Validator Selection

Learn more about Meta Pool Principles of management

Meta Pool follows the NEAR Foundation strategy to spread the stake among the long tail of high-performance, low-commission, non-concentrated validators, in order to increase decentralization and censorship resistance for NEAR.

We evaluate all validators, in our main product we do not have whitelists, and we publish the formula, the code we use to compute the formula and also the results that will be on chain. The ethos is to be totally transparent and open to all validators. =

We have 4 criteria for selecting validators on the NEAR network, that are in line with our ethos of transparency and fairness:

1. Uptime > 95%

2. Fees < 10%

3. Making sure the validator doesn't reduce the Nakamoto Coefficient

4. Have enough tokens to accomplish the seat price of NEAR Protocol

How the NEAR stake is distributed?

Every epoch, this tree stage process is executed to distribute the NEAR available to stake:

1. Stake: if people stake NEAR during the epoch, the NEAR is distributed with this procedure:

   1. We create a list of top 50 validators by performance ( average APY for the last 5 epochs)

   2. We create a list of most voted validators (with more than 5% of votes)

   3. We merge both lists and start going thru the list staking if the following conditions are met:

      1. The validator has more assigned NEAR (performance+votes) than staked NEAR

      2. The validator APY is in the top 50 performant validators

      3. The validator comission is <= 8%

      4. The validator has received less than 20K NEAR this epoch

      5. If all the conditions are met, the validator is staked with a cap of 20K NEAR per epoch, and never more NEAR than currently assigned (performance+votes)

2. Unstake: If some validator goes offline for several epochs, it will be fully unstaked, and after 4 epochs, the above procedure applies to distribute the stake

3. Rebalance: Every 4 epochs, at most 0.15% of all the stake is unstaked from the worst performant validators with more staked than assigned. After the unstaking delay, that 0.15% goes to the procdure 1) to be distributed. This rebalancing process has a cost so it must be limited in the amount, because we don't receive rewards for that 0.15% being rebalanced during the 4 epoch delay for unstaking.

Note: The data, performance, votes, NEAR assigned, can be viewed in real-time here: www.metapool.app/stakevote

Note 2: You can verify the actual list of Meta Pool staked validators using this 3rd party site by entering meta-pool.near in the account filter