This is consistent with the notion that specific coins may be targeted more often than others. Also interesting to note is that five of the top ten most pumped coins were pumped on the Bittrex exchange. Further research could perhaps investigate the properties of these coins, in an attempt to see if there are links between the most pumped coins. The individual spikes have been muted in the figure, to highlight only the pump-and-dumps. The resulting graph depicts rather suspicious trading activity, with many periods of lower price and volume, followed by significant spikes in both.
During the 9-day period shown eight pumps were detected. Regardless of whether it is directly the result of nefarious activity, it is still a pattern which raises question. A core test of a pump-and-dump identification system is its real-world detectability. In Case 1 Fig. As a result of their coordinated efforts a large price and volume spike is visible, beginning exactly at the time at which the announcement took place.
The chart depicts the results of a pump-and-dump promoted by the group Moonlight Signal , which was signalled to commence at 4 pm UTC on the 17th of August. Exchange: Binance. Once again, the warning signals of corresponding price and volume spikes are present, and the system correctly marks the strange activity at the announced starting time as fraudulent. In this case we also observe the price and volume beginning to increase just prior to the announcement time, perhaps indicating insider trading by the group leaders.
The chart depicts the results of a pump-and-dump promoted by the group Moonlight Signal , which was signalled to commence at 4 pm UTC on the 21st of August. The pump announcement in this case was given on the 4th of September , at p. Once again, we observe corresponding price and volume spikes Fig.
The reason for this is that the price continued to climb for a while after the pump, instead of immediately dumping. Thus, we can observe that sometimes the momentum caused by a pump group may actually persist for a period of time in this case about 24 h. The chart depicts the results of a pump-and-dump promoted by the group Moonlight Signal , which was signalled to commence at p.
While our system correctly marked the corresponding price and volume spikes at the specified time, it failed to identify them as being the result of a pump-and-dump. In Case 4 Fig. Similarly, to Case 3, our system again fails to mark the anomalous spikes as a pump-and-dump, for the same reason of the price not dipping quickly enough afterwards.
In order to correctly identify these cases in which the price maintains momentum for some time after the announcement, a potential improvement could be made to the algorithm whereby decreasing volume is also taken into consideration. This paper attempted to introduce to the crime science community the problem of cryptocurrency pump-and-dump schemes.
With cryptocurrencies becoming increasingly popular, they are also becoming a more likely target for criminal activity. Cryptocurrency pump-and-dump schemes are orchestrated attempts to inflate the price of a cryptocurrency artificially. We identified breakout indicators and reinforcers as criteria for locating a pump-and-dump and investigated the data using an anomaly detection approach. We were also able to show that using a limited set of parameters it is possible to detect pumping activity in the data as well as subsequent dumping activity.
Moreover, we monitored two pump-and-dump groups in order to obtain several cases of real life pump-and-dump schemes which we then applied our detection algorithm to, in order to demonstrate its performance in real scenarios. Besides locating potential pump-and-dumps, we found evidence of clustering in the data. Translated to the environmental criminology literature, this pattern resembles repeat victimisation Farrell and Pease ; Kleemans ; Weisel ; Farrell The clustering can be exploited for preventative purposes since efforts can be concentrated towards the clusters, finding out what makes them attractive targets, and implementing strategies to help mitigate potentially nefarious activity.
Consider an exchange which requires additional verification for users trading certain symbol pairs which are determined to be vulnerable. Such an intervention would increase the effort required to trade and hence to pump the vulnerable coin.
When considering how to increase the risk, an example could be a system in which the automated detection of anomalous trading activity is used in cooperation with humans. A major challenge for pump-and-dump prevention might lie in coordinating the efforts between private bodies such as cryptocurrency exchanges and government bodies. While governments are catching up on the problem and have allocated more resources to the mitigation of pump-and-dump schemes, exchanges might have little incentive to cooperate because they benefit from trading activity on their platforms.
Finally, a move towards more government regulation—in our data less regulated exchanges were targeted disproportionately more frequently—might undermine the very concept of cryptocurrency trading as a decentralised exchange without government interference. In the current investigation, we resorted to publicly available data and provided a framework for the future analysis of cryptocurrency pump-and-dumps. However, several limitations merit attention. First, the accuracy of flagging an alleged pump-and-dump is dependent upon the parameters chosen and cannot be ascertained absent a ground truth of confirmed pump-and-dumps.
Our analysis should be treated as a first attempt to place the topic in the academic literature. Second, the dataset only covers 20 days of data with hourly granularity. While this was sufficient for the scope of this paper, future research would want to attempt to collect more substantial quantities of data and at a smaller granularity e. Third, as with any flagging system, there is a decision to be made how many false positives are acceptable i.
Arguably, an exchange would want to avoid announcing a coin of being used for fraudulent activity if this were not the case. This compromise is particularly complex in real-time settings so an interesting alternative avenue for future research might be to move towards the identification of early warning signals that can highlight suspicious trading at a point in time where the costs of false positives are relatively low e.
In order to minimise the likelihood of Type I errors i. Thus, a cost for both Type I and Type II errors needs to be determined, and a balance struck between the two. Thus, a desirable area for future research would be to create of a database of confirmed pumps.
While labour intensive to do in a fully manual way, the creation of such a database could likely be achieved through a smart combination of automated and manual tasks e. Such a database could be used as a means of testing the accuracy of a detection algorithm, as well as allowing for the use of supervised machine learning methods. Two lines of research seem particularly interesting for an extension of cryptocurrency pump-and-dump identification.
First, identifying vulnerable coins and understanding the characteristics of those coins that are repeatedly targeted in more detail would allow for efficient resource allocation of detection systems e. Second, moving away from exchange trading data, the modus operandi of pump-and-dumps could be examined in more detail.
A particularly promising path for future studies could be the linguistic analysis of the coordination of pump-and-dumps in online chat groups, on the one hand; and the means by which misinformation about specific coins is spread on, for example, social media, on the other hand. This paper has attempted to provide a first look into research for cryptocurrency pump-and-dump schemes.
Ultimately, it is the hope that the information presented in this paper will serve useful as a basis for further research into the detection of these fraudulent schemes. Bartels, K. Click here to buy the next Microsoft: the penny stock rules, online microcap fraud, and the unwary investor. Indiana Law Journal, 75, Google Scholar. Bitcoin Magazine. What is an Altcoin? Borrion, H. Quality assurance in crime scripting. Crime Science, 2 1 , 6. Article Google Scholar.
Bouraoui, T. Stock spams: An empirical study on penny stock market. International Review of Business Research Papers, 5 4 , — Brooker, K. The scary rise of internet stock scams on the net. Chandola, V. Anomaly detection: A survey. Clarke, R. Opportunity makes the thief. And so what? Crime Science, 1 1 , 3. Cryptocurrency market capitalizations. Cryptocurrency Prices. Compare cryptos to GDP of countries. Developments in Banking and Financial Law: Review of banking and financial law , 33 , 1.
Dugan, B. The internet and the law part two—Commercial matters: Facilitating and regulating commerce. Victoria University of Wellington Law Review, 33, Farrell, G. Crime concentration theory. Crime Prevention and Community Safety, 17 4 , — Once bitten, twice bitten: repeat victimisation and its implications for crime prevention. Keatley, D. Crime script analysis. Pathways in crime: An introduction to behaviour sequence analysis pp. Cham: Springer International Publishing. Chapter Google Scholar.
Khan, M. How to avoid getting duped by cryptocurrency pump and dump schemes like I did. Kleemans, E. Repeat burglary victimization. Results of empirical research in the Netherlands. Pease Eds. Crime Prevention Studies pp. Monsey: Criminal Justice Press. Kramer, D. The way it is and the way it should be: liability under sec. University of Miami Business Law Review, 13, Li, T. Cryptocurrency Pump-and-Dump Schemes. Available at SSRN Mac, R. Martineau, P.
Inside the group chats where people pump and dump cryptocurrency. Nakamoto, S. Bitcoin: A peer-to-peer electronic cash system. Shifflett, S. Some traders are talking up cryptocurrencies, then dumping them, costing others millions. Siris, V. Application of anomaly detection algorithms for detecting SYN flooding attacks.
In Global Telecommunications Conference, IEEE vol. Song, X. Conditional anomaly detection. Temple, S. Cybertrading: Financial markets and the internet. Australian Law Librarian, 8, Thompson, P. Pump and dump in crypto: cases, measures, warnings. Town, S. How to spot a pump and dump and avoid it. Commodity Futures Trading Commission.
CFTC issues first pump-and-dump virtual currency customer protection advisory. US Securities and Exchange Commission. Microcap fraud. Warren, S. How might crime-scripts be used to support the understanding and policing of cloud crime?. Weisel, D. Analyzing repeat victimization. Yang, E. The treacherous terrain of penny stocks and how firms are attempting to navigate it. Download references. JK collected the data, ran the analyses and wrote the first draft of the paper; BK and JK conceived of the idea, concept, and analysis; BK wrote the final version of the manuscript.
Both authors read and approved the final manuscript. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. You can also search for this author in PubMed Google Scholar.
Correspondence to Bennett Kleinberg. See Table 7. Reprints and Permissions. Kamps, J. To the moon: defining and detecting cryptocurrency pump-and-dumps. Crime Sci 7, 18 Download citation. Received : 03 July Accepted : 16 November Published : 26 November Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative.
Skip to main content. Search all BMC articles Search. To purchase fractional shares of cryptocurrency, specify a dollar amount instead of the number of coins. Daria Uhlig contributed to the reporting for this article. We fact-check every single statistic, quote and fact using trusted primary resources to make sure the information we provide is correct.
Check Out Our Free Newsletters! Subscribe Now. In Case You Missed It. By Jordan Rosenfeld. The pandemic put only a temporary damper on the housing market, as the market rebounded robustly, hotter than ever, particularly in As we've rolled into , the market is starting to simmer Read More. By Andrew Lisa. By Josephine Nesbit. It's important to know what you're getting into before you invest in cryptocurrency. Here's what you need to consider before buying into these top picks. Housing Market: Buying the Fixer-Upper vs.
Less Work at a Higher Price. By Cynthia Measom. The current housing market means you'll likely pay top dollar for a home that's considered turnkey -- immediately ready for you to move in. Plus, the competition is steep. Perhaps those two reasons Metaverse real estate is growing at a fast pace, both in interest and prices, and has been one of the top-grossing sectors in the NFT space in and in the past couple of months.
While in March the By Jami Farkas. Rising home values can quickly transition a reasonable housing market into the type of real estate monster that has consumed places like the San Francisco Bay The LEGO Group and Epic Games announced a partnership to shape the future of the metaverse, "to make it safe and fun for children and families," the companies said in a recent press release.
Pandemic-darling Shopify's Board of Directors has approved a proposed for-1 split of its stock, "which will make share ownership more accessible to all investors," the company announced in an April Toggle navigation Back. Sending you timely financial stories that you can bank on.