Why adding a third tier sells the second

A pricing page has two plans and the cheaper one wins. A third plan is added at the top, absurdly priced, and suddenly the middle plan starts winning. Nothing about the middle plan changed — not the price, not the features, not the description. Only its neighbours did.

What everyone sees

Three tiers, each with a clear feature list, arranged for easy comparison. You examine them, reject the extremes as either inadequate or extravagant, and select the sensible one. The reasoning is explicit and you could reconstruct it on request.

What is actually happening

Preferences are constructed by comparison, not retrieved from storage. Adding an option changes the comparison set, and changing the comparison set changes the ranking — which is why the effect breaks a rule economists once assumed was safe. A dominated third option makes the middle one look moderate; without it, the middle one was simply the expensive one. Restaurant wine lists, subscription pages and hardware configurators are all built on this.

Why it stays hidden

Because you did evaluate the options. The evaluation was real, careful and honest — it was just performed on a set that somebody else assembled. Nobody experiences a pricing table as an argument, so nobody audits it as one.