Pump and Dumps in the Bitcoin Era: Real Time Detection of Cryptocurrency Market Manipulations

TL;DR

This paper investigates pump-and-dump manipulation in cryptocurrency markets, detailing how online groups organize, signal, and execute coordinated attacks across exchanges. It presents two case studies, including the large Big Pump Signal community, to uncover group structure, signaling practices, and the role of obfuscated messages and social amplification. A real-time detector based on abnormal rush-order activity is developed and evaluated on Binance data, outperforming prior work with a $F1$-score around $92\%$ and detection within roughly $5$ seconds. The authors also release a dataset of $104$ pump-and-dump events and discuss implications for investors and regulatory policy.

Abstract

In the last years, cryptocurrencies are increasingly popular. Even people who are not experts have started to invest in these securities and nowadays cryptocurrency exchanges process transactions for over 100 billion US dollars per month. However, many cryptocurrencies have low liquidity and therefore they are highly prone to market manipulation schemes. In this paper, we perform an in-depth analysis of pump and dump schemes organized by communities over the Internet. We observe how these communities are organized and how they carry out the fraud. Then, we report on two case studies related to pump and dump groups. Lastly, we introduce an approach to detect the fraud in real time that outperforms the current state of the art, so to help investors stay out of the market when a pump and dump scheme is in action.

Figures (7)

• Figure 1: Screenshot of a Discord Info room, where it is explained how the affiliation system works.
• Figure 2: Messages that indicate the start of a pump and dump operation on the Streamr DATAcoin (on the left) and the NevaCoin (on the right).
• Figure 3: Big Pump Signal pump and dumps
• Figure 4: Pump on the OAX
• Figure 5: Precision recall curve for train and test sets.