“We have elected to put our money and faith in a mathematical framework that is free of politics and human error.” - Tyler Winklevoss

Pancakeswap is among the most used Dexes in the whole of crypto, often its 24H volume even exceeds that of Uniswap, thus I thought it would be interesting to compare its top two Liquidity pools.

WBNB-USDT AND WBNB-BUSD

Unsurprisingly, WBNB is part of both pools. This is also a comparison between the two most popular centralized stablecoins Binance native BUSD and USDT.

A blockbuster clash is loading……

Tool used

For this analysis, I have used Covalent’s new data analytics tool called Increment which is in beta mode and available to the data alchemist bootcamp cohort.

Methodology and goal

All the key metrics of the pools have been, categorized under the RRR model(Reach, Retention and Revenue).

The analysis serves two important purposes:

• Help the Pancakeswap team to decide the reward split of any future Liquidity mining program rewards for the pools.

• Make traders and Liquidity providers smarter with their liquidity.

Reach

Reach metrics would evaluate the user penetration of both the pools.

• No of Traders

  The number of Traders graph is a story of two halves,until early 2022 BUSD pool dominated the USDT pool, however, post that period USDT pool has had more users almost every week. One reason behind it could be the ambiguity surrounding the reserves of Tether during that time which encouraged people to prefer BUSD.

  Since the start of 2022, both pools have also witnessed a decline in the number of users which is in line with current market conditions.

• No of Swaps

  Though lagging in the number of Traders in the previous chart, USDT pool has had more swaps than WBNB pool in almost all the weeks and the gap has significantly widened since February 2022. Clearly USDT is in more demand than BUSD.

• New Addresses

  Supporting the antecedent graphs USDT pool has edged BUSD in terms of gaining new swap addresses throughout this quarter.

• Pancakeswap vs Aggregator Volume

Both the pools have a negligible volume coming from an aggregator which is expected as BSC doesn’t has a huge variety of Dexes. Pancake is well ahead of any other DEX which makes aggregators less effective and users prefer to directly use the best DEX which is Pancakeswap in this case.

Retention

Retention Metrics would evaluate the user retention of the pools.

• Mint to Burn Ratio vs Volatility

Mint to Burn ratio is the ratio between the amount of Liquidity added and removed from the pools. The higher the ratio the better it is for pool retention.

It’s natural that the first week of the pools has a towering ratio as people rush in to deposit liquidity and withdrawals are rare. The Mint to Burn ratio in both pools has been identical and on average hovers in the range of 25-30. The ratio shows no correlation with the WBNB price and it has held well despite the tumbling WBNB prices.

• Trader Cohort Retention Analysis

  Cohort Retention analysis shows the Month-on-Month retention rate. The addresses are grouped into cohorts according to their first swap month.

  MoM retention rate is critical for any Liquidity pool particularly in current bearish conditions when the growth of new users is low and the pools would have to rely on the OGs to keep coming back for swaps. Both USDT and WBNB pools do well in the initial three months however post that they are able to retain only a margin of the 1st month’s users.

• Stickiness Ratio

  Stickiness ratio is the ratio between the Daily active users and monthly active users. The higher the ratio the lesser is the gap between the number of daily users and monthly users. In all the weeks since its inception, USDT pool has fared marginally better than WBNB.

  Despite the volatile landscape in the last 4-5 months the ratio for both pools has been quite stable.

Revenue

Revenue metrics would evaluate the upside for the DEX and liquidity providers of the pools.

• TVL

  TVL is the lifeblood of liquidity pools. A high TVL leads to a more efficient trading experience. During their peaks both USDT crossed a TVL of $260 million while BUSD pool reached a whopping $900 million. After the 2021 highs, the TVL of BUSD started dropping while the USDT was still soaring. It coincides with the turbulent times of Luna crash and Three Arrows Capital insolvency. In such circumstances, the users convert the volatile tokens into stablecoins and USDT having the highest market capitalization becomes the obvious choice.

  The gap between the USDT and BUSD TVL is reducing week by week and from what was a almost a 3.2X gap is now almost at parity.

  An important point to note is that the TVL below considers only the LP liquidity and doesn’t constitutes swap fee and doesn’t account for impermanent loss.

• Total Volume

  During the bull market in early 2021 both BUSD crossed a massive $5.5 Billion in weekly volume, making them among the most swapped pools in DeFi while USDT peaked at $2.3 Billion.

  But after the end of 2021, USDT pool volume has always exceeded that of BUSD pool. Both pools have witnessed a spike in the last 3 weeks probably due to the FTX saga. The lack of trust in centralised exchanges would mean that more people are moving to self-custody and using Dexes.

• Median Liquidity Added per Transaction

  Median liquidity added per transaction would give us an indication of which pool the liquidity providers trust more with their liquidity. In the early 2021 weeks, BUSD pool had an edge but the median liquidity added dropped for both pools following the 2021 Q1 only to be picked up again in mid-2022. As discussed previously this period coincided with the Luna crash when the TVL of USDT pool reached an all-time high which is in line with the high median liquidity added in USDT pool during that period.

  After the 2022 Q2, numbers have drastically dropped again but USDT has been consistently leading BUSD since then.

• APR vs Utilisation Ratio

  The utilisation ratio is the ratio between the TVL and volume of the pools the higher the utilisation ratio the more value liquidity providers are getting out of their liquidity. The graphs below validate that the APR moves in line with the pool utilisation ratio. Both pools have similar utilization and APR patterns. The APR on average has been in the range of 25-30% for both pools but as of the current quarter the APY of USDT has been slightly higher.

• Average Daily Liquidity Utilisation

  Average Daily Liquidity Utilisation indicates the efficiency of the liquidity of pools. Pancakeswap still follows the traditional constant product method where at all times the total liquidity is x*y = k. When a user adds liquidity it is then supplied at all the points at this curve. Thus the liquidity that is on the farther end of the curves seldom gets utilized. If pancakeswap had a concentrated liquidity approach like that of uniswap v3 the utilisation of pools probably would have been much higher.

  From the metrics, it is clear that the USDT pool has a far better daily liquidity utilization on average than that of the BUSD pool. This was bound to happen as USDT with lower TVL has been generating more volume than BUSD pool.

  Another point to note is that the average utilisation ratio for higher time periods like weekly, monthly would be much higher than that of daily or hourly because the liquidity of a pool can be traded infinite number of times. Thus even if the liquidity stays constant the volume will keep increasing resulting in higher utilization ratio over time which was seen in the chart above where the average weekly utilisation of both pools was over 150%+ every week.

• BNB Millionaire swap Volume

BNB Millionaire swap volume is a metric to measure the whale transaction volume in the pools. Users who have a cumulative BNB transaction volume of over 1 million BNB or roughly $256 million are categorized as BNB millionaires. Both pools have had a decent percentage of volume coming from Millionaires. USDT pool dominated in Millionaire volume in the entire 2021 but in 2022 both pools have similar volumes from the Millionaires with BUSD marginally ahead of USDT.

Despite the current bearish conditions, the Millionaire volume% in the BUSD pool is much higher than that of the early 2021 bull market.

Liquidity Mining reward allocation and Final thoughts

The goal of a Liquidity Mining program is to boost the targeted pool in terms of liquidity and in lot of cases to boost the usage of the targeted token.

In the case of an established Dex like pancakeswap liquidity mining programs can help it boost any pools in which it is seeing a decline and is losing market share.

Assuming that pancakeswap has $100,000 for the Liquidity mining program to allocate between the WBNB-USDT and WBNB-BUSD pools it makes sense to allocate the bulk of the rewards in the USDT pool as from our analysis so far it is clear that though the metrics of both pools have been similar but USDT pool has an edge in most of them and in recent months has outperformed BUSD pool in both trader and liquidity provider metrics. Moreover, BUSD has a higher TVL with a lower utlisation rate thus it would be better for the LPs if the USDT pool gets the liquidity boost instead

In my opinion, an allocation of 70% of $100,000 to USDT pool and 30% to BUSD pool would be an ideal split of the rewards. This would help USDT pool match the TVL of BUSD pool and there is a possibility that many LPs from BUSD pool migrate to USDT pool which would also improve the utilisation ratio for the BUSD pool and create a win-win situation for both pools.

The metrics have shown why these pools are among the top pools in DeFi . Both pools have shown strong resilience against the changing market conditions and have given consistent APY to the users barring the impermanent loss obviously which is a topic for another day.

Liquidity providers can expect to get a similar return from both pools and it’s just a matter of which centralized stablecoin they prefer.