A big feature of the V3 PoolTogether Protocol has been the ability to provide both deposit rewards and referral rewards. These rewards were initially funded by the PoolTogether core team for the first 8 weeks of the protocol.
Strictly analyzing the numbers, rewards in PoolTogether V3 have not been successful. The initial 8 week trial period is almost complete and by analyzing the rewards contract we can see:
Only $5,351 of the $16,000 in award allocations have been claimed
Only ~10% of current depositors have claimed rewards at all (387 unique claim events)
255 of the claims were for reward amounts of LESS than $1 meaning they users actually lost money (due to gas fees) in claiming the rewards.
Additional to these stats, the current implementation of rewards has a major downside of adding roughly 200k gas limit to every deposit and withdraw (roughly $4 at a gas price of 40 Gwei)!
For these reasons, I do not believe continuing rewards in the current iteration makes sense. They are not widely used and add to much to gas costs.
Rewards 2.0
The above notwithstanding, I do believe rewards are still an important part of protocol growth – they just need to be implemented better.
At a high level I would propose creating a rewards staking contract. This would keep the gas fees for depositing to an absolute minimum but once depositors have tickets they could deposit them into the staking contract. This also opens the door for us to reward users from multiple different pools and many other advantages a more flexible approach can have. I haven’t thought through the details of this but I think this can be a starting point for community discussion!
I don’t think the rewards going unclaimed is necessarily an indication that they weren’t successful.
It makes sense to not claim them very often, as gas fees are high. I assume once the rewards end, a large part will be claimed. (Basically every account whose claimable balance is higher than the gas fee).
I do agree that the gas costs are very prohibitive.
Can you elaborate on how you imagine the staking contract would work roughly, and why that would reduce the gas fees? It makes sense that every deposit/withdrawal will be cheaper, but if on top of that every user needs to deposit/withdraw into the staking contract and claim rewards from there aswell, that might just move the problem, instead of adressing it.
It would be interesting to do the gas math on that.
Is there a future planned for referral rewards? In traditional marketing, this is one of the most widespread techniques, so there should be more potential to be unlocked there.
I think the current iteration wasn’t advertised enough to users and not explained very well on the website.
The same is the case for deposit rewards. Displaying the APR for those is a great first step.
I agree, I don’t quite understand how a staking contract would mitigate the gas issue.
When farming, I tend to withdraw the staking rewards only once the farming period has ended to move my funds to another staking contract and continue farming.
I haven’t touched to my PoolTogether rewards cause I know it’s there and won’t disappear. I have no incentive to withdraw it yet and I think a lot of people think this way too.
So, I agree with some of the above thoughts. I’m not sure that claimed rewards are an indicator of success, rather the size of the prize pool without the rewards (or with greater rewards). For example, I haven’t claimed any of the rewards yet, although I plan on doing so soon. Personally, the deposit rewards are what made me deposit to v3, a continuous ‘yield’ similar to compound + the chance of a big prize was a pretty strong selling point.
In regards to a staking contract, I see how it could work for deposit rewards. You could move the balancedrip logic out of the deposit and withdraw functions in the comptroller, so deposits/withdraws use less gas. But, that adds an extra step in terms of UX, when I believe most users would expect to receive the deposit rewards regardless. Although this does benefit people who don’t care about the additional rewards, and want to deposit a few Dai for the prize. From the UX standpoint of a referral reward (volume drip) having to ‘stake’ the tickets feels odd.
I also agree with the point @Pierrick and @Torgin bring up – It’s likely too soon to evaluate reward usage because many people are likely simply waiting to claim them.
It’s good to hear how important rewards were for you to join V3 @treebeard.
On the gas point. It might not lower total gas costs but it would at least separate the transactions so if people simply wanted to deposit it would be much cheaper. Also, ideally a specifically designed rewards staking contract could use less gas.
From my perspective the most important thing is the flexibility a staking reward contract provides. I have two thoughts on this:
We are going to have more and more prize pools in the future. As we have more prize pools it would make sense for governance to maintain some sort of white list of what prize pools are eligible for rewards. Similar to how Balancer and Uniswap do this. Having a reward staking contract lets us do that rather than having it baked into every prize pool
It would be great to be able to customize reward rates more closely. I’m thinking of offering an accelerating reward rate the longer you are staked and therefore rewarding holders. A reward staking contract gives us flexibility to do that.
From a technical side, the one major issue here is that if people are staking their tickets in a contract and those tickets win then technically that contract will win not the person. I’m not sure how to overcome this one though…
Looks like 9500/16000 of the referral rewards have now been claimed, 2 weeks after the end of referrals.
Gas fees have been high, so there are probably still people with small balances that haven’t claimed.
But it’s clear that many people were waiting for the end of the rewards before claiming.
I have 4.21 rewards. It will cost me $17 to claim. So, I was waiting to see if it would keep growing. Would love to claim, but can’t justify it with the fees so high.
