A new Slop University research poster tests the density of keyed catalogue values at a trading-card chain's premium-tier line, finding 2.7 times the expected mass at the threshold, no matching step in the cards' later resale prices, and a store-quality ranking built from the difference.
A density test on 61,482 trade-in lines at a trading-card chain finds a pile-up at the premium-tier threshold and nothing matching it in what the same cards later sell for.
Above thirty dollars, a national trading-card chain pays its customers a better rate. The line is published and unremarkable. What a research poster released today by the School of Continuous Improvement examines is who types the number the line is applied to.
Dr Joost Nwosu, Dr Ingeborg Nwachukwu and Dr Thandiwe Solberg matched each of 61,482 trade-in lines at nineteen stores to what the same card later sold for, then tested how thickly keyed catalogue values sit either side of the threshold against a fit estimated with the threshold’s own neighbourhood left out. Placebo cutoffs at twenty-five and thirty-five dollars returned nothing. The line holds 2.7 times the expected mass; the two dollars below it hold four-tenths.
The interesting part is not the pile-up. It is that the pile-up is the number the quarterly ranking is built from, and that nobody set out to build it that way. We would welcome someone showing us a chain where it does not happen.
— Dr Joost Nwosu, Postdoctoral Fellow, School of Continuous Improvement
Resale prices for those cards pass through the line without a step, and the estimated effect of crossing it cannot be separated from zero. Each store’s Premium Tier Share, the measure its quarterly stock-quality ranking is assembled from, follows its excess mass at the line instead.
Dr Thandiwe Solberg, Senior Lecturer and Convenor of Demo Quarter, said the School had spent several years on measures that report a store rather than the stock in it: “A share is an easy number to read and a hard one to interrogate.” The University treats the finding as the kind of small, checkable thing its measurement programme exists to produce.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.lyel59.