How to understand Solana validator performance: the metrics that matter

Staking on Solana means delegating SOL to a validator that participates in the network’s consensus. In return, the delegator may receive rewards. But once it is time to assess a validator, the picture quickly becomes more complex: different explorers display APR, commission, stake, vote credits, skipped slots, or availability status, and not all of these figures mean the same thing.

The useful question is not which validator displays the highest percentage today. It is how the validator operates, where its rewards come from, and which metrics provide enough context to assess its performance.

This guide explains the most relevant signals.

Two validators can show similar returns at a given moment while having very different operational profiles.

One may be voting consistently with low latency. Another may have experienced a temporary period of higher transaction-fee income. A third may advertise a very low commission while not maintaining the same level of operational quality over time.

When assessing a validator, it is useful to separate three questions:

• Does it participate correctly in consensus?
• How are its rewards generated?
• What share of those rewards reaches the delegator?

APR can be a useful starting point, but it should not be the end of the analysis.

Solana validators process transactions, vote on the network state, and produce blocks when selected as leaders. Their economics can include several revenue sources that should not be automatically grouped together.

Staking rewards

The network issues inflation rewards according to the protocol rules. Validators earn credits through their votes, and those rewards are distributed between the operator and its delegators according to the configured commission.

On Solana, each stake account can only be delegated to one validator at a time. If a user wants to distribute their stake across several operators, they need to use multiple stake accounts or a solution that manages that distribution. Solana’s stake account documentation explains this structure and the activation and deactivation process for a delegation.

Transaction fees

In addition to staking rewards, validators may receive revenue linked to the transactions they process.

Solana fees have two main components:

• The base fee, of which 50% is burned and the remaining 50% is assigned to the validator.
• The priority fee, which allows a user to prioritize a transaction and is fully assigned to the validator that processes it.

These fees matter when assessing an operator’s economics, but they do not automatically represent the staking rewards received by a delegator. How a validator manages or shares additional revenue depends on its configuration and product terms.

MEV and tips

Solana can also generate revenue related to MEV and tips associated with certain transaction flows. This is another layer of validator economics, but it should not be assumed that every operator captures, distributes, or presents it in the same way.

The important point is simple: an aggregated rewards figure can combine different dynamics. To compare validators rigorously, it is necessary to understand what is actually being measured.

Vote credits and Timely Vote Credits

Validators participate in consensus by submitting votes. Vote credits reflect that participation and are part of the calculation for staking rewards.

With the activation of Timely Vote Credits (TVC), voting is not enough: voting on time also matters. The mechanism assigns more credits to lower-latency votes and progressively reduces credits when a vote arrives later. Its purpose is to prevent delaying a vote to observe which fork prevails from becoming economically advantageous, aligning incentives with faster confirmations.

For that reason, vote credits should not be read as a simple counter. They help show the quality and timeliness of a validator’s participation in consensus.

At Stakely, we developed the Solana TVC Live Tracker to make this operational layer more visible. It allows users to observe voting activity, accumulated credits, and potential credit losses in real time.

Skip rate

The skip rate measures the share of leader slots assigned to a validator in which it does not produce a block.

It is a relevant signal because leader slots are moments when the validator must be available and ready to propose a block. A high and sustained skip rate can indicate issues with infrastructure, connectivity, configuration, or responsiveness.

However, it also requires context:

• A very short period can produce a misleading picture.
• A validator assigned few leader slots will have a smaller sample.
• An isolated incident does not necessarily represent normal operations.

The useful approach is not to look for an isolated zero, but to assess whether there is consistency over time and whether anomalous periods deserve an explanation.

Availability and delinquent status

A validator can appear as delinquent when it stops voting for a defined period. This is a clear signal that it is not participating correctly at that time.

However, delinquent status is an alert, not a complete uptime metric. An operator may not appear as delinquent while having experienced brief performance issues; similarly, a one-off incident does not by itself prove that the entire operation is unreliable.

It is reasonable to review:

• whether incidents are recurrent;
• how long they last;
• whether they coincide with upgrades, public incidents, or infrastructure changes;
• and how the validator performs before and after those periods.

The Solana validator network is critical infrastructure for processing transactions and maintaining consensus. Availability is not just a figure: it determines whether an operator can fulfil its role when the network needs it.

Commission

The commission is the percentage of staking rewards retained by the operator before the rest is distributed to delegators.

It is an important figure, but a lower commission does not automatically make a validator the better choice. Commission should be considered alongside operational quality, metric consistency, and the operator’s service model.

It is also worth distinguishing between:

• commission on staking rewards;
• revenue generated from transaction fees;
• and any distribution policies for other revenue sources.

Comparing commission alone, without looking at what sits behind it, can lead to overly quick conclusions.

A sound comparison does not depend on a single screenshot. It requires reviewing data over a meaningful period and making sure the definitions are comparable.

These are useful questions to ask:

• Does the validator maintain consistent voting activity?
• How does its skip rate evolve over time?
• Has it had recurring delinquency events?
• What is its commission, and which rewards does it apply to?
• What share of the displayed return comes from staking, and what share comes from other sources?
• Does the tool used show current and historical data with a clear methodology?

It is also worth avoiding total stake as an isolated quality signal. A large delegated volume may reflect prior trust, capital distribution, or participation in specific programmes, but it does not replace an assessment of current operations.

Validator metrics change across epochs, software upgrades, network conditions, and transaction activity. That is why a point-in-time view usually says little.

The most useful way to assess an operation is to observe its evolution:

• vote-credit trends;
• consistency of participation;
• skip-rate behaviour;
• incidents and recovery;
• commission changes;
• and the consistency of information shared by the operator.

For delegators with meaningful positions, this periodic review is more valuable than chasing the highest figure in a single week.

Choosing a validator should not be reduced to a return promise. It is a decision about which operation receives delegated stake and how that operation contributes to the network’s consensus.

Staking rewards, vote credits, skip rate, availability, and commission are all parts of the same picture. None explains a validator’s quality on its own, but together they make it possible to distinguish between an appealing figure and an operation that can be assessed with sound judgement.

In Stakely’s upcoming technical report on Solana validators, we will apply this framework to data from a specific period, with methodology, charts, and network context.