Peak versus longevity: the choice that decides every greatest-of list

Two questions wearing one label

The phrase greatest of all time is used to cover two separate enquiries that happen to share a vocabulary and almost never share an answer. One asks how high the ceiling was, which is a question about the very best stretch a competitor ever produced under real conditions. The other asks how much was produced in total, which rewards durability, availability and the ability to remain useful long after the ceiling has lowered.

A competitor can be outstanding on the first measure and unremarkable on the second, and the reverse combination is at least as common. Once the two questions are separated the argument usually dissolves, because the disputants discover they were answering different things all along.

Why peak is hard to measure honestly

Peak assessment depends on choosing a window, and the shorter that window is, the more a single extraordinary run can dominate the judgement. A very short window rewards luck, since anyone can enjoy a stretch where marginal calls and favourable circumstances line up unusually well. A longer window blunts the very thing peak is meant to capture, because it begins folding ordinary periods back into the estimate.

There is no principled length that settles this, which is why peak arguments so often turn into arguments about which season counts as the real one. The best available discipline is fixing the window before looking at the candidates, so the definition is not quietly chosen to favour a preference.

Why longevity flatters the ordinary

Cumulative measures grow with time, so a competitor who is merely good for a very long period will eventually pass someone brilliant and brief. That is not a flaw if the question genuinely concerns total contribution, since remaining at a high level for years is itself a demanding achievement. It becomes a flaw when a cumulative total is presented as evidence of quality, because the total conflates level with the opportunity to accumulate.

Rate measures correct for this by dividing by the opportunity, but they then punish long careers that include a decline phase everyone expected. The honest compromise is to report both, and to say plainly which one the ranking is treating as decisive.

Composite scores and the hidden weighting

Many career ranking systems blend a peak component with a cumulative component, which sounds balanced until the weighting between them is examined. The chosen ratio is doing the entire job, because moving it a little reorders the top of the table without any new information entering. Systems rarely justify that ratio, and the reader is left to assume that a number chosen for tidiness reflects some underlying truth about value.

A more defensible presentation shows the two components separately and lets the reader apply the trade-off that matches their own definition. That approach produces a less satisfying single answer and a considerably more useful piece of analysis.

Deciding the criterion before the candidates

The discipline that improves these debates most is choosing the criterion before anybody names a competitor they would like to see finish first. Criteria selected after the candidates are known tend to be reverse-engineered, since it is easy to favour whichever measure supports a conclusion already held. Announcing the definition in advance also makes the resulting order falsifiable, because a reader can check whether the stated rule was actually applied consistently.

It further allows several orders to coexist without contradiction, each clearly labelled with the question it answers. A page carrying three explicitly different orders teaches more than a single confident list that refuses to say what it means.