How missile defence shields work (or don’t).

<< Review: House of Nitroglycerine

TBC >>

Hello again. I really had intended to move on from my A House of Dynamite obsession this week (honestly!), but I kept coming back to one important plot point in the show, and thought it would be worth spending on more week talking about missile defence shields.

A coin toss? That’s what we get for $50 billion?

—SECDEF, A House of Dynamite

As I mentioned the last week, this figure underestimates by a factor of about ten how much money the US has pumped into ballistic missile defence since they first started. And yet, as the film accurately points out, the success rate is roughly 50%. Why is it so hard for the most technologically and economically advanced society in world history to shoot down one measly missile? In other words, why is missile defence so hard?

I’m going to begin by describing what’s so great about missile defence systems, aside from the obvious. Then I’ll move on to the typical launch to impact journey of a ballistic missile, and where along this path is best to stop it. I’ll talk about the kind of payload you need to actually destroy a ballistic missile. Finally, I’ll discuss the practical and technical difficulties to finding and destroying the incoming ballistic missile.

Throughout the article, I’ll draw contrasts between what actually exists in the US and elsewhere, and what the laws of physics and engineering could potentially allow in the future.

In case you missed my review of AHoD over the last two weeks, here’s Part 1 and Part 2.

Before we start, let me encourage you to subscribe to the blog using the link below. As always, I’d love to see your comments below, or you can contact me via webform here or email here. Finally, if you’ve enjoyed this post, please share it with a friend.

If you enjoy this blog and want to support it, please consider a donation. Keeping this blog going doesn’t cost much, but it isn’t free either, so any help would be very much appreciated👍

Missile defence “domes” give a powerful defence capability

One of the main jobs of a government is to protect its people from external threats. It’s hard to imagine a scarier threat than high explosives or nuclear warheads raining down from above, so any protection against this threat is bound to be a good investment by any government. Provided it works, of course.

The most well-known missile defence system is probably Israel’s Iron Dome which, since 2011, has defended against artillery and short-range rockets fired from as far away as 70 km.

The Iron Dome has been extremely successful at defending Israeli towns and villages against attacks from Lebanon and Gaza, albeit at a very asymmetric cost: each interception costs about a hundred times more than the munition fired. It has also been a victim of its own success, in that laypeople assume it is a defence against all types of aerial threat.

In fact, the Iron Dome was not designed to protect against ballistic missiles, although it still managed a 30% success rate against Iranian ballistic missiles during the 2025 war. It’s part of a multi-layered defence system:

The idea is becoming more popular: Leonardo, the Italian defence giant, today announced the development of software which would link various European air defence systems into an integrated “Michaelangelo Dome.”

So far, we’ve looked at a very real-world example which deals with short- to medium-range threats. What about a true missile defence shield, something like the Golden Dome which President Donald Trump has promised?

Aside from making citizens feel very safe, a reliable missile defence shield would also significantly lengthen the decision-making window which world leaders have when faced with a nuclear attack scenario. As we discussed last week, the drama of AHoD and other nuclear war stories is the compressed time in which our heroes need to make a life-or-death decision for their countries and the world¹. If you could be confident that any incoming missiles would be shot down, then you have a lot more breathing space to figure out what’s going on, who’s at fault, and what you’re going to do about it. It negates the need for launch on warning.

The US does not have a Golden Dome, but it does have a somewhat integrated National Missile Defense (NMD) system. This can attack intercontinental ballistic missiles (ICBMs) in mid-course, i.e. when they’re in space—I’ll explain more in the next section. There is considerable scepticism as to how it would fare in a real situation. Aside from its anti-ICBM capability, it also consists of countermeasures for shorter range ballistic missiles, in a similar way to Israel’s Arrow and David’s Sling we saw above.

If missile defence shields are so great, then why aren’t they more common? There’s one big technical challenge: hitting something that fast is incredibly hard.

There are three places to stop a ballistic missile

The incoming North Korean warhead will be travelling at… 14,000 mph, while the interceptor’s kill vehicle will be travelling at… 20,000 mph, making this action, if successful, “akin to shooting a bullet with a bullet”, according to the Missile Defense Agency’s spokesperson.

—Annie Jacobsen, Nuclear War: A Scenario

The spokesperson quoted above actually undersells the difficulty of this engagement. A bullet travels at around 2,000 mph, so the closing speed of a missile defence engagement is about nine times faster, at 34,000 mph instead of ~4,000 mph.

A ballistic missile goes through three phases: boost, mid-course, and terminal.

The distances, speeds, and horror of ballistic missile exchange is beyond our easy comprehension, so let’s use an analogy that’s a bit easier.

You’re living an idyllic life in a small village (let’s call it Moonbrush Wood²). It’s got a big wall around it and guys with guns to defend the walls, so you don’t need to worry about the unfriendly village nearby: Grimsforth.

Sometimes it’s impossible not to think about it, though, when you see the big sniper tower in distant Grimsforth. So, how does Moonbrush Wood defend against a sniper on the tower in Grimsforth? 

Here are your options:

• Have a spy in the enemy camp and shoot their sniper before he climbs the tower. This is like boost-phase intercept, neutralising the threat before it gains all its energy and is harder to stop. The easiest option, but you need to get up close to the enemy.
• Station some shooters in the fields in no-man’s land between the villages. Give them a radar so they can track enemy sniper fire and get them to shoot the bullets out of the sky with their own bullets. This is like mid-course intercept. The projectile is going fast and is very high up, so you need a lot of speed and pinpoint accuracy to get it. Plus, it’s hard to find and track the threats.
• Put some shooters in your own village with their guns pointing towards the incoming bullets. This is like terminal phase intercept. It’s a bit easier to see the threats incoming, but it’s nearly hitting you, so your window for success is narrow.

Hitting the ballistic missile is even harder than hitting a bullet, because of the immense speeds and attitudes involved. Let’s take a look at what’s needed in the next section.

To match the energy of a ballistic missile you need… Another ballistic missile

The first human object in space was one of Nazi Germany’s V-2 ballistic missiles. These rockets, which were precursors to the US Redstone rocket which sent the first US astronauts into space, reached altitudes of 88 km on a ballistic arc which gave it a 200 km range.

The terrible descendants of the V-2, the modern intercontinental ballistic missiles (ICBMs) which can carry nuclear warheads anywhere in the world, reach altitudes of over 1,000 km on ranges of over 5,000 km.

Let’s take a look at what this means in scale:

An ICBM, and its singular or multiple warheads, is in every sense a spacecraft: a sub-orbital one like Blue Origin. Admittedly, it’s a single-use version, but otherwise it’s exactly the same.

If you think about how hard it is to shoot down a satellite, then you’ll appreciate how hard it is to shoot down a ballistic missile, because they are pretty much the same thing. The only thing that can reach and take out a ballistic missile is another ballistic missile: something else that can get as far and as high as the target.

Getting to the target is just one challenge: destroying it is another. Nuclear warheads are designed to withstand the extreme temperatures of re-entering the Earth’s atmosphere, so they will have a certain amount of hardening. On the other hand, nuclear weapons are very finicky machines, so if you can get through the heat shield and damage the warhead itself, you’ll probably prevent a full nuclear explosion.

Some of the earlier interceptors used their own nuclear warheads to destroy incoming ballistic missiles. This won’t be news to you if, like me, you spend an inordinate amount of time playing Missile Command³, where the anti-ballistic missiles you use are definitely “coded” nuclear:

This reflects reality, where some of the earliest missile defence systems used nuclear interceptors. One example is the Soviet A-135 “Gorgon” missile defence shield around Moscow. Its early iteration used nuclear-tipped interceptors, but its successor system will use conventional interceptors.

Why use nukes? Put simply, when hitting the target is difficult, you need to give yourself as much opportunity as possible. A nuclear explosion will release thermal radiation along with x-rays and neutrons which can damage the enemy’s nuke at a distance. However, it also generates a hefty electromagnetic pulse which damages or destroys friendly (well, any) satellites in space and power grids down on the ground. As guidance and control options improved (see next section), and we learned more about the dangers of nuclear explosions high in the atmosphere, the idea of a nuclear interceptors became less prevalent.

Guidance and control are hard, even without decoys

Guided weapons have improved in the last half century or so as electronic components became smaller, more reliable, and more powerful. This has improved the ability of interceptors to destroy their ballistic missile targets at a closer range and without the need for a large nuclear explosion.

The “exoatmospheric kill vehicle” (EKV) used by the US ground-based midcourse interceptor is a kinetic energy weapon: it relies on the smashy energy of colliding with the enemy warhead to do its damage. This means it doesn’t need to carry a complicated warhead and fuzing system. It does, however, need to actually hit the target or else it’s a wasted effort.

The EKV above has rockets on board for manoeuvring into the path of its target. Even with all this control, it still fails about half the time, as we saw in A House of Dynamite. This is a problem when we are faced with multiple incoming warheads and decoys such as metal strips or balloons which mimic the radar cross-section and ballistic trajectory⁴ of a warhead:

So although the stated reliability of the US’s missile defence system is on the order of 50%, this is without adding multiple warheads and decoys into the mix. After all, you only have one interceptor per launch platform.

What about lasers? Light is weightless, so it doesn’t need to be lofted into space on a big hulking rocket. It also has the advantage of travelling at the speed of, well, light, so if you can see the target, you can hit it.

It’s a promising idea, but there are many practical problems to even getting a powerful enough laser developed, let alone the issue of seeing and tracking the target for as long as it would take to destroy it.

The video below from Real Engineering is worth a watch for the lasers alone, but if you want to see the challenge of targeting nuclear warheads, skip to about 9:00:

The 13 minutes it would take to destroy an ICBM warhead is obviously no use: even if you saw and shot it the moment it came over the horizon, it would still be on top of you before you pierced the heat shield. Even if we upped the power a hundredfold, to the 1 MW laser he describes at the start, it would still take about eight seconds of continuous tracking to destroy it.

Conclusion: Even harder than hitting a bullet with a bullet

Hitting a bullet with a bullet is hard. It took the guys from SmarterEveryDay years of careful experimentation and tightly controlled rigs to accomplish it:

Hitting a ballistic missile which is travelling ten times faster is a much bigger problem, even with modern guidance and control systems. It’s a miracle the tests work as often as they do.

Most people don’t understand and appreciate the difficulties, and Hollywood (as always!) bears some of the blame. To take just one example, here’s the scene from Mission: Impossible – Ghost Protocol where a Russian submarine-launched ballistic missile (i.e. an ICBM) is about to hit San Francisco:

There are many things wrong with this scene, but the main one is how the missile is pottering along about as fast as a jet aircraft and tracking across the city on a shallow glide path. Remember: ballistic missiles (it’s in the name) go up high, then gravity brings them back down very quickly. An ICBM on its terminal phase will reach 24,000 km/hr, although atmospheric drag will slow it down to about 3,200 km/hr (.pdf link) before it reaches its target. This should be coming straight down onto the city from a great height and probably travelling much faster as well⁶.

In summary, stopping a ballistic missile isn’t like “hitting a bullet with a bullet.” It’s much harder.

Thanks for reading and please remember, if you haven’t already, to subscribe using the link below. Please also share this article with a friend. See you next week.

Featured Image: Common “two bullets colliding⁷” image found on Facebook, Reddit, Quora, etc. Often said to be from the “Battle of Gallipoli” in 1915/16, often with the additional explainer that this has a “one in a billion” chance of happening⁸. Google link here.

1. Although, as I mentioned last week, this drama shouldn’t actually apply in the AHoD scenario, since the destruction of Chicago (tragic though that would be) doesn’t affect the US’s strategic response capabilities in any way. ↩︎
2. With credit to Dungeon Keeper. ↩︎
3. I should clarify; this was a very old title when I was playing it. I wouldn’t want anyone to accuse me of “stolen valour” in my video game vintage. ↩︎
4. Because this is all happening in space, and space is a vacuum! You might be rolling your eyes and saying “duh” at me, but I still think it’s mind-boggling that you can blow up a balloon in space and it will travel exactly like a big heavy lump of metal and complicated physics. ↩︎
5. Stages, from original source, are: 1) The missile launches out of its silo by firing its 1st stage boost motor (A). 2) About 60 seconds after launch, the 1st stage drops off and the 2nd stage motor (B) ignites. The missile shroud (E) is ejected. 3) About 120 seconds after launch, the 3rd stage motor (C) ignites and separates from the 2nd stage. 4) About 180 seconds after launch, 3rd stage thrust terminates and the Post-Boost Vehicle (D) separates from the rocket. 5) The Post-Boost Vehicle manoeuvres itself and prepares for re-entry vehicle (RV) deployment. 6) The RVs, as well as decoys and chaff, are deployed during backaway. 7) The RVs and chaff re-enter the atmosphere at high speeds and are armed in flight. 8 ) The nuclear warheads detonate, either as air bursts or ground bursts ↩︎
6. Some of the other issues are the fact that the “rocket” part is still attached and still burning (it would have burned out a few seconds after launch and then detached); the warhead appears to be fuzed for impact, when most nukes would detonate at a set altitude for maximum lethality, and the missile stays intact after smashing into a building. ↩︎
7. The caption often implies that these two bullets collided in midair. However, if you read most of the online commentary, they point out that one of the bullets (the lighter-coloured one) doesn’t have rifling marks on it (the grooves or scratches running along its body). This means that it wasn’t fired out of a rifle barrel, which means that it was probably hit by the other bullet while in a magazine or ammunition box or otherwise not flying through the air. ↩︎
8. I don’t know where or how you would to this one-in-a-billion stat. It seems, like 85.4% of all statistics, to have been made up on the spot. ↩︎