If judges rate works relative to their contemporaries (keeping distributions constant) and call something 'great' only when it exceeds all that came before, the chance a new work qualifies falls roughly as 1/n. This can make later eras look artistically poorer even when underlying quality hasn’t declined. The same artifact could affect 'greatest' lists in sports, film, and literature.
— It reframes cultural‑decline narratives as potential artifacts of ranking methods, urging media and audiences to scrutinize how 'greatness' is defined before drawing civilizational conclusions.
Tyler Cowen
2025.10.15
100% relevant
The email’s model: percentile normalization plus 'greater than all predecessors' criterion implies a ~1/n probability a new observation beats the past, producing fewer 'iconic' picks over time.
← Back to All Ideas