By：CapitalismLab
Recently, Trade joe V2 became popular because it grabbed a large amount of Arb trading shares from Uniswap, and the currency price quickly doubled. So how exactly is it done? What should we pay attention to when providing liquidity?
This article will give you an in-depth understanding of the mechanism of Joe V2, analyze why it can obtain a large share of $ARB transactions, as well as the advantages and disadvantages of the product, and help you better understand DEX.
Trade Joe V2 in a nutshell:
AMM is similar to an order book, using discontinuous liquidity
Minimum price precision is based on a ratio rather than a fixed value
Vertically aggregated liquidity brings better composability
Liquidity incentives based on transaction fees and effective TVL obtained by LP
AMM Mechanism
Next, we will take the ARB/ETH 20bps pool in Joe V2 as an example for an in-depth analysis. First, let’s look at the liquidity distribution of the pool. At first glance, it is quite similar to UNI V3.
So what is the fundamental difference? A long bar (bin) of Joe V2 corresponds to a single point price, which means that unless you consume all the liquidity of this bin, the price will not change. For example, in the figure below, you can see that there are about 35k Arb and 2 ETH in the bin. If someone sells 2E worth of Arb at 0.00066, the transaction bin will move one position to the left, and the price will change.
We try to add liquidity to the pool again and find that we can add single-point liquidity, which means that the added liquidity is at one price, which is actually equivalent to placing a Maker order in the order book. What about the Pct Rage = 0.20% next to it?
We move the right slider slightly to expand the price range to 2Bin, and then calculate the relative difference between the two prices.
(0.00067237272-0.00067103065)/0.00067103065 = 0.20%
That is to say, 0.20% is the minimum price accuracy.
Let's look at what happens when the price deviates significantly from the current price, such as ARB/ETH = 0.0001.
（0.00010035504-0.00010015473）/0.00010015473 = 0.20%
Yes, it can be understood that the difference between the upper and lower prices at any position is 0.2%, which is based on a ratio rather than a fixed value. This is quite different from the traditional order book, which generally gives a fixed minimum precision, such as 0.01 USDT.
Joe's UI also provides 4 ways to add liquidity. Click "learn more" in the upper right corner of the above picture to learn the detailed definition. However, except for Spot, the other four use the parameter values set by the official and cannot be controlled. So I suggest using Spot, and the storage method is similar to UNI V3. At most, you can store a few more with different price ranges. In addition, Joe V2 currently has liquidity incentives, and the benefits of this must be considered.
Liquidity Incentives
Currently, the pools with the "Rewards" tag after their names have liquidity mining incentives.
Joe V2’s liquidity mining incentive distribution model:
The score is calculated based on the transaction fees actually obtained by LP and MakerTVL, see the figure below
MakerTVL currently only looks at the current price within +- 5 bins, such as the Bin width of Arb/ETH 0.2%, so only TVL within +-1% of the current price will be recorded.
When an Epoch ends, the scores within the Epoch are counted and distributed in proportion.
That is to say, if you want to obtain liquidity incentives, you still need to provide liquidity in a more concentrated manner.
Why $ARB transactions get a large share
Assume that the current market is continuously buying Arb, Uniswap Arb/USDC = 1.005, and assume that the price accuracy of Joe V2 is 1%, and the bin distribution is [0.99, 1.00, 1.01…]. At this time, Joe's current bin should be 1.00, which is 0.5% cheaper than Uni. As long as the transaction fee is smaller than this difference, transactions buying Arb through aggregators such as 1inch will naturally go to Joe first. On the contrary, if it is selling Arb, Joe has no advantage. In other words, it has an advantage in a high-volatility one-sided market, but is mediocre in a low-volatility monkey market.
In addition, Joe set a 0.2% fee for the ARB/ETH trading pair. At the time, due to high volatility expectations, UNI could only set four levels: 0.01%/0.05%/0.3%/1%. Most LPs were above 0.3%, which was disadvantaged compared to Joe's 0.2%.
The regular exchange rate advantage under high volatility + relatively low fees enabled Joe V2 to gain a large share in the early stage of ARB 0.00 launch. Currently, with the decline in volatility, its exchange rate no longer has a regular advantage; UNI V3 LP has basically lost its fee advantage after returning to the pool with a fee of 0.05%, but fortunately, its reputation has been established, and it has a good incentive mechanism, which gives the project party more room for operation.
Product advantages and disadvantages
In fact, in the above discussion of Arb transaction shares, the advantages and disadvantages of its AMM mechanism have been explained. This section discusses other aspects:
Advantage:
Vertically aggregated liquidity brings better composability
High efficiency + support incentives, you can get incentives from partners, such as the possibility of benefiting from LSD war
Disadvantages:
There is no mature token empowerment mechanism such as Bribe, and the benefits to token holders are limited
Impermanent loss is relatively large
Advantage: Composability
Since in Joe V2, liquidity is aggregated vertically through each bin, while in Uniswap V3, liquidity is aggregated horizontally. The main benefit of vertical aggregation is that it allows liquidity to be fungible.
If we look at a specific transaction for adding liquidity, we will find that after the user adds ETH/USDC, Joe returns a large number of ERC-1155 LBT with different Token IDs to the user, reflecting the user's liquidity distribution at different price levels.
This is quite different from UNI V3 which returns a single NFT. Because the same Token ID represents liquidity within a single Bin/price level, which is homogenized and has better composability.
Advantages: Partner incentive expectations
The high efficiency of concentrated liquidity + support for incentives is expected to attract partners with liquidity needs such as LSD to provide incentives. For example, Kyberswap, the previous UNI V3+ incentive model, received an incentive allocation from Lido second only to Curve (see the tweet below), and occupied a large share of Alt-l1/L2's LSD trading volume.
Then Joe V2, which can provide similar value, theoretically also has this opportunity, thereby increasing TVL and transaction volume.
Disadvantage: Token empowerment issues
As I mentioned in the tweet below, a pure spot DEX would be difficult without additional enabling mechanisms such as bribery, as TVL and trading volume alone would not be able to be converted into returns for token holders.
Currently, Joe earns revenue by taking a cut of the transaction fee, but if he takes too much, it will inevitably affect his market share. Although Joe has Ve Joe, it is not a copy of Curve and is not very successful, so token empowerment is still a problem.
Disadvantage: Impermanent loss is relatively large
As mentioned above, Joe V2 has a regular exchange rate advantage under one-sided market conditions, but this actually means that LPs incur greater impermanent losses, which is equivalent to selling the coins at a cheaper price.
Joe will set the fees reasonably to make up for this, for example, the fee for ARB/USDC with a 1% bin width is 0.8%, and the fee for ARB/ETH with a 0.2% bin width is 0.2%. Plus this point is not easy to notice, so it’s okay (laughs)
Summarize
Compared with UNI V3, Trade Joe V2 has more of a differentiated advantage in efficiency in specific scenarios, and the high composability advantage is very dependent on its own scale. Therefore, we should continue to pay attention to the project's BD and other operations in the future to see if we can get enough cooperation incentives through BD and build a growth flywheel.
