Osmosis is an advanced automated market maker (AMM) protocol that allows developers to build customized AMMs with sovereign liquidity pools. Built using the Cosmos SDK, Osmosis utilizes Inter-Blockchain Communication (IBC) to enable cross-chain transactions.

In Osmosis, users can stake their OSMO to vote on governance proposals. The users with higher amount of holdings have more power in the voting system.

The validators network on Osmosis is a vital part of the ecosystem and, among other things, guarantees the security of the Proof Of Stake blockchain. You're probably wondering, how so?

Think of a validator like a banker or, even better, like a Bitcoin or Ethereum miner. They are responsible for authoring transactions and make sure to only add valid and non-tampered transactions to a block on the blockchain. Moreso, they are also required to regularly cast votes for the prices of Osmosis tokens. Misbehaviors get punished by slashing a validator. This translates to a loss of funds (including delegators' OSMO) equal to the gravity of the offense.

Validator candidates can bond their own Osmo and have Osmo "delegated", or staked, to them by token holders. The validators are determined by who has the most stake delegated to them — the top XX validator candidates with the most stake will become Osmosis validators (depending on the proposal approved).

In this dashboard we are gonna explore how decentralized is the Osmosis ecosystem. Mainly, we are gonna focus on the Props 114, 196 & 337 which were all passed in an effort to promote decentralization within the ecosystem.

To do so, we are gonna track several governance metrics to see if the proposals had some effect on the decentralization:

• How did the voting power of the top 66% of the active set validators change after these proposals were passed?

• How did the voting power of the top 66% changed after t+30, t+60 and t+90 days after each proposal?

• Power share distribution by validators rank

• How did the Nakamoto Coefficient change after these proposals were passed?

• How did the Nakamoto Coefficient changed after each proposal?

  \n

Osmosis voting power

• After the first proposal #114, the adoption didn’t be so good, so the amount of voting power for top 66% validators remained steady at 60M OSMO.
• After the proposal #196, the adoption was so good and the validators voting power jumped from the previous 60M to more than 100M in less than 5 months.
• From the last proposal number 337, it seems that the voting power is still increasing and surpassed the 120M OSMO. Let’s see what happens in the upcoming weeks.
• During the period previous to the proposals, the Nakamoto coefficient was at around 4.
• After the first porposal adoption, the Nakamoto coefficient increased to 6. After the proposal #196 the Nakamoto coefficient also increased to 7 but previous days after the most recent proposal it dropped to 6 again.
• Since the proposal #337 adoption, the Nakamotto coefficient is at 6.
• There were more old validators voting for the governance than new validators. But if we take a global view over all the proposals carried out so far, we can see that the activity has been similar over time.
• Both, the amount of active validators as voters and the number of votes by validators have been higher for old validators that the new ones. However, if we consider the average number of votes per validator over time, we can see how in this case the activity of new validators is similar to the old validators. Then, we can assume that when a new validator joined the network rapidly is adapted.
• For the first proposal, it is curious that a new validator vote against the option to increase the number of set validators, however, the amount of NOs in the old validators group reached 15.
• For the other proposals, new validators only voted YES, while some of the old validators (aprox 10%) voted NO or Abstain.

Looking at the voting power of the top 66% of the active set validators change after these proposals were passed, we can see how after the first one 114 proposal, the adoption didn’t be so good, so the amount of voting power for top 66% validators remained steady over time at around 60M OSMO.

However, after the proposal #196, the adoption was so good and the validators voting power jumped from the previous 60M to more than 100M in less than 5 months.

From the last proposal number 337, it seems that the voting power is still increasing and surpassed the 120M OSMO. Let’s see what happens in the upcoming weeks.

In the three charts above, we can see the difference of the voting power change of the top 66% during the next 30, 60 and 90 days after the proposals ends. We can see how for all of them the voting power is growing fast.

In the image on the left, we can see the power share distribution change based on the validators rank. As we can see, it seems that the top 10 validators has lost an important percentage of power from the beginning, passing from more than 0.7 power share to less than 0.45. The top 10-50 validators power share has increased from 0.3 to 0.35. These changes are good for the decentralization.

Osmosis Nakamoto Coefficient

Regarding the Nakamoto coefficient, we can see how during the period previous to the proposals, it was at 4, then it is to say that 4 validators were needed to reach the bizantine tolerance fault to have more than a half of the voting power.

After the first porposal adoption, the Nakamoto coefficient increased to 6. The number registered an increase after the proposal #196 adoption, passing from 6 to 7, but few weeks before the last proposal adoption, the number come back to 6. Now, since the proposal #337 adoption, the Nakamotto coefficient is at 6. This uptrend is good for the decentralization as more validators are needed for the fault tolerance.

Overview

Methodology

Key insights

Validators activity

In the first chart we can see the validators’ activity for the three aforementioned proposals. We can see how in this case there were more old validators voting for the governance than new validators. But if we take a global view over all the proposals carried out so far, we can see that the activity has been similar over time. Both, the amount of active validators as voters and the number of votes by validators have been higher for old validators that the new ones.

However, if we consider the average number of votes per validator over time, we can see how in this case the activity of new validators is similar to the old validators. Then, we can assume that when a new validator joined the network rapidly is adapted.

Governance proposals results

Finally, if we take a look at the proposal results and splitted them by type of validator (new or old), we can see how at some point, their intention differ.

For exampe, for the first proposal, it is curious that a new validator vote against the option to increase the number of set validators, however, the amount of NOs in the old validators group reached 15. For the other proposals, new validators only voted YES, while some of the old validators (aprox 10%) voted NO or Abstain.

In all of the cases, we can see how the governance proposals to increase the set validator number has been approved.