Results

Introduction

THORChain is a decentralised cross-chain liquidity protocol which does not peg or wrap assets, it simply determines how to move them in response to user-actions. THORChain's key objective is to be resistant to centralisation and capture whilst facilitating cross-chain liquidity. THORChain only secures the assets in its vaults, and has economic guarantees that those assets are safe.

There are four key roles in the system:

• Liquidity providers who add liquidity to pools and earn fees and rewards

• Swappers who use the liquidity to swap assets ad-hoc, paying fees

• Traders who monitor pools and rebalance continually, paying fees but with the intent to earn a profit

• Node Operators who provide a bond and are paid to secure the system

This dashboard will explore the Liquidity providers (LP-ers) retention since network's launch.

Source

Methodology

The first and last liquidity providing transaction for each address will be idenfitified with two different CTE, querying the thorchain.liquidity_actions tables with min(block_timestamp) and max(block_timestamp) respectively.

An additional CTE will be created to aggregate total liquidity provided in USD (both RUNE and asset) and the number of liquidity adding txs per address. The info will be then joined in to a final table as a result.

An address still providing liquidity will be defined by a parameter as the number of days since the last transaction providing liquidity. Per default, this value will be 7 days, but it can be changed by the users to see the implications.

Two visualizations for the % still providing liquidity will be shown. The first would be with no constraints, the second will be controlled by a parameter for the lower limit of provided liquidity as the sum per address. Default value will be 10k USD

Figure 1. shows the current number of addresses by month of first liquidity providing transaction. We can see a period of big new activity on May 2021, followed by a decline until November 2021, where similar level as in May appear and a new uptrend in the new year, reaching in March 2022 almost a 100% increase compared to January 2022.

Note the period of inactivity in August and September 2021 due to a missing data issue.

Defining the address which provided liquidity in the last 7 days as current LP-ers, most of them concentrate in March and April, with very little numbers before those months.

Figure 2. shows the % of addresses still providing liquidity, which with 7 days is low - 6.4%.

Figure 3. shows the % of addresses which provided at least 10k USD to the pools and still providing liquidity (in the past 7 days). Percentage increases unitl 8.3%

Figure 4. shows the total liquidity provided monthly by status. As opposite to Figure 1. we can see here how currently active LP-er play a significant role in June and July 2021 in terms of total liquidity in USD. Those levels are only comparable to April 2022 levels.

Conclusion

With the fairly small limit in the parameter which defines what a current LP-er is of 7 days, the dominance of "extinct wallet" is very clear. Both in monthly distribution based on LP-ers first transaction and total liquidity provided.

Surpringsingly, neither time factors or frecuancy of transactions seem strongly correlated to total liquidity providedd.

Figures 5. and 6. show the correlation betweenthe total liquidity provided per address and the number of LP transactions (frequency) and the days between first and last liquidity providing respectively, by status. Figure 5. have both axes in log scale, Figure 6 only the Y-axis.

Currently active LP-er seem like a group with less outliers in terms of high txs and low liquidity provided as the extinct group.

The distribution of current LP-er with regards to the number of days between first and last liquidity providing tx is also surprising. Despite the low time given for the current LP-ers, there are quite a lot of such users which did a tx 10-12 months ago (and have done one in the last week). There is also a bound cluster 6 months ago and a dispersed one 3 months ago which merges with the cluster of addresses with below 50 days between first and last.

A significant observation is also the big number of really small transactions ( <1 uUSD!!) that have happened in the last days. This is worth a deeper analysis on a additional analysis.