Polymarket skill test raises questions about high-price contract buyers
A Medium write-up rebuilding a London Business School and Yale sign-randomization skill test on 1.8 million Polymarket trades found the test flagged the sample account and 612 others like it, whose common trait was buying contracts at 99 cents or more. The author also found 59 percent of simulated no-edge favourite buyers were called skilled, and a standard multiple-testing correction removed none of them.
A reconstruction of a London Business School and Yale sign-randomization skill test using 1.8 million Polymarket trades identified a sample account and 612 similar accounts as skilled, according to a Medium write-up. Their shared characteristic was buying contracts priced at 99 cents or more.
Simulations challenge the skill label
The Medium author also tested simulated favourite buyers with no edge. The procedure classified 59 percent of those simulated buyers as skilled, raising questions about whether its classifications distinguish trading ability from the strategy being used.
A standard correction for multiple testing removed none of those classifications, according to the write-up. In this exercise, that adjustment therefore did not resolve the concern exposed by the no-edge simulations.
Reading the results cautiously
The distinction is between receiving a statistical skill label and demonstrating an actual edge. The reported results suggest that buying very high-priced favourites can produce the former under this test even when the simulated traders have no edge.
That does not establish whether the sample account or the 612 similar accounts possess genuine skill. It does mean the classifications, on their own, leave that question unresolved.
What it means for prediction markets
- Account rankings need scrutiny: A skill designation from this test may reflect purchases at 99 cents or more rather than independently establish an edge.
- No-edge simulations matter: The 59 percent classification rate provides a direct check on how the procedure treats simulated favourite buyers.
- Multiple-testing adjustments have limits here: The reported correction removed none of those classifications, leaving the central interpretive concern intact.
Sources & attribution
- Published
- Source
- Medium
- Publisher
- The publisher did not list a web address.
- Source URL
- The feed did not publish this story's article address.
Full attribution under the Digital Millennium Copyright Act (DMCA), 17 U.S.C. § 512: this article is an original report by Infiniti Intelligence that summarizes the reporting of the publishers named above, links to their original work, and reproduces no protected expression beyond brief factual reference and in-line attribution (“according to Medium”). All rights in the underlying reporting remain with the publishers listed; nothing here implies endorsement by or affiliation with them. Rights holders may request correction, update or removal through the contact address published at infiniti.w3ai.io.
Related
- Russia-Ukraine peace talks by December 31, 2026 priced at 74.0% on Polymarket, down 7.5 points, on $509K traded
- Prediction markets face midterm accuracy test after misses in Brazil, Serbia, Wisconsin and Israel races, Wall Street Journal reports
- NBA names Kalshi and Polymarket its preferred prediction market exchanges and signs a CFTC integrity agreement, CoinsNews reports
- Study finds Polymarket or Kalshi on page one in half of sampled midterm forecast searches, Business Wire reports
- Tallahassee mayor market moves to Loranne Ausley at 74 percent on Kalshi and 72.5 percent on Polymarket, Tallahassee Democrat reports
- Polymarket Bitcoin Up or Down on November 1 market prices Up at 82.5 percent, up 32 points, PolymarketTrader reports