Comparing Cardano and Bitcoin Pools

At first glance, Bitcoin and Cardano pools might appear to function similarly. Ignoring the distinction between PoW and PoS, it may seem that pools merely produce blocks on behalf of delegators. However, the differences are substantial. The Bitcoin protocol is unaware of pool existence, whereas the Cardano protocol tracks how many pools have produced at least one block per epoch and can economically incentivize high-quality block production.

Introduction of Pools in Bitcoin and the Consequences

Satoshi Nakamoto designed Bitcoin so that rewards are given to those who solve computationally intensive tasks. Miners compete to solve these tasks in each round.

Bitcoin’s block time averages 10 minutes, allowing for 144 rewards to be distributed daily. With 10,000 miners in the network, each miner would, on average, receive a reward approximately every 70 days.

Not all miners are equal; those with higher hash rates have a better chance of earning rewards more frequently, while smaller miners are rewarded less often and have minimal chances of earning rewards.

Mining pools addressed this issue. By pooling their hash rates, multiple miners can work together under one pool. The pool operator, who runs a Bitcoin full node, creates new blocks and distributes the computational tasks among the miners. When a miner finds a solution, it is sent to the pool operator, who quickly broadcasts the block to the network to secure the reward. This reward is then shared proportionally among all participating miners.

The first Bitcoin mining pool was established in 2010.

Despite this, the Bitcoin protocol itself remains unchanged and does not recognize pools. It assigns the block reward to the pool operator’s address, who then manually distributes the reward based on the hash rate contributions of individual miners. The protocol is unaware of this distribution process.

The establishment of Bitcoin mining pools was driven by several factors, including concerns about centralization and the challenges faced by individual miners.

As Bitcoin gained popularity, large-scale miners (often with significant computational power) dominated the network. This concentration of mining power raised concerns about centralization and the potential impact on Bitcoin’s decentralized nature.

Mining pools emerged as a solution to distribute rewards more evenly and encourage participation from smaller miners. Pools provide a more consistent stream of rewards, even for smaller miners. Pools democratize mining, allowing small miners to participate effectively.

Without pools, individual miners might struggle to compete with large mining farms.

However, the introduction of pools created an imbalance in the network. Before pools, many block producers existed because miners had to run their (mining) full nodes.

Centralization in Bitcoin mining is now concentrated among a relatively small number of mining pools, which collectively control a significant portion of the network’s hash power. Although there are thousands of individual miners, the concentration of mining power in a few major pools can affect the network’s overall decentralization.

Pool operators’ nodes hold a much stronger position in the network compared to other (non-mining) full nodes. While operators are somewhat controlled by miners, history shows that miners often do not prioritize the quality of block production or decentralization. There have been instances where pool operators censored transactions without losing a significant share of the hash rate.

Currently, there are only about 25 block producers in the Bitcoin network, with the two dominant pools producing more than 50% of the blocks. The Bitcoin protocol cannot assess the quality of decentralization or actively influence it.

Thus, the responsibility for maintaining decentralization falls to the team, miners, and the community.

People are greedy and gravitate toward centralized leadership. The issue of having a small number of block producers remains unresolved and has been downplayed for quite some time. The argument often relies on the assumption that miners can respond promptly to potential attacks. Unfortunately, the Bitcoin protocol doesn’t offer a more robust security solution, leaving the community dependent on this assumption. Additionally, we shouldn’t overlook the fact that large miners could collaborate with a pool to carry out an attack. Large mining farms still pose a genuine security risk. That is a very similar threat as before the introduction of pools.

Imagine if the protocol could track the number of pools in the network, distribute rewards directly without relying on pool operators, and even incentivize high-quality decentralization. Cardano is designed to do just that.

Pools in the Cardano Ecosystem

The IOG team learned from the mistakes of its predecessors. Using Bitcoin as a model, they carefully considered the best way to implement pools and the delegation process for Cardano.

Recognizing that staking would generate similar interest as mining, and understanding that not every staker could be a block producer, the team integrated pool awareness directly into the Cardano protocol.

This integration unlocked numerous possibilities. Cardano can reward not only pool operators but also all stakers, with the distribution of rewards for stakers being independent of pool operators. The pool operator cannot keep staking rewards that belong to delegators. Another significant advantage is that Cardano can economically incentivize the optimal number of pools in the network and limit the growth of pool stakes to a certain size (so-called saturation point).

The parameter K determines the desired number of pools in the network, which is currently set to 500. To achieve the maximum reward, the ideal scenario would be to have exactly 500 fully saturated pools.

It is more beneficial for the network to have a smaller number of well-compensated, high-quality pool operators than a large number who do not earn from the business. Balancing decentralization and block production quality is essential. From a decentralization perspective, having the largest possible number of block producers is desirable. However, this can lead to longer block diffusion times across the network. Additionally, rewarding a larger number of operators results in an overall reduction in rewards for everyone. From a quality standpoint, it is more advantageous to have a smaller number of operators who receive relatively high and consistent rewards.

See how the protocol calculates rewards for pools taking into account the saturation.

A pool with a relative pledged stake ( s ) and a relative total stake ( 𝜎 ) earns rewards based on the following formula:

( R ) represents the total available rewards for the epoch (it will be explained later). ( 𝑎₀ ) is the pledge influence factor (currently 0.3). ( 𝑧₀ = 1/k ) is the relative pool saturation size (currently 0.2%). ( s’ = min(s, 𝑧₀) ) and ( 𝜎’ = min(𝜎, 𝑧₀) ) ia the relative pledge and total stake of the pool, capped at ( 𝑧₀ ).
The saturation factor ( 𝑧₀ ) prevents the formation of excessively large pools, while the pledge influence factor ( 𝑎₀ ) discourages large stake operators from splitting their stake across multiple pools. Pool splitting can negatively impact the Nakamoto coefficient, as fewer entities would be needed to control the majority of the stake.
To be clear, an operator can manage multiple pools; the protocol cannot prevent this. However, stakers can at least see which pools are operated by the same operator. Thus, they can decide to delegate to single pool operators.

The top 10 operators control 22% of the stake.

Approximately 1200 active pools in the Cardano network produce at least one block per epoch (5 days), qualifying them for staking rewards.

One reason staking rewards are lower than initially expected is that the number of pools exceeds the K parameter’s requirement by more than double.

Currently, staking rewards are around 2.5% to 3%, which is about half of the maximum possible reward per epoch.

Besides the large number of pools in the network, other factors contributing to lower rewards include the small amount of fees collected by the network, missed blocks (where slot leaders fail to produce blocks), slot battles, oversaturated pools, and low pledges.

The following chart illustrates the situation. The red line indicates the maximum possible rewards, while the green line shows the actual rewards paid to operators and stakers.

In both Bitcoin and Cardano, rewards consist of two components: a portion from the reserve and a portion from fees. The vast majority (99%) comes from the reserve, with fees being negligible.

There is a key difference between Bitcoin and Cardano. Bitcoin has a fixed block reward for four years, after which a halving event reduces the reward by half. Bitcoin cannot reduce rewards in response to low decentralization because it cannot detect its level of decentralization.

Cardano, on the other hand, does not guarantee a fixed reward per epoch. Instead, a defined amount is taken from the reserve, representing the maximum possible reward. However, this reward can be reduced based on block production per epoch.

Understanding the Cardano Monetary Policy and Reward Scheme

Periodically, a ‘virtual pot’ is filled with ADA coins every epoch. This pot has two income streams and two spending streams.

Income:

  1. Monetary expansion (reserve)
  2. Fees collected by the protocol during the epoch

Expenses:

  1. Staking rewards
  2. Project treasury

The mechanism is governed by protocol parameters.

The rate of monetary expansion is set by the parameter ρ, currently at 0.3%. For each epoch, 0.3% of ADA coins are taken from the reserve and added to the virtual pot. The amount of transaction fees collected can vary each epoch, increasing with network usage.

The T parameter determines the percentage of the virtual pot allocated to the treasury and staking rewards. T is set to 20%, meaning 80% is left for staking rewards.

During an epoch, the circulating supply remains stable, consisting of coins held by stakeholders and in the treasury. Between epochs, the circulating supply increases as ADA coins move from the reserve to circulation. The treasury grows, and all stakers (including SPOs) receive staking rewards.

Fees only circulate and do not affect monetary expansion. Cardano redistributes collected fees between stakers and the treasury according to the T parameter.

The circulating supply does not increase by exactly 0.3% of ADA coins each epoch. Staking rewards are adjusted based on pool performance. If conditions for maximum rewards are not met, a significant portion of ADA coins returns to the reserve.

Cardano has a potential maximum reward rate for block production each epoch. If certain conditions are unmet, only a portion of the reward is paid, with the rest returning to the reserve.

After the Chang hard fork, the protocol parameters will be managed by the community. This means that the community, through governance bodies, must agree to the change. For example, it will be possible to propose and approve a change to the K parameter, increasing it to a value of 1000. However, as described above, the K parameter is not only about decentralization. It is necessary to think about other aspects.

The Cardano protocol is aware of its level of decentralization and can calculate rewards to economically incentivize desired behaviors. If stakers want higher rewards, they need to delegate their ADA coins in a manner that aligns better with the protocol’s expectations.

Conclusion

For Bitcoin to achieve greater decentralization, miners would need to delegate to smaller pools. The two dominant pools currently offer zero fees, providing a strong economic incentive for miners and likely contributing to centralization. Another possibility is for the operators of these dominant pools to voluntarily reduce their hash rate by refusing new miners, though this is unlikely. Alternatively, the Bitcoin Core team could take inspiration from Cardano and attempt to rewrite the Bitcoin protocol.

Cardano has successfully maintained a high degree of decentralization for over three years, with a stable number of active pools and stakers. The primary issue is the presence of multi-pool operators, but community opinions vary. Some advocate for more support for single pool operators, while others believe certain entities deserve more trust.

While both Bitcoin and Cardano utilize the concepts of pools and delegation, the differences between their protocols are significant. In Bitcoin’s case, pools emerged after the protocol was implemented, and even today, Bitcoin remains unaware of their existence. Many people still say that a miner has mined a block, but in reality, blocks are almost always created and published to the network by a pool. Miners can only solve the PoW task for the block where the pool operator has inserted the reward address. Cardano economically incentivizes pool operators and stakers to achieve the desired decentralization. This approach prevents the emergence of two (or even one) dominant pools within the ecosystem. While it’s theoretically possible for two dominant multi-pool operators to arise, it’s hard to imagine a large number of stakers supporting such a scenario.

1 Like