Trading Crypto with Nansen Smart Money

Trading Crypto with Nansen Smart Money

Wondering if you can get market beating returns by following Smart Money on Nansen? We put the theory to the test and share our results in this article.

Disclosure: Members of Nansen may be participating or invested in some of the protocols mentioned below. This statement acts as a disclosure of potential conflicts of interest and is not a recommendation to purchase any token. This content is for educational and informational purposes only. Please exercise caution if you are participating in some of the protocols. The information contained in this document has been provided as general commentary only and does not constitute any form of regulated financial advice, legal, tax or any other professional advice. It is intended only to provide observations and views of the Nansen personnel. Observations and views expressed herein may be changed by the Nansen personnel at any time without notice. Nansen accepts no liability for losses arising from the use of this material.


Nansen Smart Money labels have been a simple way to introduce users to the world of cryptocurrencies and create a curated view of the market. The basis of these labels comes from ranking wallet performance and identifying top performing wallets as “Smart Money”.

Figure 1: Smart-money label distribution as of end of March 2022

However, the primary use case of following Smart Money behavior has been for discovery and early identification of possible market movements. Significant knowledge of individual cryptocurrency projects and general market conditions are still required for these insights to become actionable. This report attempts to simulate a strategy of strictly copy trading based on Nansen Smart Money labels and estimate the returns associated with such a strategy.

Description of Study and Summary Results

We created a dataset of the daily token holdings balance for all Nansen Smart Money wallets on the Ethereum blockchain up to March 2022. The data uses the “point-in-time” snapshot of Smart Money wallets for each day. This generates a token universe between 500 to 1,500 unique tokens held by Smart Money wallets during this time. Using "point-in-time" data helps us avoid look-back bias when analyzing performance of Smart Money labels by using only the data that was available at the time of the transaction.

This daily token holding balance is used to generate various momentum indicator metrics and investment strategies. Each metric is tested against a set of Monte Carlo simulation of random token buys representing a range of return possibilities. Lastly, we compared each strategy to the returns against holding ETH as a benchmark.

Momentum Indicators

For this study, we used the following six indicators:

  • a. Change in the USD value of Smart Money balance by token to select for large token purchases
  • b. Absolute value of change in USD value of Smart Money balance to select for all movements, buy or sell
  • c. Change in the number of Smart Money wallet addresses holding a token to select for number of buyers
  • d. Percentage change in the number of Smart Money addresses holding a token
  • e. Change in the wallet share of one token vs all tokens in Smart Money wallets to select for allocation
  • f. Change in the wallet share of one token vs all tokens in Smart Money wallets as a percentage, to correct for the size of different wallets

For each indicator, we took the average of 15, 30, and 60 days of momentum and purchased the top 10 tokens with the biggest momentum with an equal weighted allocation on a daily basis. This portfolio is then compare against the benchmark portfolios and a Monte Carlo simulated returns.

A screenshot of a computerDescription automatically generated with medium confidence
Figure 2: Momentum Indicators vs. ETH

Out of the six momentum indicators tested, momentum indicator d, “Percentage change in the number of Smart Money addresses holding a token”, delivered the best performance against both the ETH benchmark and the 75 percentile of Monte Carlo simulated returns. The strategy delivers an average-yearly-return-to-volatility ratio of ~3.1 vs ~1.8 for a portfolio holding ETH only and an average-return-to-maximum-price-drawdown of ~6.8 vs ~3.2 for ETH, over a period covering March 2020 to March 2022.

ChartDescription automatically generated
Figure 3: Address percent change strategy vs. Monte Carlo simulations

Nansen's Take

Nansen Smart Money labels are designed to help users quickly identify market trends and token movements on the blockchain. We’re now assessing its applicability for programmatic trading by creating a framework to assess the effectiveness of different trading strategies and identifying trading indicators that can generate the best return controlling for risk.

In this preliminary research, we estimated that using even simple strategies with Nansen Smart Money labels can potentially generate significant returns. This validates our hypothesis that Smart Money wallets are an effective way to surface the signal from blockchain data. As next steps, we’ll refine our methodology and analysis to ensure our findings are accurate and actionable with the medium term goal of creating reliable indicators that investors of all types can use with confidence.

Lastly, we hope to collaborate with different actors in the ecosystem on refining the validation framework and create different benchmarks to assess performance of different strategies. Having a community driven framework will ensure our analyses are open and replicable by others.

For the more technically inclined, please refer to our research paper "Using Nansen Smart Money and Bridge Data to Simulate Tactical Investment".

Special thanks to the following Nansen Explorers for their contributions.


  • Darren Lim, Research Analyst
  • Cristian Falcutescu, Research Engineer
  • Daniel Krupiza, Senior Research Engineer

Nansen Query:

  • Ko Miyatake, Full Stack Engineer
  • Jason Xu, Sr. Product Manager

Sign up now for free

You will hear from us soon!

Join our Discord and get exclusive strategies

I'm in