Original title: Detecting unusual trading patterns on cryptocurrency exchanges by means of complexity measures
Jakub Zwydak, Marcin Wątorek, Jarosław Kwapień, Stanisław Drożdż · July 2026
Fake transactions on cryptocurrency exchanges can warp reported activity and hide how markets really operate. To spot this problem, researchers tested a new diagnostic framework on high-frequency trading data. They tracked trades, volumes, and price changes to analyze complex mathematical patterns in how trades occur over time. The team looked at Bitcoin, Ethereum, and XRP across four major crypto exchanges between April 1 and June 30, 2025. They discovered a clear anomaly on the exchange Bitget starting after mid-May 2025 for Bitcoin and Ethereum. Transaction counts spiked sharply, but total volume and price moves did not rise to match. Instead, the market flooded with tiny, erratic trades showing weakened patterns. While this behavior does not directly prove wash trading, complexity tools successfully caught hidden, artificial-looking disruptions that regular price checks missed.
Imagine you are watching a busy marketplace. Normally, you see people buying and selling items, with the number of transactions matching the amount of goods changing hands. Now suppose someone starts making many tiny, pointless exchanges—handing a coin back and forth—so the number of transactions skyrockets but the total goods traded stays the same. In this paper, the researchers use statistical tools to spot exactly that pattern: a sudden flood of very small trades on one exchange, with no increase in total volume or price movement. That pattern is a red flag for possible fake activity.
Crypto exchanges can fake trading volume to appear more liquid. This study shows that analyzing patterns like trade size and timing can reveal anomalies that price charts ignore. For traders and regulators, it offers a way to spot potential manipulation on specific exchanges, improving market transparency.
The abstract states the findings do not provide direct proof of wash trading; they only indicate features consistent with artificially increased transaction counts. The analysis is limited to three cryptocurrencies on four exchanges over a three-month period. The framework detects anomalies but does not identify the source or intent behind the unusual patterns.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Detecting unusual trading patterns on cryptocurrency exchanges by means of complexity measures — Jakub Zwydak, Marcin Wątorek, Jarosław Kwapień et al.
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.