Product Update: AVv3 Improvements Plan
Dear alpacas,
AVv3 has now been live for one month. It has been a trial by fire where we saw the broader market move down sharply, which stress-tested our systems, algorithms, and strategies. Over this time period, we have learned a lot from real operating conditions and have identified various ways to improve the vaults’ performances. In this article, we’ll go over how we are adjusting the parameters to be more conservative and smooth out the return profile of the vaults.
💂Updated Guardrails
We discussed various guardrails for AVv3 in the launch announcement article. We are now making some adjustments to the parameters:
Max. Leverage
- New Value: 6x
We will reduce the max leverage of the vault from 8x down to 6x. If the max leverage value is hit, we will execute a deleverage transaction where we will close 10% of the positions and use 10% of the asset value to repay debt. Please note that the 10% value was chosen to have a “gentle” effect on the vault, but can be performed multiple times if the market continues to move in an adverse direction.
Example:
Initial State
- Vault’s total asset value = USD 100,000
- Debt value = USD 83,333
- Debt Ratio = 83.33%
- Leverage = 6x
We will deleverage the vault, by using 10% to pay back debt
Updated State
- Vault’s total asset value = USD 90,000
- Debt value = USD 73,333
- Debt Ratio = 81.48%
- Leverage = ~5.4x
Max Exposure
- New Value: + / - 3x from the vault’s target exposure
We are introducing a new guardrail that monitors each vault’s current exposure against the vault’s target exposure. For example, the target exposure for a BNB Savings Vault is 1xLong BNB. A target exposure for a BTCB Savings Vault is 1xLong BTCB, and a target exposure for a Market-Neutral Vault is 1xLong Stablecoin. So in a case where the 1xLong BNB vault goes beyond 4xLong BNB (+3x from target) or 2xShort BNB (-3x from target), it would hit the guardrail, and we will reduce 10% of the exposure.
Example:
Initial State
BNB Savings Vault: (1x BNB Long)
- Equity Value: 100BNB
- Exposure: 400BNB Long
At this state, the exposure is 4xLong
New State
- Equity Value: 100BNB
- Exposure: 360BNB Long
Note: The new guardrails will go into effect next week around September 14th for all vaults except the BNB Savings Vault. Given the BNB Savings Vault’s current state, we will wait to implement it so as not to be forced to take actions during the current adverse market conditions
❓Shouldn’t we also have a guardrail on the Price Range?
At a surface level, placing a guardrail on the price range might seem like a good risk management strategy to not allow for the LP position to be too concentrated. However, this is not necessarily always the case and would limit our flexibility to manage the vault in a more efficient manner, as illustrated below:
Better response to a sharp price movement
With a tighter range, your LP is actually not forced to buy/sell for the whole range during a sharp price movement (because with a tight range, the price would move completely out of the active range and become inactive, no longer buying/selling) Once the price settles in a new range, you can then re-range and come out better. Of course, the tight range would have to be at a reasonably low leverage.
“Hidden Range”
The hidden range concept can also be thought of as “just-in-time” liquidity, where we create an LP position with a very tight price range on the borrowed asset side (at a lower leverage) What this means is that you are only borrowing the “minimum” necessary amount of asset to create an LP, minimize the borrowing interest cost. For example, let’s look at the BNB Savings Vault farming on a BNB-USDT pool. This strategy would mean that we create an LP position with a tight lower price range (borrowing USDT.) And if the BNB price gradually moves down, then we would only borrow more USDT as required to extend the range.
Based on the logic described above, it’s better to have the flexibility on the price range and control risks from the leverage and exposure.
💰Upcoming Vaults
The performance of the crypto-crypto pairs have been satisfactory as evidenced by the BTCB Savings and ETH Savings Vaults.
Due to their high correlation, the IL in these pairs is much lower. Thus, we will be releasing more of such vaults in the coming weeks, focusing on various crypto-crypto pairs -e.g., BTCB-BNB, ETH- BNB, etc. So be sure to check our social for the announcements!
