An older authoritative page beats a fresh one when the query has no freshness demand, and a fresh page wins when the query does, so the answer splits cleanly on the query’s freshness need. For a stable, evergreen question, the page that has accumulated trust, links, and a track record of satisfying searchers usually holds the position, because age there is a proxy for proven value. For a query about something that changes, recency itself is the value, and the older page falls behind no matter how strong its history.
The deciding factor is what the searcher needs from the answer. Evergreen queries, how something works, a definition, a timeless how-to, reward depth and reliability that build over time, and an older page that has earned its authority is exactly what the searcher wants. Disturbing nothing, it keeps winning because nothing about the right answer has moved. Here, “fresh always wins” is simply false; freshness is not what the query is asking for.
When the query carries a freshness demand, the balance flips. News, anything dated like “best X 2026,” fast-evolving topics, prices, releases, rules in flux, all of these reward the page that reflects the current state. An authoritative page from three years ago can be both trusted and wrong, and trust does not rescue stale facts. Here, “authority always wins” is just as false; the searcher needs current, and current beats credentialed. So neither pole holds across the board, and the only reliable read is per query.
For your next ranking call, classify the query before you weigh the pages. Ask whether a good answer would change if it were written today: if not, the older authoritative page is likely the one to back or to match on its own strengths, and if so, prioritize recency and currency. Judge by whether the query demands freshness, and let that decide which page should win.