"Eat Prey, Leave." Mechanic concept feedback for distributed partial rage quit in V2 / V2.1

General
1

There have been previous discussions raised on the Forums that have proposed a concept of allowing negative voted collections to be liquidated to fund the purchase of winning collections.

There was some good feedback in this thread, and I’d like to expand upon it with some concepts I’ve designed around it.

Two community desired mechanics in the system are:

Eat-Eat

Allowing for a percentage of the bottom voted collection(s) to be dynamically liquidated, with funds moved to a Revenue Staking strategy that then subsequently funds a future epoch’s yield.

Rage Quit

The ability for members of the DAO to exit their position against a value equivalent to the TVL. If this was achieved in a sustainable manner then we would, in theory, bring token value more in line with the TVL organically over time. In theory (again) that would be beneficial to everyone and hopefully render this mechanic moot in the process.

So what’s the plan?

What I’m proposing is that these two mechanics get jiggy with it and make a horrific, yet mysteriously beautiful half-Eat, half-Quit baby. Welcome: Eat Prey, Leave.

The concept of this is as follows:

  • The bottom voted collection of an epoch is liquidated at a percentage based on the amount of negative votes to fund an intermediary pool.
  • Token holders then have the next epoch to rage quit their token holdings
  • At the end of this subsequent epoch, any residual funds will be added to a strategy that will distribute sweeping funds over the next x epochs, capping the amount released each epoch to prevent large sweeps on smaller collections (for example, selling 10% PUNK and sweeping BGAN wouldn’t make sense)

Imagine that in a current sweep war we see PUNK take the bottom spot. Through a calculation, this takes 10% of our current PUNK position, liquidates it into (W)ETH and creates a pool with a value of approximately $500,000.

This amount could then be targeted by rage quitters to exit at TVL, minus a fee based on the yield that specific collection has generated on average over a set period of time.

This penalty determined by the yield is proposed as it shows the damage caused to the DAO by actioning an early exit. The aim of this penalty is to promote exiting against collections that are not beneficial to the DAO, rather than collections that generate the majority of our yield. In this instance, PUNK is a good example as although it holds a high, stable price the yield that it generates doesn’t benefit the DAO in long term holding.

So continuing with our PUNK example, there has been 0.02 in PUNK fees collected over the past week. This is equivalent to roughly 0.1 ETH. If we assume this is average then the annual yield of our PUNK position is 4.8 ETH.

Below is a snapshot of values that were taken a couple of weeks back whilst initially looking into this:

  • TVL = 6,013 ETH

  • PUNK liquidation = 263 ETH

  • PUNK annual yield = 4.8 ETH

  • Tokens in circulation = 1,697,135

  • Current token price = $4.21

  • TVL token price = $6.71

I haven’t put together a formula for calculating the true exit price, but considering that yield is so low, I would imagine the following would be roughly accurate and we can determine a pure calculation if approved:

  • Exit token price = $6.60

The amount of yield remaining following this rage quit epoch window will then be put up as sweep funds for the next epoch up to a ceiling cap, with excess returned to the Treasury, or stay in the exit pool for future ragers.

As collections become targeted more often, we will continue to have smaller and smaller allocations, which will deter motivation for voting against it. For example, a subsequent vote against PUNK after the initial 10% would instead only yield $450k → $400k, etc.

FLOOR tokens that are used to rage quit are locked in the contract to remove from the ecosystem / burnt.

Eat Prey → Leave → Fund

I welcome any questions, thoughts and comments to this. This is something that has come from me individually, and I hold no ill will to how the team, community or public react to this.

1 Like
2

Super interesting proposal @Twade, thanks for putting this together and keen to see how the conversation unfolds. Very glad we are trying to do something more meaningful wrt restoring backing and closing the gap.

Just some initial takes after quick read:

  • If there are a large number of folk looking to leave at tv this mechanism would let a chunk leave fairly until they’re unable to negative vote punks. In return, you probably get the gap to exist again until that share grows big enough again to once again negative vote punks. It’s kind of a weird game that doesn’t seem particularly great for either side?

  • How can we better make the RQ pool bot-free. If there’s limited capacity each time I assume a FIFO system would not work great given someone can snipe in first easily. It’d have to be something like stake FLOOR if u want to RQ this epoch, after we know how many do, given the pool size do some percentage distribution.

  • I think there should be a bigger exit tax than just the yield generated by the assets. Maybe something like 5% discount to backing?

  • liquidations probably done through a Fjord pool or twap overtime or treasury manager handled? Is liquidating 10% of our fat punk stack easy?

  • Backing price should remove FLOOR tokens used in the LP as they don’t have claim on the treasury since they’d have claim on themselves making a weird circular loop. This is why buybacks etc are usually burned. Should be Backing = ($TVL - $floor_tokens_in_lp) / (circ_supply - floor_tokens_in_lp)

  • is this something that could be in place by the v2 upgrade? Seems it would have the be post v2 given audits are underway already?

it’s hard to imagine how this plays out without knowing the number of folks that would want to exit at tv rn vs hold if a protection mechanism was put in place. Do you have any estimations?

All in all, appreciate the step

3

hello this is an interesting idea

At the end of this subsequent epoch, any residual funds will be added to a strategy that will distribute sweeping funds over the next x epochs, capping the amount released each epoch to prevent large sweeps on smaller collections (for example, selling 10% PUNK and sweeping BGAN wouldn’t make sense)

Let’s say there are no rage quitters. Then the full amount of the funds will be added to a strategy that sweeps funds. Would this be too costly in terms of transaction costs of entering and exiting positions? Like would it make sense for there to be a way for rage quitters to signal if they will participate, and only then are the necessary funds liquidated, up to the set percentage as a maximum? I just worry constant exit-and-reenter will be costly for the protocol.

4

Thanks for the feedback @zon and @doodee!

If there are a large number of folk looking to leave at tv this mechanism would let a chunk leave fairly until they’re unable to negative vote punks. In return, you probably get the gap to exist again until that share grows big enough again to once again negative vote punks. It’s kind of a weird game that doesn’t seem particularly great for either side?

It would be my hope that the potential to exit at these prices would mean that the base token price would normalise towards TV, so there would be less benefit in attempting to buy tokens → negative vote → sell tokens → repeat. I see your point in when our collection bags begin to lessen, we would no longer have the appeal of TV value and price could drop.

This could be a potential issue, depending on how heavily this gets actioned. It wouldn’t just be PUNK tokens either, but that is probably the most relevant example with it holding the most and granting the least yield to the DAO. Yes, it makes up Treasury value, but realistically it won’t benefit to the extent a more active collection such as Milady makes.

I think your next point begins to touch on a potential safeguard and I can expand a little there.

How can we better make the RQ pool bot-free. If there’s limited capacity each time I assume a FIFO system would not work great given someone can snipe in first easily. It’d have to be something like stake FLOOR if u want to RQ this epoch, after we know how many do, given the pool size do some percentage distribution.

This was something I think would need to be focussed on and would help to reduce the ability for people to just buy tokens and then dump them within a singular epoch.

I’ve tried to keep solutions that keep things decentralised, but my current working theory would be that only accounts that have held FLOOR tokens for at least x epochs before the liquidation to benefit longer term holders. We could also look at what @doodee suggested and having users stake FLOOR they will want to action an RQ with when their window becomes available on a FIFO approach. Unstaking would move to back of queue, and this could either be a separate staking mechanism, or joined to the existing planning staking mechanism so that voting power is not lost.

I think there should be a bigger exit tax than just the yield generated by the assets. Maybe something like 5% discount to backing?

I like the idea; when determining the formula it would be good to look at a small tax that would continue to subsidise sweeps (collection or floor).

Liquidations probably done through a Fjord pool or twap overtime or treasury manager handled? Is liquidating 10% of our fat punk stack easy?

Fjord or Cowswap were two I looked at. 10% of PUNK would be substantial and just dumping directly into NFTX could cause quite a bit of loss. Money brains will need to ponder this one more :slight_smile:

Backing price should remove FLOOR tokens used in the LP as they don’t have claim on the treasury since they’d have claim on themselves making a weird circular loop. This is why buybacks etc are usually burned. Should be Backing = ($TVL - $floor_tokens_in_lp) / (circ_supply - floor_tokens_in_lp)

Agreed on both fronts, that FLOOR shouldn’t factor in and that RQ tokens are burnt.

Is this something that could be in place by the v2 upgrade? Seems it would have the be post v2 given audits are underway already?

I think after version 2 there will be a small version 2.1 that will introduce some small additions.

It’s hard to imagine how this plays out without knowing the number of folks that would want to exit at tv rn vs hold if a protection mechanism was put in place. Do you have any estimations?

This is a solid point and something that we should raise with the community. It is raised quite a bit in the Discord community that I see, but this could just be a loud minority, but I don’t want to make an assumption. This vote could act as a good indicator to offer “Full RQ”, “Eat Prey, Leave” or “Neither”.

At the end of this subsequent epoch, any residual funds will be added to a strategy that will distribute sweeping funds over the next x epochs, capping the amount released each epoch to prevent large sweeps on smaller collections (for example, selling 10% PUNK and sweeping BGAN wouldn’t make sense)

Let’s say there are no rage quitters. Then the full amount of the funds will be added to a strategy that sweeps funds. Would this be too costly in terms of transaction costs of entering and exiting positions? Like would it make sense for there to be a way for rage quitters to signal if they will participate, and only then are the necessary funds liquidated, up to the set percentage as a maximum? I just worry constant exit-and-reenter will be costly for the protocol.

This is a good point. The gas overheads should be relatively minimal as we would be moving an ETH value into an escrow and then remaining ETH into a strategy. If this was to go ahead, would be good to have this logic as an optional trigger so if it is no longer used, becomes defunct or unwanted, then DAO can vote to disable it / re-enable it.

1 Like
5

Twade, thanks for thinking up this concept! It’s a very FloorDAO way of doing a redemption and I appreciate the time you took to craft the concept.

I agree with zon around a small exit tax from RQ to fund future sweeps. Depending on the slippage when it comes to selling assets 3-5% seems fair.

For me, I think a redemption should be put in place before v2 goes live since v2 mechanisms have been contentious in the community and the timeline around v2s ‘go live’ date is uncertain at the moment.

If the idea is that this RQ mech is added in v2.1 then the timeline to allow people to exit is further extended. Likely 6+ months in the future??

What I would love to see is a partial redemption for folks who want to exit Floor before v2 and then the implementation of a mechanic like this to allow people to RQ periodically if FloorDAO wishes to continue promising a backed token to its holders (which it sounds like the team doesn’t want??).