Unveiling the secrets of the "Coinbase Mafia" in the world of cryptocurrency

Deep dive into Coinbase's relationship network, filtering out the 10 cryptocurrency projects most closely related to Coinbase.

This article is one of the "Proof of Value" blockchain data interpretation series, interpreting blockchain with data science. This article is jointly presented by "X-Order × ChainNews".

X-Order is an innovative research institution focusing on cryptocurrency investment, open finance, and data science. It has previously collaborated with ChainNews to launch the "Proof of Value Blockchain Venture Capital Fund Ranking".

Advertisement - Keep scrolling for more content

In the Silicon Valley venture capital circle, the "PayPal Mafia" is an unavoidable topic. This term first appeared in Fortune magazine, describing the early employees of PayPal and their status in the U.S. internet industry. Since PayPal's acquisition by eBay in 2002, early key employees of PayPal chose to start their own ventures, maintaining very close connections with each other, investing in each other, and supporting each other, much like a mafia organization controlling the U.S. tech industry.

In the cryptocurrency industry, there is a similar "mafia" known as the "Coinbase Mafia."

Based in San Francisco, Coinbase is one of the most important companies in the cryptocurrency industry with immense influence. The recent valuation disclosed by the company reached $8 billion, but the employees derived from Coinbase are equally impressive. Similar to PayPal, many early employees of Coinbase chose to venture into the blockchain world, launching cryptocurrency investment funds, various underlying protocols, decentralized applications, active in various segments of the blockchain industry, and they support each other, investing in each other, and have become the backbone of this industry.

The influence of the "Coinbase Mafia" cannot be ignored. Therefore, we have decided to utilize the database resources of X-Order and ChainNews, leveraging X-Order's knowledge graph algorithm and machine learning model to deeply research Coinbase's relationship network and uncover the secrets of the "Coinbase Mafia."

We believe that uncovering the relationship network of the "Coinbase Gang" can help the industry gain a deeper understanding of the following information:

• Based on Coinbase's network and influence, explore high-quality projects

Coinbase has proven to be one of the most successful companies in the cryptocurrency industry. Its early team is more experienced, and former Coinbase employees have an advantage in identifying high-quality projects. Additionally, their network attracts high-quality resources. Therefore, projects associated with Coinbase are more likely to be of high quality, and startups within the Coinbase ecosystem have a higher probability of success. In this regard, the success of startups associated with the "PayPal Gang" in the internet industry has already been demonstrated.

• Speculate on the likelihood of cryptocurrencies being listed on the Coinbase platform based on their relationship with the Coinbase ecosystem

The "Coinbase Effect" is a well-known secret in the cryptocurrency market. This effect refers to the temporary price surge of a cryptocurrency once Coinbase announces its listing on the platform. This phenomenon is influenced by Coinbase's strict listing standards for digital assets and its liquidity advantage. There have always been rumors about "which token will be listed next on Coinbase." Instead of blindly following rumors and hearsay, it is more logical to speculate based on the relationship between a certain token and the Coinbase ecosystem.

To reveal the relationship between blockchain projects associated with the "Coinbase Gang" and their closeness to Coinbase, we have introduced a "Coinbase Content" list to filter out projects that are most closely related to Coinbase.

Utilizing the data resources of X-Order and ChainNews, we analyzed Coinbase's relationship network using X-Order's knowledge graph algorithm and machine learning model. It is important to note that the relationships we analyzed mainly include two categories:

• Work relationships: Within this layer of relationships, we further divide them into "regular employees" and "executives." For the purpose of statistical comparison, "executives" here include developers, advisors, and investors;
• Investment relationships: This mainly refers to the relationship between a project's investment source and Coinbase.

Through the knowledge graph relationship network, we identified numerous projects closely related to Coinbase. Below are the 10 projects that are most closely related:

Coda

dYdX

Oasis Labs

Filecoin

0x

Zcash

Blockstack

NEAR Protocol

XRP

Note: The investment amount indicated in the graph is the total amount of the financing round and may not necessarily be contributed entirely by that institution.

We also hope to use this knowledge graph to predict which digital assets are likely to be listed on the Coinbase platform in the future.

We assume that projects with a closer relationship network to Coinbase are more likely to be included in the inspection list. Therefore, we designed a prediction model that includes 11 relationship network feature variables and used a decision tree combined with manual intervention to select three main feature variables:

• Second-degree relationship paths for executive connections;
• Second-degree relationship paths for project connections;
• Third-degree connection paths as a secondary influencing variable in the model.

Up to now, Coinbase has announced three listing inspection lists, providing us with the opportunity for backtesting the model. We modeled the projects in the previous three listing inspection lists and trained a k-means clustering model using the last two Coinbase inspection lists and tested the Logit regression model. We used our designed model to simulate the "prediction" of the third-phase inspection list and backtested the model, which validated our assumptions.

Below are the results of the 17 projects from the third-phase listing inspection list announced by Coinbase in September 2019 in the backtesting list:

This result proves that our Logit model has a certain level of confidence and aligns with our expected assumptions. Next, we used the training set of the previous three inspection lists to train the model and made a prediction for the fourth phase inspection list:

It is worth mentioning that in this predicted list, we excluded some projects that have not issued tokens yet, as well as projects that have ceased updates or whose popularity has declined.

As Coinbase is a gateway exchange in the industry, listing is done cautiously, and projects that enter the inspection list have a certain probability of being listed, which in turn can bring a significant amount of liquidity to the projects. We hope that our model and prediction results can provide some clues for investors who are interested in Coinbase's listing plans and hope to capture investment opportunities within the "Coinbase Effect."

About the Researcher: X-Order

X-Order is an innovative research institution focusing on cryptocurrency investment, open finance, and data science. It conducts and supports interdisciplinary research in social sciences and technological developments, such as token economics, complex systems, artificial intelligence, network topology, and cognitive neuroscience. X-Order has established a non-profit fund with NGC Ventures to support cutting-edge social technology research and experiments globally. Current partners include leading investment firms, higher education institutions, scientific communities, and AI companies such as Purdue University, Swarma Club, Neureality, Shannon Technology, and more.

This article is reproduced with the permission of ChainNews, article source: ChainNews (ID: chainnewscom)

Related Reading

• Asian Exchanges Need to Tighten Up! Coinbase Paves the Way for Asia, Joins Japanese Virtual Currency Self-Regulation Organization, and Applies for Trading License

• "LongHash Column" Coinbase's Bitcoin Balance in Cold Wallets is About to Exceed One Million

Join Telegram now for the most accurate blockchain news and cryptocurrency updates!