Reliability, era and the arithmetic of a career record

Opportunity as the first factor

A career total depends on how many events were available to contest, and the length of a season has changed substantially across periods. A competitor racing in a period with a longer calendar accumulates more opportunities per season without performing any better. Career length compounds this, since remaining competitive for longer produces a larger total independently of the level maintained.

Comparing totals across periods therefore compares calendars and career lengths alongside whatever it intended to compare. Expressing achievement as a proportion of events contested removes the calendar effect and introduces the usual problem with rates. Short careers can post extreme proportions, so the rate needs a sample size beside it before it means very much at all.

Reliability and the results that never happened

Mechanical failure removes results that would otherwise have been recorded, and failure rates have fallen dramatically as engineering has matured. A competitor from a period of frequent failure has a record containing many retirements that say nothing about how they drove. Adjusting for this requires estimating what would have happened in the absence of failure, which depends on position at the time.

Such estimates are possible and uncertain, and they systematically raise the assessed level of competitors from less reliable periods. Ignoring reliability entirely treats an engineering property of the era as though it were a property of the individual.

Depth of the competitive field

The number of competitors realistically capable of winning has varied, and winning against a shallow field is a smaller achievement than against a deep one. Periods where a single team held a substantial advantage produce concentrated records that overstate the difficulty of what was achieved. Periods of close competition spread results across many competitors, which understates individual level when measured by totals.

Estimating depth requires comparing across eras, which is precisely the comparison depth adjustment is meant to enable, so circularity returns. The available response is to describe the competitive environment qualitatively rather than to apply a numerical correction that cannot be justified.

Rates, totals and what each answers

A total answers how much was achieved and is therefore sensitive to opportunity, longevity and the reliability of the machinery. A rate answers how often success followed participation and is therefore sensitive to the strength of the equipment during that period. Neither is a measure of ability, and each is a measure of a partnership under a particular set of conditions.

Reporting both, with the number of events and the era clearly attached, allows a reader to interpret them appropriately. Reporting one alone, as career lists usually do, invites exactly the cross-era comparison the numbers cannot support.

What a career record can honestly establish

A long record of success across multiple teams, regulation changes and teammates is strong evidence of a high individual level. A record concentrated in one team during a period of clear equipment advantage is much weaker evidence about the individual. The distinction between those two shapes is visible in the record itself and is almost never noted in the totals that get quoted.

A ranking that reported the shape alongside the total would communicate far more than one that ordered by the number alone. The shape is available to anyone willing to read the record properly, which makes its absence a choice rather than a limitation of the evidence. That reporting is inexpensive, and its absence is the reason motorsport career lists generate arguments that never resolve.