Occam's Razor: Why Assumptions Are Expensive

Two explanations for the same set of facts. Your colleague did not reply to an important email for three days. Explanation one: they are deliberately freezing you out, having decided — perhaps after a conversation you were not part of — to marginalise your project. Explanation two: they missed it.

Both explanations fit the evidence perfectly. Both are internally consistent. Only one requires assuming a hidden decision, a motive, a conversation you did not witness, and a sustained intent to deceive. The other requires assuming that an inbox is busy.

Occam's Razor is the principle that when competing explanations account for the evidence equally well, the one requiring the fewest assumptions should be preferred. It is the most widely quoted rule of thumb in reasoning and the most consistently mangled — usually by people who think it means "the simple answer is always right."

What the razor actually says

Two qualifications carry all the weight, and dropping either turns the principle into nonsense.

It applies only among explanations that fit the evidence. The razor is a tiebreaker, not a first filter. A simpler theory that fails to account for the data is not preferred over a complex one that succeeds — it is simply wrong, and simplicity does not rescue it. "The engine stopped because it ran out of fuel" is beautifully simple and worthless if the tank is full.

It counts assumptions, not syllables. "Simple" does not mean easy to understand or comfortable to accept. General relativity is conceptually harder than the theory it replaced but assumes less. The razor asks how many independent, unverified things a claim requires you to accept — not how difficult the mathematics is.

Get these right and the principle stops being a vague preference for tidiness and becomes something closer to arithmetic.

Why it works: the probability argument

The razor is not merely an aesthetic taste. It has a justification, and once you see it, the principle becomes hard to argue with.

Every assumption in a chain of reasoning is a claim that might be false. Independent probabilities multiply. A theory requiring five assumptions, each generously 90% likely, is 0.9⁵ ≈ 59% likely overall. Add three more equally plausible assumptions and it drops below 43%. Meanwhile a rival theory requiring one assumption at 80% sits comfortably above both.

This is why complexity is probabilistically expensive, not just inelegant. Every additional moving part is another opportunity for the whole structure to be wrong. And the cost compounds silently — elaborate theories feel more thorough precisely because they account for more detail, which disguises the fact that each accounted-for detail was purchased with another assumption.

This also explains why the razor sits underneath several other well-known heuristics. Hanlon's Razor — never attribute to malice what is adequately explained by incompetence — is a direct application. Malice requires assuming intent, coordination, competence at concealment, and sustained secrecy. Incompetence requires assuming someone was busy or careless, which is universally attested. The unflattering-but-mundane explanation is nearly always assumption-cheaper than the sinister one.

Where the razor misleads

An honest treatment has to include the failure modes, because the principle is over-applied at least as often as it is under-applied.

Reality is sometimes genuinely complicated. Immune systems, economies, and ecosystems are not simple, and a satisfyingly simple model of any of them will be wrong. The razor selects among available explanations; it does not promise that the truth is tidy. Preferring simplicity is a betting strategy, not a description of the universe.

It can entrench the status quo. What counts as an "assumption" depends on what you already believe. To someone steeped in an existing paradigm, a new theory looks assumption-heavy while the incumbent's assumptions are invisible, having been absorbed as background facts. The razor applied carelessly is a machine for rejecting unfamiliar ideas — which is why it is a heuristic and not a proof.

Some things really are conspiracies. Coordinated deception exists, price-fixing happens, and cover-ups are documented. The razor says to demand more evidence before accepting the assumption-heavy explanation; it does not say to rule it out. Where evidence accumulates, the complex explanation earns its assumptions and should be accepted.

The practical discipline is narrower than the slogan and more useful. When you catch yourself constructing an explanation with multiple hidden actors, unstated motives, and coordinated behaviour — pause and ask what a single ordinary mistake would explain. Usually: everything. Then ask what evidence would distinguish the two, and go find it, because the razor's real job is not to settle the question. It is to tell you which explanation should carry the burden of proof.