In the context of Ethereum and Proof-of-Stake (PoS) networks, the Annual Percentage Rate (APR) is often presented as a clear and accessible measure of validator performance. As a summary statistic, APR seeks to answer a straightforward question: If I stake 32 ETH today, how much can I expect to have after one year?
However, APR is fundamentally an oversimplification of a highly complex system. Its role as both a measure of past returns and a forecasting tool obscures the intricate dynamics that govern validator rewards on Ethereum.
For example, using APR to predict future returns is like:
- > Using a small sample of stocks from the S&P 500 to estimate the average yearly return —similar to how APR behaves for small validators.
-> Using just 1-2 years of S&P 500 data to forecast long-term returns —similar to relying on short-term APR data like 7-day or 30-day rates.
This article aims to unpack the underlying biases of APR, explore the stochastic nature of validator rewards, and propose alternative metrics that offer a more accurate assessment of node operator performance—metrics which align more closely with operational realities. Finally, we will examine how Chorus One’s approach, incorporating our proprietary MEV-boost fork, Adagio, captures a more refined understanding of Ethereum staking dynamics. By optimizing the interaction with Ethereum’s proposer-builder separation, Adagio allows us to consistently improve validator efficiency, resulting in tangible improvements in performance without relying on the variability of APR metrics.
Ethereum validators are compensated through two primary reward streams:
While the attestation process is deterministic, rewards from block proposals and MEV are inherently probabilistic. This variability introduces a fundamental challenge: APR assumes a uniform distribution of rewards across validators, which is far from reality. The skewed nature of the reward distribution makes APR a poor proxy for expected returns, especially over shorter time horizons.
The central flaw in using APR as a measure of validator performance lies in its failure to account for the randomness that defines much of the reward structure. To illustrate this, consider the following:
As these rewards are driven by skewed distributions, their mean value—a key input for APR—becomes a biased estimator. Skewness is a measure of how asymmetrically data is distributed , see e.g. here. In probability theory, the mean of a skewed distribution is a poor representation of the typical outcome. Validators who are fortunate enough to receive multiple block proposals or sync committee assignments will see a disproportionately higher APR compared to validators who, through no fault of their own, are assigned fewer opportunities.
To further understand how randomness impacts APR, it is useful to visualize the reward distribution for validators operating at different scales.
The plot above shows how reward skewness changes based on the number of validators controlled. Precisely, higher is the skewness, longer is the upper tail, indicating that the overall distribution is asymmetric on the right. The consequence is that the mean is higher than the median. MEV rewards are the most skewed (bottom-right), meaning they vary the most between validators. Sync committee selection also has a significant impact (top-left), while block proposals have the least skew (bottom-left).
What’s clear is that as the number of validators increases, the skewness in rewards drops significantly. This means larger validator sets see more consistent rewards, while smaller sets face more variability due to randomness. The same holds true even by accounting for only a smaller time period instead of the whole year data.
-> This highlights why APR, when viewed in isolation, is not a reliable measure of performance, particularly for node operators running fewer validators.
This plot shows the distribution of simulated APR assuming different number of validators controlled. It is evident how the APR becomes reliable only when the number of controlled validators is high compared with the number of active validators (purple and cyan histograms). This is because, as we saw earlier, when more validators are controlled, the skewness in rewards decreases, making APR more reliable.
It is worth noting that, the aggregate APR of an entity controlling more validators is not the APR of a single customer, usually holding a lower number of validators. In this case, the APR of the small subset is affected by higher variance as in the case of low number of validators controlled.
However, when rewards are pooled—such as in solutions like Chorus One’s ETH staking vault on Stakewise —this variance is minimized. By pooling rewards across many validators, customers gain exposure to the performance of top-tier node operators while benefiting from a more consistent and stable APR.
In light of these insights, what should we look at when evaluating a validator’s true performance? A more reliable framework involves focusing on the operational aspects that are within the control of the validator:
These metrics provide a far more grounded understanding of validator performance than APR, which often serves more as a reflection of stochastic luck than actual skill or operational consistency.
At Chorus One, we approach Ethereum staking with a deep commitment to performance optimization. While APR figures may fluctuate due to the randomness of block proposals, we have developed sophisticated tools to maximize validator returns and minimize variance.
Central to this approach is Adagio, our internally optimized MEV-boost client. Adagio improves Execution Layer rewards by optimizing the way we interact with block builders. Specifically, we have introduced latency parameters that allow us to extract higher MEV rewards without compromising slot accuracy. This gives our validators a distinct advantage in capturing Execution Layer rewards, effectively smoothing out the variability that undermines traditional APR metrics.
Moreover, our focus on uptime and effectiveness ensures that our validators consistently outperform industry benchmarks. By maintaining near-perfect operational performance and leveraging cutting-edge tools like Adagio, Chorus One is able to deliver superior returns over the long term, irrespective of the randomness that defines APR calculations.
Source: Ethereum Network
Source: Chorus One
Over the past 30 days, Adagio has delivered an 8.45% increase in MEV rewards compared to a standard configuration without Adagio.
For real-time tracking of Adagio's MEV rewards and to explore its performance further, visit our live dashboard: Adagio Dashboard.
APR, while often used as a shorthand for node operator performance, is a fundamentally flawed metric. Its reliance on skewed distributions and random events, such as block proposals and sync committee participation, makes it a biased estimator for expected returns. Instead of focusing on APR, a more reliable approach to evaluating validator performance involves analyzing metrics like effectiveness, uptime, and frequency of fulfilled duties.
At Chorus One, our focus on operational precision and technical advancement allows us to consistently deliver reliable performance. With solutions like Adagio, we enhance reward optimization, offering staking outcomes that navigate the inherent volatility and randomness of APR-based assessments.
Staking ETH with Chorus One is effortless—just a few clicks, and you’re on your way to earning rewards. No hassle, just seamless staking.
Start staking today: https://opus.chorus.one/pool/stake/
Or, speak to our team to learn more.
Learn more about MEV and Ethereum node operator performance:
MEV:Metrics that Matter
Timing Games and Implications on MEV extraction
Check out all our research reports
About Chorus One
Chorus One is one of the largest institutional staking providers globally, operating infrastructure for over 60 Proof-of-Stake (PoS) networks, including Ethereum, Cosmos, Solana, Avalanche, Near, and others. Since 2018, we have been at the forefront of the PoS industry, offering easy-to-use, enterprise-grade staking solutions, conducting industry-leading research, and investing in innovative protocols through Chorus One Ventures. As an ISO 27001 certified provider, Chorus One also offers slashing and double-signing insurance to its institutional clients. For more information, visit chorus.one or follow us on LinkedIn, X (formerly Twitter), and Telegram.
