Interception at 10,000 miles an hour

And what it tells us about the nature of intelligence

How do you intercept a ballistic missile travelling at over 10,000 mph?

Most people assume there is a lot of fancy tech and powerful computers. This is definitely true (and no doubt much of it is classified), but history suggests there is one principle all good interception systems rely on, a principle which is hard to improve upon. Understanding this principle will tell us something important about the nature of intelligence, and about what is so wrong about the way we commonly talk about rationality.

Interception isn’t just for missile defence. The fundamental problem of being at the right place at the right time is true when you want to catch a ball, or when a bat goes to catch a moth.

A story from the Battle of Britain in WWII reveals how humans rediscovered the same principle which evolution follows when predators catch prey (and which you instinctively follow when you catch a ball).

The interception in this case was by the British RAF of Nazi Luftwaffe bombers. At the start of the war the Nazis had better radar technology, more planes, and more experienced pilots. The shortage of British pilots, and limits to the number of hours they could be asked to fly, meant the RAF strategy for air defence relied on direct interception, rather than air patrols. When coastal radar bases picked up a signal, fighters would be sent out, guided by updated radar information. A decisive factor in the success of the air war was the RAF’s ability to rapidly and efficiently intercept inbound planes. The way they did it remained a military secret, and kept from their enemies, for the duration of the war.

The RAF had initially approached the interception problem as one of calculation - bringing in teams of human calculators, and increasingly sophisticated technology, to marshal the required information and make the trajectory calculations to allow interception, all without success. A frustrated RAF commander, E. O. Grenfell said he could do a better job ‘by eye’ rather than calculation¹, and following successful demonstration of this, a debrief was done to figure out how exactly he’d done it. The system that emerged was beautifully simple, and it was that simplicity which made it successful in the high stakes, high uncertainty, environment of aerial interception. It’s a similar principle to that which allows interception of missile-by-missile today, as the RAF used for plane-by-plane in WWII.

Grenfell’s method, refined by Sir Henry Tizard (then President of Imperial College), was to fix a constant angle of approach between the to-be-intercepted bomber and the intercepting fight. This simple rule relies only on knowing two pieces of information: the heading of the two planes, and produces only one output - the course adjustment that the pilot of the intercepting fighter needs to make. It is also robust if the to-be-intercepted bomber changes course.

This figure illustrates. The fighter (top, in black) is keeping a constant angle T° to the bomber (bottom, in outline):

Figure 3b from Figure 2 from Hamlin (2017) https://doi.org/10.1111/tops.12253

The Nazi air-force never discovered this method, putting them at a disadvantage when RAF bombers started incursions into their airspace. The Luftwaffe threw vast technical and human resources at the interception problem, but during the war remained fixed on a method based on continuous narration from ground control to the pilot, incorporating radar information and feedback from the pilot to inform course adjustments. Hamlin (2017), describes how a single pursuit could occupy a crew of 80 people in the information collation and calculations required. The Nazis’ superior technology, richer information supply and a more sophisticated system of calculation ultimately proved a disadvantage. It ate up more and more resources, and the RAF hit upon the strategy of sending 1000 planes at once to pass through an area controlled by a single radar station, and so completely overwhelming the Nazi capacity to plan interceptions.

So here we now see the moral for understanding intelligence. The RAF superiority was not in equipment, nor in information, nor in resources, but in the control system that combined them. RAF interceptions used a simple rule. It was easy to calculate, frugal in the information it required, fast in how it could be communicated to the pilot and robust to changes in the environment.

You use a similar system when you go to catch a ball or frisbee. Yes, technically, there is a complex set of equations which determined the trajectory of the object, factoring in the initial projection force, wind speed, air resistance and so on. None of us do that in our head, we just find ourselves in the right place at the right time, and the secret is analogous to how the RAF intercepted Luftwaffe bombers - maintaining a fixed angle to the ball as you move to catch it:

Figure 2 from Gigerenzer G (2021) Embodied Heuristics. Front. Psychol. 12:711289. doi: 10.3389/fpsyg.2021.711289

In fact, the information processing required by your brain is made even easier through a simple trick. No dedicated circuit needs to represent the relative velocity of the ball and the myriad factors which affect its flight. The catcher just moves so that the image of the ball stays in the same place on their retina. As long as this can be maintained - if the catcher is able to move so the image doesn’t creep towards one of the edges of their vision - they are on track to intercept.

No calculation is needed here, apart from detecting drift in either direction and speeding up or slowing down to compensate. The information required for the decision is available immediately in the environment, in a form which can be readily fed into the control of movement.

§

There’s a prejudice to assume that more complex reasoning is better. In the dual process models of psychology, made famous by the title of Kahneman’s Thinking Fast and Slow, the tendency is to contrast slow (rational, deliberate) processes with fast, (biased or intuitive) processes. Quick and simple is associated with inferior reasoning.

The interception story shows why this is misleading. The better strategy, tested in both evolution and in human combat, is the simple one. This strategy is easy to implement, fast and adaptable. It allows the tight coupling of the agent with the environment which is the hallmark of true intelligence - being able to do the right thing at the right time.

Using something like the angle of approach strategy, and iterative approximation to adjust for interception, is so successful it was built into the post-war generation of guided missiles. The successors of these were the interceptor missiles protecting the skies in the Middle East and they take advantage of the same principle at their core.

§

Interception is one example of how complex problems don’t always require complex solutions. Hamlin (2017) provides these two schemes: A complex decision strategy, relying on multiple information sources (top) with the simple, iterative, strategy (bottom).

Figure 2 from Hamlin (2017) https://doi.org/10.1111/tops.12253

Often, faced with complexity, we assume we need something like the top strategy. Gather as much information as possible, reconcile it and produce a single consequential decision.

Much has been made of the inherent uncertainty in decision making, but the real world is also changing as well as uncertain. By the time you have produced a decision the information you based it on will be out of date. In a rapidly changing world the computational costs of decision making are real limitations on effectiveness.

The interception strategy is an example of the bottom strategy: a simple rule that uses one piece of information to produce rapid updates, allowing (repeated) actions, solving the problem in stages.

It can be a mistake to meet complexity with more complexity in decision making. Sometimes simple rules which provide rapid guidance by taking advantage of the structure of information available in the environment outperform alternatives. The interception strategy is one example.

---

This newsletter is free for everyone to read and always will be. If you can afford it, feel free to chip in to help me keep writing for everyone: upgrading to a paid subscription (more on why here)

Below, further reading and other things I’ve been thinking about.

---

For more on fast and frugal heuristics:

• Gigerenzer, G., Todd, P. M., & ABC Research Group, T. (2000). Simple heuristics that make us smart. Oxford University Press.

For a history of the interception heuristic I have followed Hamlin (2017), whose article is exceptionally readable for a scholarly paper:

• Hamlin, R. P. (2017). “the gaze heuristic:” biography of an adaptively rational decision process. Topics in cognitive science, 9(2), 264-288. https://doi.org/10.1111/tops.12253

A commentary putting Hamlin’s history in context is:

• Gigerenzer, G., & Gray, W. D. (2017). A simple heuristic successfully used by humans, animals, and machines: the story of the RAF and Luftwaffe, hawks and ducks, dogs and frisbees, baseball outfielders and sidewinder missiles-oh my!. Topics in Cognitive Science, 9(2), 260-263.

Gigerenzer influenced Andy Haldane, then Executive Director for Financial Stability at the Bank of England, who wrote an essay about applying these principles to avoiding financial crises:

• The Dog and The Frisbee (2012)

Other things…

PREPRINT: Accurate Information Can Substantially and Durably Increase Republicans’ Beliefs in Election Integrity

Another result which confirms that the intractability of partisan, motivated, cognition is often far exaggerated. From the abstract, bold mine:

To examine whether election integrity beliefs are indeed resistant to corrective information, we develop and test an informational treatment that provides a high volume of politically balanced accurate evidence on election integrity. Immediately prior to the 2024 general election, we randomly assigned N = 871 Republicans to either the experimental group or a control group engaging with general political information. The treatment substantially increased participants’ overall beliefs about the integrity of US elections, retrospective beliefs about the integrity of the 2020 election, and prospective beliefs about the expected integrity of the upcoming 2024 election (.6 < ds < .8). Furthermore, a follow-up shows that the effects persist two weeks later, following the 2024 election. These findings demonstrate that even beliefs closely tied to partisan identity are responsive to credible factual information.

Stagnaro, M. N., Amsalem, E., Mullin, A. B., Lin, H., Berinsky, A. J., & Rand, D. G. (2026, February 10). Accurate Information Can Substantially and Durably Increase Republicans’ Beliefs in Election Integrity. https://doi.org/10.31234/osf.io/42pum_v1

Catch-up

Recent newsletters from me:

• Specifically right, generally wrong. Have I stumbled upon a dark pattern for generating engagement?

• The effects of political disappointment. Testing how we feel about each other when democracy doesn’t go our way, a work in progress.

• Good bias. Let’s untangle what people mean when they say the B word

• Gambling with research quality. How you get 244 different ways to measure performance on the same test of decision making. And what it means for the reliability of behavioural science

PAPER: The shades of (un)productive disagreement: A systematic literature review and a hierarchical taxonomy

So many ways to disagree!

Salman, A., & Kolikant, Y. B. D. (2026). The shades of (un) productive disagreement: A systematic literature review and a hierarchical taxonomy. Educational Research Review, 100769. https://doi.org/10.1016/j.edurev.2026.100769

PREPRINT: Hyperactive Minority Alters the Stability of Community Notes

"[We] conduct counterfactual simulations that modify the display status of notes by varying the pool of raters. Our results reveal that the system is structurally unstable: the emergence and visibility of notes often depend on the behavior of a few dozen highly active users"

Nudo, J., Nemmi, E. N., Loru, E., Mei, A., Quattrociocchi, W., & Cinelli, M. (2026). Hyperactive Minority Alter the Stability of Community Notes. arXiv preprint arXiv:2602.08970.

Will Manidis & Nabeel S. Qureshi: Rented Virtue

Of note:

"The work that lasts from this era will be no different. The companies that endure, the technologies that serve rather than consume, the institutions that hold their shape across generations when the founders are dead and the capital is restless and the market is telling them to be something other than what they are — they will have something at their center that resists justification in purely secular terms."

Link: Rented Virtue

Sound the Nerd Victory klaxon

Tax nerd throws life savings at prediction markets, after realising he could take the opposite side of bets by Musk fans. DOGE, inevitably, fails to decrease year on year federal spending. He wins big.

WSJ: The Tax Nerd Who Bet His Life Savings Against DOGE (Feb 25)

…And finally

END

Comments? Feedback? Want to be intercepted? I am tom@idiolect.org.uk and on Mastodon at @tomstafford@mastodon.online

1. “As tempers, ideas, and time were all short by that point, he was immediately and expensively given an opportunity to publicly prove his claim.” Hamlin (2017)