Liquidity Provision with Gravity Silos
Automation vs Manual LP Management
When it comes to Concentrated Liquidity Provision, the majority of users manage their LP positions manually on their chosen DEX. In this article we examine the latest Silos Strategies for Liquidity Provision on Shadow DEX, and we compare our Automation results with our Manually Managed Positions.
TLDR:
- This past week we ran 4 separate LP positions on Shadow DEX for stS/USDC
- 2x using Gravity Silos with Automation and 2x manually managed positions directly on Shadow UI
- Each position started with $500 USDC
- During the testing week, the value of S declined approx. 6-7%
- At the end of the week, both Silos ended with a net positive balances, while both manually managed positions ended with negative balances
- Silos are set and forget and require 1 single deposit
- Manually managing liquidity requires literally hours of monitoring and manually actioning rebalancing and claiming rewards
Liquidity Provision, Risks, and Rewards
It’s no secret that providing liquidity on a DEX carries risk, namely Impermanent Loss. These risks are often offset by the DEX issuing rewards to Liquidation Providers.
With concentrated liquidity provision, these risks are somewhat reduced through the use of “active ranges”, however this introduces a new issue, “staying in range” or you stop earning rewards!
The latest Silo Strategy — stS/USDC
This week, we have been testing the latest Gravity Silos strategy, a “simple” liquidity provision on Shadow DEX for stS/USDC pair.
We created two versions of this LP strategy, the first uses a 2% active range (considered aggressive) earning significant APR, and the second uses a 4% active range (considered less aggressive compared to the 2% range), with approx. half the rewards of the 2% range.
Why 2% and 4%?
You might be wondering why we decided to create two versions of the same strategy, so first lets explain a couple of things about manually providing liquidity and how Silos differ.
Manually providing liquidity to a DEX requires the users to do several approvals (for the two tokens being used), and then a deposit for the two tokens to actually add the liquidity to the DEX (lets say 4–5 txs at most). On a blockchain like Sonic, this will generally cost less than 0.02 S for the approvals and the deposit, and using the Shadow UI to provide liquidity is a straightforward process if you already have the two required tokens.
By comparison, Silos allows a user to deposit 1 single token (in the case of these strategies, USDC, stS, or wS) and the Silo automatically does all of the approvals, calculations and liquidity positions for the user, but at a slightly increased cost (approx. 0.15 S in our recent tests).
When providing liquidity manually, there might be 4–5 onchain transactions, by comparison, the initial deposit into a Silo can trigger 20–30 actions (with a single click), with the Silo handling every required action for the user, i.e. all approvals, swaps, liquidity provision etc.
The manual process takes the average user around 5 minutes of clicking buttons (longer if they have to calculate positions or swap tokens prior to providing liquidity, but in the interest of fairness we started with the correct tokens and required balances). Silos allows for one single transaction which completes all actions almost instantly due to the super speed of the Sonic blockchain combined with the pure power that is Gravity Silos!
LP Rebalancing, Costs and Considerations
The cost of the initial deposits, be it manual or with Silos, are fairly similar to each other, manual does use less gas but not by much, however as you can see from above, Silos does a lot for the user, and the main reason for the two options is due to the volatile nature of S token in recent weeks and the cost and number of LP Rebalancing seen in previous rounds of testing.
During initial internal QA testing, we found that the 2% range was having to be rebalanced often as the liquidity position kept exiting the reward-range, we wanted to see how well a 4% range would react to market fluctuations and compare the two ranges for rewards vs costs of operation.
For comparison, manually rebalancing a liquidity position on Shadow requires the user to;
- Check the current LP make-up (stS amount vs USDC amount)
- Calculate how much of the liquidity to remove from the existing position
- Swap some of the removed tokens from one token to another
- Add the new liquidity to the existing position
In testing, we found it was simpler to just remove all of the LP then convert the required portion from one token to the other, then re-add the LP.
The cost to manually rebalance an LP in either case worked out to be less than 0.05 S, however manually rebalancing an LP took around 10 minutes each time the position had to be altered, this also includes claiming and compounding the reward tokens into the LP.
The Rules for our Test
Due to the time requirements of manually managing the LP, we decided that it would be “fair” to check the LP on a regular schedule which we felt would align with the majority of users availability, so once in the morning “before work time”, once at lunch time, once “after work time” in the evening, and once “before bed” late at night. We set ourselves similar parameters to our Silos, i.e. if the LP was out of range when checked, we had to rebalance the position. If the earned rewards were above $10 when checked, we had to claim and compound into the position.
Similarly, Silos was set up to stay in range (2% and 4% respectively on the 2 strategy versions), and to claim and compound when rewards were above $10.
We opened the 2 Silos and the 2 manual positions within a couple of minutes of each other to give them all a starting position as close as possible to each other, and we added $500 USDC to each Silo and $500 worth of USDC/stS to each Manual position.
The Results
Over the course of a week, we monitored the manual positions, claiming rewards when above $10 and rebalancing our LP manually if needed when we checked the positions at our pre-determined 4 times per day. At the same time that we checked the manual positions, we also checked and noted down the values and gas usage of the automated Silos.
After 1 week, the results were clear;
Silo 1, 2% Range, finished with a final balance of $514.10 and 18.11 S total gas used (approx. $6), giving this Silo a final balance of 1.62% profit.
Silo 2, 4% Range, finished with a final balance of $508.55 and a 6.69 S total gas used (approx. $2.20), giving this Silo a final balance of 1.27% profit.
Manual LP 1, 2% Range, finished with a final balance of $485.26 and 0.89 S used (approx. $0.29), giving this manual LP a final balance of approx. 3% loss.
Manual LP 2, 4% Range, finished with a final balance of $485.79 and 1.074 S used (approx. $0.35), giving this manual LP a final balance of approx. 2.91% loss.
Considerations
- During this week of testing, the value of S declined by approx. 6–7%
- Both Manual LP’s lost approx. half of the declined percentage of S value
- A user who is inclined or able to monitor and manually reposition a manual position more often could possibly reduce the loss with enough time and effort invested into managing the positions
- Both Silos managed to make a profit during this time when the market declined significantly, despite the increased gas costs to rebalance the LP to stay in range
- All positions compounded the rewards earned into the primary LP position
Note that in a sideways market, or in a market where the value of the underlying assets are trending up, Silos can perform even better than in a declining market. During previous testing we found that the value of the Silos with compounding rewards was trending up at approx. 2–3% per day, however in a declining market Impermanent Loss is still a consideration even with Silos, and as such the returns are reduced somewhat.
Closing
In closing, the Automated Silos outperformed the market and faired considerably better than the manually managed liquidity positions.
Even in a declining market, the “2% Range” Silo was on track to earn approx. 135% APY with daily compounding, and that is after considering impermanent loss and gas costs, and on a fairly small initial deposit of just $500.
As an example, a deposit of $5,000 would be expected to see 10x the rewards compared to the $500 deposit and the gas costs would remain unchanged, i.e. $5,000 initial deposit would have ended the week around $5,141 with $6 worth of gas costs, resulting in a 2.7% return in 1 week (over 285% APY if compounded daily).
Want to get involved with the Gravity Community, come and join our Discord server and make sure to follow us on Twitter to stay up to date with the latest announcements and news.
Links
Website: https://gravityfinance.io
Twitter: https://twitter.com/Gravity_Finance (@Gravity_Finance)
Discord: https://discord.gg/gfi
