Study Finds Repeat 'Rug Pulls' on Pump.fun Crypto Platform

A study of repeated, team-associated selling around sharp collapses of Pump.fun tokens in deliberately selected data windows separates measurement from claims of fraud, common ownership or realized profit. Because price-based measures are not fully reliable, the researchers compare a price-based screen with one using only bonding-curve reserve and SOL amounts; most price-based episodes also appear under the reserve screen. Episodes cluster well beyond a within-hour label-shuffle benchmark, and most identified collapses were sold by a single creator wallet.

Recorded own-token cash flow for identified groups is positive in most windows and negative in a few, but the negative sign is fragile—it changes with the screen, group definition, window length, and trading costs of about two per cent per side. A daily SOL price covariate shows no association with episode incidence. An agent-based simulation of a team exit on a bonding curve, extended with latent attention and calibrated to marginal moments, finds the team’s result depends mainly on token attention, fees, and the team’s own trading; continual team trading is costly, and reacting to other traders’ inflow adds little. The model reproduces the sign of group-level results but not the common early exits. Missing token transfers, verified migration labels, and human-labelled validation prevent identifying operator profit or fraud prevalence.