Overview

In this dashboard an easy-to use "Eclipse Score" is represented. this score is meant to rank validators who are not effectively representing their delegators via voting.

when users stake their tokens to a validator, they trust them with voting power. so they should feel responsible and participate in governance. for example: if a validator participates in only 5% of votes but are in the top 5% of validators by LUNA delegated, they will have a terrible Eclipse Score here.

Methodology

Each validator has two addresses: operator address starts with terravaloper and Wallet address. \n \n When other wallets delegate Luna to a validator, they engage with the operator address. When the validator itself casts a vote, the wallet address not the operator one signs the voting transaction. Two type of addresses can be connected to each other by the creation transaction. when a validator is created, the wallet address signs a transaction creating the operator address So by searching for transactions creating a validator, a table is built: each row represents a validator containing both addresses. \n \n Wallets can stake their tokens to validators. This gave validators a voting power that they can use to cast vote in different governance proposals. Participation in governance proposals is one of the main factors of a healthy network. When a lot of wallets trust a validator or the delegated volume is high, the validator should be responsible and cast the votes it has received. \n \n Validators that aren't so active in voting in spite of high voting power should be considered weak ones. We need a metric called eclipse score determining the voting behavior with respect to voting power. \n \n ==For voting behavior, number of votes is found for each validator. \n For staking, number of different wallets delegated to a validator and the net amount staked to it are found.== \n \n these different metrics don't have same scales. So all should be normalized. For this purpose all metrics are divided by the maximum value of each metric. Consequently all will be in 0-1 range. Foe a better visualization, they will be multiplied by 100 so all in 0-100 range. \n \n Consider a case, two validators have voted in the same number of proposals so they have an identical score in voting. But one has a much higher staked volume and delegators compared to the other one. There is a question here, should they have the same eclipse score?

\n Validator with higher staked volume and delegators has done its job as expected but the other one has shown a great performance and deserves more attention. Terra wallets should consider this smaller validator when staking since it's really dedicated to terra ecoaystem and deserves higher staking volume. \n \n ==The eclipse score is defined by V/0.5*(n +s) \n V stands for voting, n for number of delegators, s for the net staked volume.== \n \n s+n stands for total staking metrics of a validator. If this number is higher, we say the validator is bigger. for validators with the same V score, the eclipse score would be higher for smaller ones. For wallets with close size in s+n, wallets with higher v would ger higher ecclipse score as expected. \n \n If a validator is active in almost all proposals and has a high staked volume and delegators, its score would be around 1. if the coefficient 0.5 wasn’t resent in the formula. this validator would get 0.5 score. it’s the reason for this coefficeint.

I will reresent validators with respect to each metric separately then I will discuss eclipse score. in the end five best and worst validators would be shown.

There are two metrics to define the staking part. first how many unique wallets have delegated to a specific validator and second one the net amount staked to a validator. for voting behavior there is one metric: number of times a validator votes in governance proposals. in each metric the maximum amount is found. and all metrics are divided by this one and finally multiplied by 100. so all validators would have a score between 0-100.

This process establishes a rich framework to compare validators.

in the charts below we can see the distribution of validators with respect to each metric separately.

• There are a lot of validators with no votes. they will have a 0 eclipse score.
• apparently there are a few top validators in each category that has made the score of others so less. pay attention that many validators have a score below 1 in a 1-100 range.

now it’s the time to look at these metrics in a pair. we will have 3 pairs that their scatter plots are below.

• There is a slight linear trend in all three plots. as number of one metric increases, the other metric increases as well. It’s more obvious in volume-delegators plot.
• There are many small wallets in both volume and delegators that have a well voting score. they will have a good eclipse score and deserves more attention.
• There are a few all around validators with high score in voting and volume or delegators. they will get an eclipse score close to 1. they have done their jobs.

now it’s time to look at the eclipse score itself.

first you can enter any validator in parameters section to see its different scores. and then we have a fully detailed table of validators with all four scores.

Now let’s see the distribution of wallets with respect to eclipse score.

• There were a lot of validators with no voting. all will have 0 eclipse score. 66 ones specifically. in the chart below they have been omitted for better visualization. all these validators are the worst case ever they haven’t cast even one vote. it’s more than 65% of total validators.
• there are a lot of small ones. less than 5% have a voting score above 10.
• ==pay attention that this score is not between 0-100.== in theory between 0 and infinity. a small wallet with high voting score and low volume and delegators would get a high score here.
• if a wallet gets a score around 1, it means that it has done a balanced job. and has voted according to its volume.

It’s worthy to look at the 5 best wallets and 5 worst wallets.

I had to omit two validators with unusual high eclipse score. they were inactive ones with just one vote and a very staked volume that made their eclipse score high. we are not interested in disabled ones.

Top 5 wallets are small ones in delegators and volume metrics. they deserve more attention. they have casted votes in many proposals. so a wallet when staking should consider these top 5. they will do their duty well.

On the other hand we had a lot of wallets with no votes all 0 eclipse score. from them I selected 5 with biggest staking volumes. they aren’t active in voting at all and don’t deserve high staking volume.

Conclusion

there are a lot of validators with high staking volume that aren’t active in governance proposals. on the other hand, there are many small validators highly active in voting.

when delegating, wallets should pay attention to this factor and choose validators that do their job and help the ecosystem be more healthy and stable.