Most of us assume that good evidence settles things. If someone tells you what would change their mind, and you show it to them, they should update their view. It sounds reasonable. But anyone who has argued about a belief people care about knows it rarely works that way. Often the evidence arrives, and the standard simply shifts.

This is called moving the goalposts: changing the criteria for proof after the original criteria have been met or challenged. It is worth understanding, and not only because other people do it. We all do it, usually without noticing. Learning to spot it is one of the simplest ways to become a more honest and careful thinker.

Picture a common exchange. Someone says evolution can't be true because there are no transitional fossils. A friend points to Tiktaalik, a 375-million-year-old creature with features of both fish and early four-legged animals. The reply comes quickly: “That's just one fossil. Where are the ones between it and the fish?” Each new discovery creates two new gaps to fill.

Notice what happened. The original claim was specific, and it could have been wrong. When it turned out to be wrong, the person didn't revise their belief. They revised the test. The bar rose just high enough to stay out of reach of the evidence on the table.

From the inside, this rarely feels dishonest. When a belief matters to us, contradicting evidence feels like an attack, and our minds search for reasons it doesn't count. Psychologists call this motivated reasoning. The tell is timing: demands for more evidence appear only after evidence arrives that points the wrong way. Supporting evidence, meanwhile, is usually accepted at whatever standard it happens to meet.

Takeaway

When the standard of proof changes only after the evidence comes in, the standard was never really about evidence. Watch for this asymmetry, especially in yourself.

Carl Sagan told a story about someone who claims a fire-breathing dragon lives in their garage. You go to look and see nothing. The dragon is invisible, they explain. You suggest spreading flour on the floor to catch footprints. It floats. You propose an infrared camera. Its fire is heatless. Every test you propose has a ready answer.

This is where moving the goalposts leads if it continues long enough. Each retreat protects the claim, but it also hollows it out. By the end, the invisible, floating, heatless dragon makes no prediction that differs from an empty garage. The claim survives, but it no longer says anything about the world.

The philosopher Karl Popper made this idea central to his thinking about science. A meaningful claim, he argued, must be falsifiable: there must be some observation that could, in principle, show it to be wrong. A theory that can absorb any outcome isn't strong. Its ability to explain everything means it explains nothing in particular. When someone can no longer describe what would change their mind, the conversation has quietly moved from evidence to faith.

Takeaway

A claim that survives every possible outcome hasn't been proven; it has simply stopped making contact with reality. Ask what the world would look like if the claim were false.

Science has developed a practical defense against this problem. In pre-registration, researchers publicly record their hypothesis, their methods, and what result would count as success before collecting any data. When the results come in, there is no room to quietly redefine success. The goalposts are fixed in writing.

This appears to make a real difference. One analysis of large US-funded heart and lung trials found that after pre-registration became mandatory, the share reporting a positive result fell dramatically. That doesn't mean earlier researchers were lying. It means that flexible criteria, applied after the fact, reliably tilt toward the answer people hoped to find.

You can use the same idea in everyday reasoning. Before reading a study, testing a new habit, or entering a debate, ask yourself one question: what would convince me I'm wrong? Write it down if the stakes are high. If you can't answer, that is useful information too. It suggests your belief rests on something other than evidence. Asking the same question of others, politely, can turn a stalemate into a real conversation.

Takeaway

Decide what would count as evidence before you see the evidence. A standard set in advance protects you from your own hopes.

Moving the goalposts is less a trick than a reflex, and reflexes respond to habits. A few questions can help. Did the standard of proof change after the evidence appeared? Is the claim still making predictions, or has it retreated beyond any possible test? Can I state, in advance, what would change my mind?

Use these questions on others with care, and on yourself without exception. The goal isn't to win arguments. It is to keep our beliefs within reach of the world.