jamming, and they are immune to electronic countermeasures. A defence whose economics depend entirely on jamming is a defence with a known, already-fielded counter.

"Harden the target." Correct, underrated, and only partial. Hardening works for fixed assets and does nothing for the things that have to move, be seen, or be open.

The uncomfortable common thread: every response above is an attempt to reduce cost per engagement. None is an attempt to remove the per-engagement cost entirely — and that is the only move that actually breaks the arithmetic.

What actually changes the exchange rate

A defence mechanism whose cost does not scale with the number of engagements.

That is the whole idea, and it is why I have spent five years arguing for a physical layer rather than a better electronic one. A physical net perimeter does not care how cheap the drone was, how it navigates, whether it is transmitting, or whether the jammers are saturated. It has an installation cost and a maintenance cost, and then it has approximately the same cost whether it stops one aircraft or four hundred. It fails only physically.

That is not a claim that a net replaces radar, jammers or interceptors. It is a claim about where the load-bearing element belongs. Electronic layers are excellent at detection, classification and cheap early defeat, and they degrade precisely when the attack is most serious. A physical layer is indifferent to the thing that degrades them.

The full argument, and the district-level team architecture it sits inside, is the Shield Curtain doctrine. Its practical application to a facility is the layered design guide.

How the inversion happened

It is worth understanding why the exchange rate flipped, because the mechanism tells you whether it can flip back.

For most of the twentieth century, air threats were expensive because they were manned. An aircraft carried a pilot, and a pilot required an airframe engineered to keep a human alive, trained over years, and recovered after the mission. Every one of those requirements imposed cost, and every one of them capped the number of sorties an adversary could generate. Air defence was expensive too, but it was defending against something even more expensive, and — critically — against something that existed in small numbers.

Three things removed those constraints simultaneously.

The human came out. No life-support engineering, no ejection system, no recovery requirement, no training pipeline per airframe, and no political cost to losing one. The airframe collapses toward the cost of its components.

The components became consumer goods. Flight controllers, cameras, radio links, batteries and motors are commodity items produced at consumer volumes for entirely unrelated markets. The defence industry does not set their price; the consumer electronics industry does, and it drives it downward relentlessly.

Manufacture became parallelisable. A missile is produced by a defence contractor on a defence timescale. An FPV drone is produced by many small facilities simultaneously — which is how you arrive at 100,000+ annually, per side, by 2025.

The defender's side of the equation experienced none of these changes. An interceptor is still a precision-engineered, low-volume, defence-procurement item. The manufacturing base is consolidated, the lead times are long, and the unit cost does not fall.

That asymmetry is structural. It is not a temporary consequence of one conflict, and there is no visible mechanism by which it reverses — because the thing that would reverse it is interceptors becoming commodity consumer electronics, and interceptors are not going to become commodity consumer electronics.

Calculating your own exchange rate

This is a number every facility and every force should have, and almost none do. It takes an afternoon.

Establish your cost per engagement. Not the sticker price of the interceptor — the fully-loaded cost of responding once. Munition, crew time, system wear, any operational disruption the response itself causes, and the cost of any collateral effect the response creates. That last term is frequently the largest and is almost never counted.

Estimate the adversary's cost per engagement generated. The airframe, the payload if there is one, and the operator's time. Use a published figure rather than an optimistic one — the ranges in this article are the public record.

Divide. That is your exchange rate.

Then ask the second question, which is the one that matters: how many engagements can you sustain before the magazine is empty, and how long does replenishment take? Express both as time, not as quantity. A defence that holds forty engagements against an adversary who can generate ten an hour has four hours, not forty units.

Most facilities that run this exercise discover two things. Their exchange rate is worse than they assumed, because they never counted the collateral and disruption terms. And their magazine, expressed in hours, is far shorter than the duration of any campaign that has actually occurred — the March 2026 campaign ran for 36 hours.

What this means if you are responsible for something

Three conclusions, in order of how uncomfortable they are.

Your cost per engagement is a threat surface. If you can be made to spend seven figures by an adversary spending four, you have handed over a lever. Audit it. Most facilities have never calculated the number.

Magazine depth is a schedule, not a quantity. The question is not how many interceptors you hold; it is how many hours of a saturation event you hold, against an adversary with a production line. Those are very different numbers and only one of them is on the inventory sheet.

At least one layer must be non-consumable. Whether that is physical, directed-energy, or something else, the architecture needs an element that does not run out. A defence composed entirely of consumables has a defined end, and the attacker chooses when it arrives.

The uncomfortable summary

The economics of drone warfare do not favour the attacker because the attacker is clever. They favour the attacker because the attacker's unit is cheap, replaceable and produced at industrial rate, and the defender's unit is expensive, scarce and produced at defence-procurement rate. That is a structural condition, not a temporary technological accident, and no interceptor improvement I have seen changes the structure.

You cannot buy your way out of an unfavourable exchange rate. You can only change what you are exchanging.

---

What I bring that an analyst's note cannot: I run the operational side as well as the analytical one, which means I am the person who has to justify the exchange rate to a client — not the one who gets to describe it and move on.

---

Carlos Kfoury is GM/CEO of CIS Security and founder of RAGE X Corp. Conflict intelligence with explicit confidence and urgency labels is published through RAGE Intel.

Related: Counter-Drone Architecture for Critical Infrastructure · Inside the March 2026 Iran–GCC Campaign · The Shield Curtain Doctrine