Bitcoin mining is facing growing suspicions, which can be summed up in a single question: what if the apparent diversity of mining pools is just a facade, and there are a few entities coordinating them all from the shadows? There is no concrete proof of this, but there is circumstantial evidence and solid arguments in favor of the idea, which will be explored in this article.

The Bitcoin transaction templates proposed by the world's largest pools are virtually identical. This similarity can be observed in real time through stratum.work.

An analysis published on February 28, 2025, details how the transaction templates submitted through pools such as Poolin, Binance, Luxor, Cloverpool, and Ultimus reflect significant similarity.

Mining pools operating under the incentive model known as FPPS (full pay per share) tend to prioritize transactions that maximize their revenue, leaving little room for individual miners to influence block composition. For some commentators in the Bitcoin ecosystem, this may pose a problem.

The FPPS model shifts the risk of not finding a block to pool operators, who, to ensure profitability, centralize decisions about which transactions to include. As a result, miners, who receive a fixed payment for their computational work, have little incentive or real decision-making power to worry about transaction diversity, leading to homogeneity in the templates submitted to the network.

While the similarity between the templates may serve as reasonable evidence that there is a coordinating entity for all large pools, there is a simpler and less dramatic explanation for the phenomenon. All large mining pools follow the maximum extractable value (MEV) logic; therefore, all transaction templates prioritize transactions with the most profitable rates, which are roughly the same.

However, a recent discovery shared by user @boerst on X, reinforces suspicions of a coordinating entity behind all the pools.

In his post, @boerst analyzed data from the Stratum protocol, used by pools to communicate with miners, and discovered a bug that simultaneously affected several supposedly independent pools, including Antpool, Binance Pool, Cloverpool, and Rawpool (the latter identified as a WhitePool proxy).

According to the analysis, a few weeks ago, these pools faced issues with their Stratum V1 job server, and over a period of 33 blocks, they all submitted approximately five templates per second.

“I found another flaw in the proxy pool matrix (…) and they were all running ~5 templates per second for ~33 blocks.” — @boerst, via X

This synchronous behavior is, in the words of @boerst, “incredibly unlikely” if the pools were truly independent. The coincidence suggests that, behind their distinct names, these pools could be operating as a single entity or sharing a common infrastructure, which would point to greater centralization than meets the eye.

Rob Warren, author of The Bitcoin Miner's Almanac, delves deeper into @boerst's templates and gives a name to the entity supposedly behind the scenes, heavily controlling or coordinating Bitcoin transaction selection. Warren mentions Antpool, rather than Foundry, the world's largest mining pool.

While Foundry may seem like the big fish of the bitcoin mining world, Antpool can actually function as the backend for seemingly disparate pools.

Warren claims that similarities can be seen in pool payouts, block size, transaction organization, distribution timing, and even code errors to identify the actual source of a template—not just the OP_RETURN attribution in the Coinbase transaction, which has the following format: “Mined by Foundry USA.” This is the method Warren assumes Mempool.space uses to determine which pools the templates come from.

The author, with technical knowledge of Bitcoin, explores further avenues to uncover organizational patterns in transaction selection that point to Antpool. “For example, if you can find templates with the same organization, arrival times, and even technical errors, you can begin to paint a picture of a network where Antpool controls substantially more hashrate than it appears,” commented Warren.

Although Warren reveals the correct method for determining what's happening in the mempool beyond Coinbase attribution, he doesn't clearly explain how he came to the conclusion that Antpool could be the entity exercising secret control over Bitcoin mining. He therefore doesn't name specific pools that are allegedly under Antpool's influence.

The question then remains: If Antpool is acting as the backend for other pools, what are these pools and why?

What is certain, however, is that there is now an alternative way—similar to the methods described by Warren—to evaluate a template's origin and data beyond Coinbase. It's called miningpool.observer, and was created by @boerst, who sparked the discussion and prompted Warren's subsequent reflections.

Mining Pool Observer uses the following methodology:  

Compares a recent block template with a newly mined block. For an ideal comparison, a template built simultaneously with the mining pool's block is required. Since this is not possible without actively capturing data from different mining pools (e.g., capturing Stratum jobs), the template creation time is assumed. 

Mining Pool Observer, mining block explorer. 

Another Developer Believes Bitcoin Mining Pools Are a Masquerade

Luke Dashjr, Bitcoin Core developer and CTO of OCEAN pool, has been one of the most vocal critics of this dynamic and one of the bitcoiners who maintains that there is no independence between the large mining pools.

Dashjr has emphasized the commonality in transaction templates:

"The templates are only the same when they are not different pools."

With this statement, he suggests that the uniformity in templates is proof that the pools do not operate independently but are coordinated or even identical.

He also criticized the infrastructure behind these pools, stating:

"They're putting different names on the same servers."

For Dashjr, this covert centralization undermines one of Bitcoin’s fundamental principles: the autonomy of network participants.

The hypothesis of centralization and hidden interests among entities participating in Bitcoin mining is not new. Before 2021, Bitmain owned Antpool, a pool that, together with Foundry USA, controlled more than half of the global hashrate. Today, Antpool owns nearly 17% of the global hashrate, according to data from Hashrate Index, and Bitmain remains the dominant manufacturer in the ASIC market.

In December 2020, ViaBTC did business with a Bitmain subsidiary called AntSentry, further demonstrating the interconnection between mining operations and manufacturers.

The combination of these elements, the uniformity of transaction templates, Dashjr's criticism of “masked pools,” @boerst's finding about synchronized errors paints a shadow of concern for Bitcoin mining, although final, conclusive evidence is lacking. However, faced with this situation, for the reasons mentioned or for others that threaten Bitcoin's decentralization, new pools have taken up the challenge.

This is the case of OCEAN Pool, which has been on the market longer and has mined more than 200 blocks, and of DMND (phonetically, “demand”), a mining protocol that uses Stratum V2, an infrastructure that allows for work negotiation by miners and, consequently, greater power for them to decide on transactions and their inclusion in blocks.