Issue

Last month the fall of LUNA price in combination with UST depeg led to the collapse of Terra blockchain and a crypto market crash.

This incident put liquidity pools work with the mechanism of pegging into a center stage. Arbitrageurs take advantage of depegs and intensify the crashing.

In theory stETH and ETH should follow the same price as stETH is minted in equal amount of ETH staked. However, the data shows that is not always true.

The graph below shows the stETH/ETH rate of LIDO: Curve farming pool over time. We can see that since May 7th the stETH/ETH rate is fluctuating.

The question is what factor(s) explain these fluctuations

Potential Explanatories

These some 'usual suspects' that ate most likely the cause of the fluctuation, or correlated with the stETH-ETH rate that can be use as indicator of the issue. In this dashboard the correlation between stETH-ETH rate and the following variables will be examined:

• ETH Price
• Add-remove liquidity volume
• Pool tvl
• Number of deposits
• Number of withdrawals
• Average amount of deposits
• Average amount of withdrawals

ETH price

Instead of staking 32ETH to become a validator, users can stake ETH in Lido and get stETH. Then take stETH as collateral at Aave to borrow ETH. Continue to stake ETH in Lido to perform the loop and increase staking interest.

When the price of ETH falls the risk of liquidation increases, many sell stETH with lower price that leads to steTH/ETH depeg.

The graph below shows the hourly price of ETH over time. We can see that after an ATH in November the price has been declining.

Add-remove liquidity volume

The symmetry of adding and removing assets into and from a pool is important factor in stabilizing the pool value. The graph below shows the daily net amount of deposit-withdraw of LIDO: Curve pool.

TVL

Total value locked can be also important in stabilizing a pool. We can see that on May 9th the TVL was at the highest level, however, since then the TVL has sharply substantially decrease.

Number of deposits and withdrawals

In addition to amount of liquidity removed or added, the number of actions can be also important. Number of actions can be an indicator of user's behaviours.

The graph below shows hoe the number of actions on the CUrve pool has been changed over time. We can see the recent increase in the number of asset removals.

Whales behaviours in crypto market are example of many. Therefore, it is very likely that their behaviours indicate phenomena.

To measure whale actions on liquidity pools we can look into the average of deposit/withdrawal actions. The more whales are involved the higher average can see.

The graph below shows the daily average of deposit and withdrawal amounts.

We can see increase in the frequency of high average amount of removing liquidity on Curve last week.

Average amount of deposits and withdrawals

Possible Explanations

The graph below shows correlation between stETH-ETH rate and daily average price.

We can see that in ETH price more than $4000 there is very unlikely to have 1:1 rate. in ETH price less than $2000 the depeg is very likely.

Discussion of findings

The historical data shows that:

• The ETH price between $2000 and $4000 is the safe zone for Curve pool. Less than $2000 and more than $4000 the depeg is very likely.

• There is no correlation between the stETH-ETH rate and LIDO:Curve can be found.

• A positive correlation can be found between the number of deposits and the stETH-ETH rate. The higher the number of deposit the less likely it is the low rate. This should be noted that this relationship can be be necessarily a causal effect.

• In the number of remove more than 50 we have always a depeg.

• When whales deposit we have 1:1 rate.

• On the other hand, in times of depeg the average amount of removals are at the lowest level.

The graph below shows the correlation between Rate and net liquidity flow. We cannot see a clear correlation between the rate and the net amount of deposit-withdraw.

The graph below shows the correlation between Rate and the Curve pool TVL (in ETH). We cannot see a clear correlation between these two variables. In any range of TVL we can see stETH depeg.

However, in the TVL around 200K there is series of depegs incidents.

The graph below shows the relationship between Rate and the number of deposits.

A positive correlation can be found between the number of deposits and rate. The higher the number of deposit the less likely it is the low rate.

The graph below shows the relationship between Rate and the number of removals.

We can see that with the number of remove more than 50 we have always a depeg.

Graphs below show the relationship between rate and the average amount of deposits as well as withdrawals in a day.

As can be seen the higher average deposit is cooccurred more frequently with 1:1 rate.

Interestingly, we can see that many depeg incidents happened when the average withdrawals were in the lowest amount.