Research
Findings built from a verified dataset of more than a thousand auction lots across five houses. Every claim carries a number, a sample size, and a way to be proven wrong.
Two thirds of everything sells under estimate. A championship is worth 4.55×. Where a lot sits in the catalogue is worth up to 14.6×. And a finding I had to withdraw.
Every sale is captured from the auction house's own published results page — the full population, not the lots I happened to call. Lot by lot, into a dataset that currently runs to more than a thousand entries across Sotheby's, Heritage, Goldin, Grey Flannel and SCP.
That sounds obvious. It isn't what I was doing three days ago, and it cost me a published finding. That retraction is here.
If you want the scoring rules and the baseline a no-skill rule achieves, that's the methodology page.