Weighted-Average LP Duration

    What is the average duration of liquidity held in each pool, weighted by the size of the LP?

    Introduction and method

    In this Bounty, I will investigate the average duration of liquidy held in each pool (weighted by the size of each liquidity pool). The method I will use to investigate is to explore the data within thorchain.liquidity_actions table and look at what happens on an hourly and a daily basis.

    Note: There are many liquidy pools available on Thorchain, this dashboard will consist of the analysis from the following assets:

    • Native assets with RUNE: 'ETH', 'BCH', 'BNB', 'TERRA', 'BTC', 'DOGE', 'LTC'
    • Non-Native assets with RUNE: 'THOR', 'XRUNE', 'BCH', 'USDT', 'RAZE', 'ALCX', 'AAVE', 'WBTC','YFI', 'HOT', 'BTCB', 'SUSHI', 'UOS', 'LUNA', 'XDEFI', 'TWT','HEGIC', 'BNB', 'CREAM', 'UST', 'ETH', 'AVA', 'FOX', 'FRM', 'ALPHA', 'SNX', 'DAI', 'BTC', 'DOGE', 'USDC', 'GUSD', 'LTC', 'ADA', 'BUSD', 'TGT', 'PERP', 'KYL'

    Also, I'm not sure whether the way I got the difference in liquidities in different pools is entirely correct, but I tried my best writing this SQL query!

    Average Liquidity Pool Duration (HOURS)

    Loading...
    Loading...

    I find these results quite interesting and counterintuitive to some extent. The reason for saying this is that the pools in which people stay the longest are ETH-ALPHA, ETH-RAZE, ETH-KYL and ETH-SNX. I have not heard about most of those pools, and therefore I'm quite surprised that users prefer these LPs instead of the more popular ones, such as the ones provided by BNB, BTC or TERRA. I think TERRA having users stay for as little as 180-220 hours probably surprised me the most, as that is very little duration compared to the other pools mentioned earlier, which have an average stay duration of 3000+ hours!

    One of the reasons why people might not be staying as long in the native pools of TERRA is the recent price volatility, which makes users scared that they will lose their assets, whether it's LUNA or UST (even with the protection that is provided by the pools). By the looks of it, one could argue that users prefer staking their assets rather than using LPs, as providing liquidity in pools might lead them to a loss of their valuable assets (in the case where the price of the token goes up and it gets exchanged for a stable coin for example).

    Average Liquidity Pool Duration (DAYS)

    Let's now have a look at the average duration in days, instead of hours.

    Loading...
    Loading...

    Here we can see the same story, with the average time of stay in LPs being in days. As mentioned above, users stay in TERRA LPs only around 7.5-9 days on average, which is not as long as I initially thought this result would be! Again, I don't want to speculate why this might be, but the most reasonable argument to me is what I stated above - that they simply don't want to exchange their LUNA for UST because they believe in LUNA's long-term potential.

    Conclusion and further work

    This dashboard has shown the native to RUNE liquidity pools and the average duration of deposited assets from users.

    With those pretty interesting results, I think it would be nice to conduct some further analysis. For example, it would be interesting to see why users don't stay in TERRA's LPs very long, and whether my arguments provided are in fact true. To do this, I would investigate whether there is any correlation between the average duration of users staying in each LP and the price of Luna. If the price of Luna is going up and users are pulling out of LUNA-UST LPs, then maybe there is some correlation!