Hey @Curia - thanks for the response! Good observations. All learnings to be thinking about before launch future seasons/version of staking for ZKsync. I can add a bit of detail given the Gov Team was involved in the Tally Delegate Tagging project that was then used in the Staking Pilot interface for the “featured” delegate sort default.
The purpose of the Delegate tagging project was an independent project meant to help make it easier for token holders to identify a) active delegates and b) understand which key ZKsync builders and partners are participating in governance (e.g. ZKsync chains, apps, technical partners, etc.).
This tagging system was used on the Staking Pilot for the same purpose. This was particularly important given that rewards were conditional to stakers delegating to active Delegates. In the design, the “featured” filter was randomize, meaning every time someone visited the page the order of the featured list was different. It was also possible to remove the filter and search for any Delegate to delegate to. Additional information is available in the Staking Pilot FAQ under “How are Delegates displayed on the staking pilot interface?”
The concentration of delegation observed to L2Beat, SyncSwap & Matter Labs was something that was observed & monitored throughout the program. These 3 delegates were removed from the featured list mid-S1 to avoid further concentration. In the future, staking programs need to account for this. For example, the “featured” filter may need to be adjusted to automatically not include any Delegate that has over (example) 100M ZK delegated. Curious what you (and others) think about that threshold?
Good observation. As you noted, this was due to a few Delegates who were previously active becoming inactive during S1. This is a more nuanced thing to measure as participation differs from proposal to proposal/proposal type, but worth taking into consideration as how to improve future set ups.
The voting record of all 6 Delegates mentioned above has been 5/5 of the last five votes. Given active delegation is defined as voting on 2/5 votes, these folks would have to miss 3 votes before becoming inactive. Stakers are able to - and incentivized to - redelegate to a different Delegate if they become inactive otherwise they will not receive rewards. As S1 ended shortly after ZIP-15, we never saw a redelegation effort of stakers, however would expect to see stakers adjust their delegations before any future iterations to ensure they are delegating to active delegates in order to earn rewards.