PickleJarV2 - Adding configurations to how your farm earns

Drawing inspiration from Banteg’s post regarding a dystopian future for Yearn https://twitter.com/bantg/status/1465781218243497993?s=20

I began to think about how to improve the current PickleJars. After having thought about it for a few days and discussing it in Discord, I pitched the following idea to Larry as a V2 for the current PickleJar. He was responsive to the notion and instructed me to post it here for further discussion by the community.

My current idea is to add configurations to the PickleJars so that users could decide how they wanted their earnings to be allocated and in what proportions they would like to do so. Currently, our jars only compound LPs. However, users generally want 1 of 3 things to happen when they enter a farm:

  1. Compound LP
  2. Keep Reward Token
  3. Dump for Stables (Or ETH/native chain token)

Currently, Pickle only covers 1 of these three scenarios.

I believe that if we created jars in the way that is shown in the diagram, Pickle could become essentially the go-to protocol for farming. In the diagram, I have set up 3 different bars to indicate in what percentages they would like their rewards to be allocated out of a total of 100%. This setup would streamline the farming experience for users significantly. Especially if we pair it with a huge multi-chain portfolio of farms and a cross-chain index, I don’t think any other farming protocol that is currently on the market could compete, the closest protocol Larry and I could find being https://www.powerbomb.finance/.

This upgrade to the current PickleJar would allow both bulls and bears to participate in virtually any farm, both adding liquidity to good protocols and sanitizing the market by allowing people to quasi-short scammy coins (this in particular is important because only larger and sophisticated players such as SBF can currently do this across many farms simultaneously while pickle could allow anyone to take such positions in the market).

The difficulty of course is in aggregating everyone’s desired configurations, setting aside the stables and reward tokens, and making them proportionally claimable for each user depending on their position size and farming configuration.

10% fees should be perfectly fine for compounding and dumping to stables and perhaps we could lower the fee to 5% as a placeholder if the user reconfigures for only farming reward token (they could save on gas just waiting for an opportune time to reconfigure again depending on market conditions instead of pulling their funds out and putting them back in the original farm which could be costly gas-wise[for example, I removed half of my Spell-ETH lp from Pickle because I wanted some exposure to pure spell rewards as well which I would have left with Pickle with the proposed V2 paradigm]). Realistically, if we could pull this off well, we should have no issue in raising fees much higher as well. 15 or 20% fees would also be fair for users that decide to mix and match their configurations.

After having discussed it with Larry, he said that he would love to build it, but the matter of course would not be trivial and would require some time to put together. Once built, however, it could function in much the same way that we have been currently working to deploy the DegenZone farms and be utilized en masse to supersede the current PickleJar in most if not all cases.

Since talking to Larry, I have thought of one additional feature that we could enable/disable on certain chains depending on the quality of the chain and the depth of the liquidity available. As shown in the diagram below, this would make the process of utilizing accumulated stablecoins even more streamlined for users.

As mentioned in my previous forum post about the DegenZone, I think this version of the Pickle Jar could have some strong synergies with other parts of Pickle that are currently under discussion such as GREEN which could be the main stablecoin in question here.

Having gotten the initial sign off by Larry, I am now inviting everyone else to give feedback about this direction in development. Thanks for your time and considerations.

~ amatureApe

4 Likes

personally, I don’t think many users would want anything other than autocompound, but I could be wrong

Can some users prove me that I’m wrong?

Being long a SPELL ETH farm and earning 100% APY, you’re exposure to SPELL-ETH is X. Neutralizing further exposure from autocompounding SPELL-ETH to USD would make around 2% difference to your exposure which is 1.02X vs X exposure. That’s negligent difference drift per week in my view.
Then for a 52% APY it’s 1% exposure a week but our average farm is around 30% APY so the effect on this will be close to 0.

Our farms and jars are audited. Changing code would make them less trustworthy.

Not worth the effort and the next audit IMO. Can the users who need this tell me where I am wrong?
Want to short SPELL, why not press the red button on FTX/Binance once a week or one ca month for a predetermined amount?

Imo, you could be long SPELL in this scenario, but if you’re truly long SPELL you’d be staking sSpell on abra and not providing liquidity. You’re providing liquidity because you want the incentives.

Dumping the incentives for stables makes a lot of sense, because you then realize that incentive. I’d be interested in this, you could also allocate portions (dump 50%, compound 50%) for example, to hedge against the downside. I used to dump a lot of my rewards for DAI and stake that in alchemix, max mint alUSD to reinvest that in my LP. Costs me a lot of gas and DAI yield has gone down a lot in recent months in Alchemix. If I could do this through pickle without lifting a finger, I’d be more than happy to (especially if I get to earn yield immediatly as well).

We’re kind of off the hook with the protocols because we’re dumping only 50% of their pool2s. If wee start dumping 100% for composition like USD or ETH or a mix, the perception of our farm-and-dump could be worse.

It looks like Amatureape and Harry are keen to use such feature.
Are there more users who would like to see such feature?