Inventory of leading projects in the derivatives Dex track: which has more potential, dYdX or GMX?
The explosion of derivatives DEX is inseparable from the support of the market environment, and the growth of dYdX and GMX is a result of user choices facilitated by the current market conditions.Author of this article: Claire, Taihe Researcher
Table of Contents:
Huge Incremental Space in the Derivatives Dex Market
dYdX vs GMX---Factors Required for the Success of Derivatives Dex
Market Environment
Product Attributes
Value Capture
Who Will Ultimately Prevail, dYdX or GMX?
dYdX Surpasses GMX in Most Characteristics
dYdX V4 May Break the Deadlock
Projecting the Performance of dYdX and GMX Based on Market Environment
1. Huge Incremental Space in the Derivatives Dex Market
The derivatives Dex sector currently only accounts for 2% of Cex, and users lack trust in centralized exchanges due to the FTX collapse. Moreover, the market capitalization of the two most noteworthy projects, dYdX and GMX, has not stabilized within the top 100. If derivatives Dex continues to grow its market share in the future, there is significant potential for incremental space. This article will analyze which project in the derivatives Dex space has more potential based on the assumption of a huge incremental space.
2. dYdX vs GMX---Factors Required for the Success of Derivatives Dex
Currently, there are mainly two models for derivatives Dex:
- Order Book: represented by dYdX
The order book model pursues matching efficiency, requiring professional market makers to provide liquidity, and is more similar to the Cex experience, with funding rates and prices aligning with Cex most of the time.
- Liquidity Pool: represented by GMX and Gains Protocol
The advantage of oracle liquidity pools is zero slippage, providing a very smooth experience for large trades.
I will analyze the two projects, dYdX and GMX, from three aspects: market environment, product attributes, and value capture.
2.1 Market Environment
Looking back at history, the performance of the two projects under the same market environment has been completely opposite. Many on-chain projects have performed consistently with the broader environment. As shown in the figure below, dYdX's trading volume and open interest (OI) surged significantly starting from September 2021 when it issued tokens and launched trading mining, a trend that continued until the end of 2021, after which it fell sharply. dYdX's performance is consistent with the vast majority of projects in the market. Therefore, if the market rebounds, there is reason to believe that all of dYdX's metrics will rise accordingly.
In contrast, GMX has shown a completely opposite trend. Since GMX launched on Arbitrum in August 2021, its trading volume and OI have steadily increased, and since June 2022, OI has experienced explosive growth, with GLP maintaining a return of 20-50% even in a bear market.
Why did GMX perform poorly in a bull market but manage to grow against the trend in the bear market of 2022?
2022 was the year of L2 explosion combined with Ethereum's merge. 2022 was truly the year of L2's real-world application, featuring Optimism's token issuance, Ethereum's merge, Arbitrum Odyssey, Nitro upgrades, and other hot topics, along with the technological aspirations of zk rollups, bringing many users and traffic to GMX.
GMX's growth in the bear market is due to the dual growth of users and GLP. Specifically, gambling mechanisms perform better in bear markets because there are fewer hot topics, which stimulates users' gambling behavior, and projects in the form of gambling mechanisms attract more users. The counter-cyclical growth of GLP stems from the fact that GLP not only acts as a market maker to earn from user losses but also receives the fees distributed by GMX, thus obtaining sufficient returns even in a bear market.
2.2 Product Attributes
Good product attributes determine the traffic of an exchange. Below, I will analyze the two exchanges through liquidity, price discovery mechanism, and funding rates.
2.2.1 Liquidity
Liquidity is defined as the ease of buying or selling an asset at a stable price in a specific market. In other words, when trading activity is active in the market, traders can enter and exit at any time, regardless of the amount. dYdX introduces market makers to provide liquidity, pursuing matching efficiency, but cannot avoid slippage, resulting in trades not being executed at stable prices but rather at the average price of market orders, with larger trades experiencing greater slippage. GMX employs a zero-slippage mechanism, allowing for quick execution at oracle prices, enabling traders to buy and sell at stable prices and enter and exit at any time, even for large trades. Thus, GMX has better liquidity than dYdX.
If more users flow into the chain in the future, and derivatives Dex captures more market share from Cex, the trading volume and user base of derivatives Dex will significantly increase. Theoretically, dYdX's liquidity ceiling is higher than GMX's. This is because dYdX does not need to consider liquidity curves or how to maximize LP profitability, as there are two types of trading matching behaviors: market makers and users. As long as there are buyers and sellers in the market, liquidity exists. In contrast, GMX does not have user-to-user matching; it can only bet against the liquidity pool. The liquidity providers of GLP rely on passive income, and the opening limit is constrained by the utilization space of assets in the GLP pool.
2.2.2 Price Discovery Mechanism
The price discovery mechanism determines whether an exchange has pricing power. Order books have pricing power and can influence prices, which means that significant deviations in OI are less likely to occur. The order book model is a bet between users and market makers, requiring a 1:1 matching of long and short positions, and most positions can be offset. The portion that remains unoffset, causing position deviations, manifests as price fluctuations, consistent with centralized exchanges. Users' profits and losses are caused by market behaviors resulting from imbalances in long and short positions. Oracle pools lack pricing power and cannot influence prices, only passively receiving oracle price feeds, which means that the receiving party can only absorb price fluctuations. This may lead to oracle attack issues. For example, GMX was attacked in September 2022 due to its zero-slippage feature.
Oracle prices are sourced from centralized exchanges. From the perspective of using oracles, GMX will never be able to become the largest exchange. If GMX were to become the top exchange in the future while still sourcing prices externally, it would be easy for price manipulation to occur. Because external liquidity is worse than GMX's, it is easier to manipulate, meaning that even the prices of highly liquid assets like BTC and ETH on GMX could become distorted, not just the previously attacked AVAX price. This situation could be very dangerous, as GMX could be continuously attacked, leading to its prices failing to reflect the market's fair value. In contrast, dYdX, by not using oracles, can reflect the market's fair price. If it surpasses Cex in the future, it will certainly be through the dYdX model.
2.2.3 Funding Rates
The funding rate mechanism of dYdX is consistent with that of Cex. Unlike the order book, which can better balance longs and shorts, opening a long or short position on GMX can be understood as "borrowing" risk assets or stablecoins from the GLP pool to establish a position, with both longs and shorts needing to pay the same fees to borrow assets. Therefore, both sides of the trade on GMX are always paying funding fees and do not receive funding fees. On GMX, it is impossible to balance longs and shorts; when a strong one-sided market occurs, leading to significant deviations in OI, the GLP pool may incur massive profits or losses. For example, on June 18, 2022, the proportion of shorts in total OI shifted to 71%, and the GLP pool's losses peaked at $7.5 million.
In a bull market, most traders will go long, so the order book model balances longs and shorts through funding fees. However, based on GMX's fee mechanism, going short on GMX in a bull market not only does not yield funding fees but also incurs borrowing costs. Therefore, traders will certainly avoid going short on GMX. This leads to a state of extreme imbalance between longs and shorts on GMX during a bull market, as traders continue to profit in a one-sided market, while the trading fees collected by the platform cannot compensate for GLP losses, further exacerbating GLP's losses. Thus, in a bull market, traders will be more inclined to choose the dYdX model.
2.3 Value Capture
dYdX provides a certain value return to traders in the form of token distribution, but there is no value capture for token holders. All trading fees are owned by the dYdX protocol. V4 may be able to change this situation. In V4, the protocol will be fully operated by the community, and the community plans to create an operational sub-DAO. These changes may help dYdX tokens unlock value capture.
GMX tokens have strong value capture capabilities, with 30% of platform fees going to GMX stakers and the remaining 70% to GLP stakers. As shown in the figure above, the returns for GLP stakers can often reach the range of 20-50%. The sources of GLP returns mainly include two:
Fee income: This income is linked to the platform's trading volume, which depends on market volatility. The fees are converted into ETH and provided to stakers, which is what is referred to as real yield.
Trader losses: As seen in the data from the figure below, traders' cumulative PnL has been continuously declining over the year, translating into sustained returns for GLP stakers.
Therefore, GMX's significant advantage currently lies in its token's value capture. However, after 100% of platform fees are distributed to GMX stakers and GLP stakers, the GMX protocol will have no income, and whether this cash-burning model can truly be sustained, or if it will be like Uniswap reopening the fee switch, remains to be seen. In contrast, dYdX tokens currently have no value capture, and whether V4 can make adjustments needs to be continuously observed.
3. Who Will Ultimately Prevail, dYdX or GMX?
3.1 dYdX Surpasses GMX in Most Characteristics
From the comparison of the above characteristics, it can be seen that dYdX outperforms GMX in all aspects except liquidity, value capture, and decentralization. However, the adoption of oracles makes the best liquidity a double-edged sword, while value capture and decentralization are likely to be areas that dYdX will improve in Q2 of this year with V4. Although GMX has strong value capture, its gambling machine nature may lead to significant losses when market liquidity improves, and the use of oracles results in a very limited range of assets supporting trading. While GMX has some shortcomings, its composability is an advantage it can leverage, which dYdX is unlikely to possess. Currently, composability is mainly used to hedge GLP, but it is precisely where on-chain projects can showcase their innovative capabilities.
3.2 dYdX V4 May Break the Deadlock
dYdX V4 claims to become fully decentralized, with full decentralization mainly reflected in a decentralized order book and matching engine. The central party will no longer operate any part of the protocol but will be controlled by the community, and it will no longer earn income based on the protocol's trading fees. I understand this is why dYdX chose to migrate to Cosmos and launch V4 there, which will allow dYdX to become a unique public chain and have its own validator set, building the chain's operational model according to its own needs. Additionally, since dYdX's current circulation is only 14%, over 70% of tokens will be directly unlocked in 2023. Once V4 goes live as a public chain, the protocol will have its own validating nodes and shift to a PoS mechanism. If V4 proceeds smoothly, this selling pressure may be absorbed by staking.
3.3 Projecting the Performance of dYdX and GMX Based on Market Environment
The explosion of derivatives Dex cannot be separated from the support of the market environment. The growth of dYdX and GMX is driven by user choices facilitated by the current market environment. Projecting the market environment after 2023, based on Goldman Sachs' research analysis and Yellen's recent public statements, the U.S. economy is expected to continue moving towards a soft landing. The year 2022 can be seen as a period below potential growth, which can better balance the labor market and curb price pressures, and the negative impacts of U.S. fiscal and monetary policies on the economy will diminish in 2023.
As 2024 is the year of Bitcoin halving, under the premise that the macro market does not drag down, it generally marks the beginning of a bull market. Projecting the situation for derivatives Dex, GMX may maintain high returns in a bear market, but in a bull market, due to the shift in hot topics, the popularity of such protocols may decline. If the market environment improves after 2023 and incremental funds enter, then dYdX is more likely to become the preferred on-chain derivatives exchange for users.
