No, Missile Defense Will Not Work 97% of the Time

, senior scientist | October 13, 2017, 11:26 pm EST
Bookmark and Share

In an October 11 interview on Fox News, President Trump claimed:

We have missiles that can knock out a missile in the air 97 percent of the time. If you send two of them, they are going to get knocked down.

This is not true. At least not in any relevant way.

The only homeland missile defense system is the Ground-based Midcourse Defense (GMD) system, which I’ve written plenty about here in these pages, and have co-authored a recent report about. If you’ve been following along, you’ll know the president’s statement was clearly untrue.  I’ll explain why.

What does the actual test record show?

The GMD interceptors have succeeded in destroying the target in nine out of 18 tests since 1999 (50%).  They have destroyed their target in four out of 10 tries (40%) since the GMD system was nominally deployed in 2004. They have destroyed their target in two of the last five tests (40%).

So there is no basis to expect it to work any better than 40 to 50% of the time even under the most generous and easiest conditions—former Pentagon testing agency director Phil Coyle calls the test conditions so far as “scripted for success.”

While the test record says something about the GMD’s capabilities under scripted conditions, the real world will be more complex and challenging. The Pentagon’s highest testing official assessed in 2014 that the test program was “insufficient to demonstrate that an operationally useful capability exists.” More on this later.

But for sake of argument, say the “single shot kill probability” has been determined via tests to be 40 to 50% in those optimistic conditions. Because reliability is low, the US would fire multiple interceptors at the missile to try to boost the system’s effectiveness. Using four-on-one targeting, and a 40 to 50% chance that a given interceptor would work, this leads to a 6 to 13% chance that the warhead gets through.

Real-world conditions

But this isn’t the right question. If it came down to a nuclear attack, would North Korea send just a single missile, and choose the most convenient conditions? That seems unlikely. Let’s say the salvo is five incoming missiles. In that case, with an interceptor kill probability of 40 to 50%, using four interceptors on each missile, the probability that one warhead gets through is 28 to 50%. Uncomfortably high.

I could not stress more that this is a best-case scenario. It assumes that:

1) Failures are uncorrelated and not, e.g., a design flaw common to all interceptors, such as the guidance system issues that took nearly a decade to diagnose and fix,

2) The intercept attempts take place under simplified conditions and that the system is not being stressed as it would in a real-world situation, and

3) The system successfully identified the five real targets from among decoys. If the system cannot distinguish decoys from the real targets, it will have to engage them all, quickly depleting the interceptor inventory. These do not need to be the Ferraris of decoys to be an issue. Some of the GMD intercept tests have included decoys, but all of those have been designed to be easily distinguished from the target warhead.

In short, one can construct situations under which missile defense might destroy missiles: a small salvo of missiles sent without countermeasures and under the limited range of conditions under which the system has been tested. The problem is that these are not by any stretch the most *likely* situations. A potential adversary has every incentive to make the attack as difficult as possible to intercept if he is going to initiate World War Three.

Note that even if the president were instead talking about one of the missile defense systems that has a better and more complete test record, such as THAAD, the issues with not having been tested in operationally realistic conditions is the same. And because THAAD defends against shorter-range missiles from North Korea, which are cheaper and more plentiful, it has the additional issue that it may be overwhelmed even if it is able to discriminate between decoys and real targets. There just may be too many targets.

Why is this dangerous?

The best-case scenario is that President Trump is trying to avoid a confrontation by allowing himself to save face: he has declared that North Korea must not be able to threaten the US mainland with nuclear-armed missiles. Or that he hopes such statements would help dissuade North Korea from considering an attack.

Certainly worse than this is the possibility that Trump actually believes that strategic missile defense provides credible protection and he has not been advised correctly. One hopes he is provided accurate information by stewards of these programs, although at least in public, government official often describe the GMD system as much more capable than it has been demonstrated to be.

This is dangerous, because common sense would say that if we have spent $40 billion on a missile defense system that the US has claimed has been “operational” for going on fifteen years, it must “work.” But it doesn’t. Look at the test record.

The problem is that believing missile defense works when it doesn’t can lead you to take actions that make you need it, and then it can’t help you.

Posted in: Missiles and Missile Defense Tags: , , , ,

Support from UCS members make work like this possible. Will you join us? Help UCS advance independent science for a healthy environment and a safer world.

Show Comments


Comment Policy

UCS welcomes comments that foster civil conversation and debate. To help maintain a healthy, respectful discussion, please focus comments on the issues, topics, and facts at hand, and refrain from personal attacks. Posts that are commercial, self-promotional, obscene, rude, or disruptive will be removed.

Please note that comments are open for two weeks following each blog post. UCS respects your privacy and will not display, lend, or sell your email address for any reason.

  • Ryan Alt

    How much longer before NK has a mirv capability? My guess would be that they are a long way from that. Is the fact that it has never once been tested realistically against even the most limited attack mean that they already know that it isn’t up to the task?