Availability as a ranking criterion rather than an excuse

Why availability keeps getting excluded

Ratings are usually built to describe how well somebody performs when present, because that is the quantity performance data most directly supports. Absence is then handled separately, as a caveat noted beside the figure rather than as information incorporated into it. The reasoning is that availability depends heavily on chance, and it would be unfair to penalise a competitor for an accident nobody could avoid.

That reasoning is sound for any single instance and much weaker across a long record, where repeated absence starts to look like a property. The awkwardness is that the same data cannot easily distinguish bad luck from a pattern, especially with the modest number of seasons a career provides.

What a side actually plans around

From the perspective of building a squad, the relevant quantity is expected contribution across a season rather than quality in the abstract. A competitor who performs at a very high level for a small fraction of a campaign requires cover to be maintained for the rest of it. That cover has a cost in squad places, budget and the development time of whoever fills the role during the absence.

A less spectacular competitor who is consistently present therefore delivers a different and sometimes larger practical benefit to the same side. A ranking intended to inform squad decisions should reflect that, while a ranking intended to describe peak quality reasonably should not.

Durability as a partly repeatable attribute

Physiologically, tolerance for repeated load varies between individuals for reasons including tissue properties, movement mechanics and accumulated history. Some of that variation is stable enough that a record of frequent absence carries genuine information about the likelihood of future absence. A great deal of it is not stable, and a single serious contact incident tells you very little about anything except that incident.

Distinguishing the two requires understanding mechanism rather than counting days, and that distinction is a matter for qualified medical assessment rather than for a ranking. A ranking can note the pattern without diagnosing it, which is the appropriate limit of what a table should attempt.

How the choice shows up in the metric

Rate-based measures implicitly ignore availability, since they divide by opportunity and therefore treat a short season as a small but complete sample. Cumulative measures implicitly include it, because time absent produces no accumulation and the total falls behind that of a present competitor. Neither approach announces itself as taking a position on availability, and readers rarely notice that the choice has been made for them.

Publishing both figures makes the difference visible immediately and lets the reader apply whichever definition suits their question. It also identifies the competitors for whom the two measures diverge most, which is usually the most interesting group in any table.

Stating the criterion in advance

The recurring discipline in ranking work applies here as clearly as anywhere, which is to decide the rule before looking at the candidates. A rule that includes availability will be attacked whenever a highly regarded competitor misses a season, and it should be applied anyway. A rule that excludes availability will be attacked whenever somebody wins a placing on a fraction of a campaign, and it should also be applied consistently.

What destroys credibility is switching between the two depending on which result the assessor prefers in a given year. Consistency here is worth more than getting the philosophical question right, because readers can adjust for a rule they understand.