A boat in a container shipyard

The Best DEX LP Performance Benchmark: The Daily Rebalancing Portfolio

The best way to evaluate the profit potential of a DEX pool is to compare its returns against a common benchmark. We recommend a daily rebalancing portfolio.

Written by

Mark Lurie

Published on

November 28, 2022

The prevailing paradigm for evaluating liquidity provider (LP) returns in DeFi is broken. 

Competing DEXs are locked in a never-ending race to attract LPs, which has resulted in many DeFi projects using artificially inflated LP returns to lure in users. This is not only misleading to LPs, but also makes it harder for ordinary DeFi users to identify which projects are well-designed and genuinely effective, versus low quality dupes that hide behind misleading numbers or offer unsustainable short-term yields. The recent focus on “real yield” is an attempt to remedy this, but at best it is an incomplete solution, for multiple reasons

Fortunately, there is a better approach. But before we dive in, it’s necessary to debunk the three main misconceptions that liquidity providers fall victim to when attempting to evaluate DEX performance. 

1. The “Fee Yield” Fallacy

The claim: LPs should allocate capital by comparing APYs across different pools with similar assets, where APY is cumulative trading fees accruing to LPs (Fees) divided by total liquidity (TVL).

Why this is misleading: This approach ignores expenses incurred by LPs, namely Impermanent Loss (IL), which is an enormous oversight. Profit = Revenue – Costs, and LP’ing based on top-line fee yields is akin to running a business to maximize revenue instead of profits. In short, it’s a surefire way to lose money.

The “Fee Yield” Fallacy seems to persist because:

  1. Semantic confusion with the historical use of the term ‘yield’ in TradFi, which typically refers to profit, not revenue.
  2. It’s the only metric reported by most DeFi projects, perhaps because it’s the only metric that can never be negative, which makes it a convenient marketing tool for projects trying to attract LPs.

2. The “Impermanent Loss” Fallacy

The claim: Even though Impermanent Loss (IL) is unique to each person’s position, it is a temporary issue that involves similar losses across pools composed of similar assets.

Why this is misleading: There is no inherent property of IL that makes it temporary, because there is no inherent reason for asset prices to revert to their starting point once they’ve shifted. TradFi markets avoid this false assumption by requiring liquid assets be marked-to-market . Sure, any loss (or gain) over time would be “temporary” if prices magically snapped back to their original value. But that doesn’t make those losses impermanent – it simply makes them unrealized until a given position is exited.

Additionally, IL is not the same across any pools with the same assets, only against any CPMM (Uniswap V2 fork) with the same assets. The truth is different protocols undergo very different degrees of IL, and some projects (such as Clipper) don’t incur any “IL” at all, although LPs may lose money in other ways due to certain price movements. Understanding what IL actually consists of and recognizing the fact that this figure varies significantly across different protocols is necessary to make prudent LP decisions.

The “Impermanent Loss” Fallacy seems to persist because:

  1. Semantic confusion over the term ‘impermanent’, which misleadingly implies that the incurred loss is temporary.
  2. IL is not reported by most DeFi projects, perhaps because it’s always negative, an inconvenient attribute for projects trying to attract LPs.

3. The “HODL’ing” Fallacy

The claim: The best way to evaluate LP returns is compare the current value of the pool to the initially deposited assets valued at today’s prices (e.g., had the LP not deposited those assets at all). 

For example, if an LP started with 10 ETH worth of assets and deposited them into an ETH-USDC pool, and later withdrew assets worth 9 ETH, then they incurred a loss of 1 ETH relative to simply HOD’Ling.

Why this is misleading: The returns from a given pool over a given period is partly a function of the macro markets, and partly a function of that specific pool. We generally refer to the impact of the macroenvironment as beta, and the outperformance of a specific DEX as alpha.

Beta often has the biggest impact on returns vs. HODLing, so LPs who judge pools against HODLing will allocate assets based on beta (which could also be hedged). But LPs will be exposed to beta unless they don’t LP at all. That means they won’t be allocating to the pools with the best alpha, resulting in systematically poor allocation decisions.

The HODL’ing Fallacy seems to persist because:

  1. It’s simple to measure and there is no widely accepted alternative paradigm. 
  2. There is a widespread bias toward doing nothing among Bitcoin millionaires who got rich from simply holding BTC over the last ten years.
  3. There is a false impression among DeFi users and fund managers that their current returns are actually benchmarked to HODL’ing.

Benchmarking LP Returns Against a Daily Rebalancing Portfolio

Instead, the best way to evaluate the profit potential of a DEX pool is to compare its returns against a common benchmark. We recommend a rebalancing weighted portfolio, specifically a daily rebalancing portfolio (DRP).

You may already be familiar with this concept as a simple diversified portfolio, where each asset has a target weight and assets are bought and sold once per period to maintain that target. For example, many people’s savings are in a portfolio that is 60% equity and 40% bonds and is rebalanced quarterly. A rebalancing portfolio is superior to HODLing not because it will outperform 100% of the time, but because it is a better risk-return tradeoff (better sharpe ratio).

Mostly, though, it will outperform HODLing. This is because it actually makes money from volatility that happens in between rebalancings. For example, assume that a given DRP is designed to hold 50% ETH, 50% USDC. If the price of ETH rises 10% on day 1, the DRP will sell some ETH for USDC to maintain its 50%-50% target. If the price of ETH then falls 5% the next day, the DRP will sell enough USDC for ETH to regain its 50%-50% target. Notice that in this case, price levels have changed but the DRP made money (the CPMM would have lost money).

The rebalancing portfolio is an especially good benchmark because it’s probably already your default. If you own some ETH, some BTC, and some USDC you must have some weights between them. Perhaps you don’t rebalance optimally at predefined intervals, but you may do so when allocations feel extremely unbalanced. For example, many long-term ETH bulls will convert to USDC if they feel ETH is especially high or frothy - in other words, when their portfolio becomes overweight ETH. Then, they will convert back to ETH when they feel ETH is at a local minimum - in other words, when the portfolio is underweight ETH. This is effectively a rebalancing weighted ETH-USD portfolio, with discretion defining the intervals rather than using the empirically optimal intervals. By targeting weights, you end up systematically buying low and selling high, making more money for less risk than HODLing.

While many portfolios are rebalancing quarterly, there is nothing theoretically preventing daily rebalancing (except practical transaction costs). Hence, we use the Daily Rebalancing Portfolio as the ideal benchmark.

Regardless of the benchmark methodology, the main benefit is in comparing pool returns to a common benchmark of any sort. That abstracts away the beta and isolates the alpha, allowing LPs to make better allocation decisions. First, you decide what portfolio you want to benchmark against (e.g. 50% ETH, 50% USDC). Then, you evaluate all pools against that same benchmark.

Clipper’s core pool tracks the theoretical DRP. It can either make or lose money depending on what happens in the market, but we judge its returns relative to the benchmark, because that’s what is unique to Clipper and in its control.

The BEST Way to Evaluate DEX Performance

Ultimately, assessing the profitability of a DEX depends on the specific time-series of prices that occurs in the market. This means benchmarking DEX returns against a DRP will more effectively capture the whole picture when attempting to compare the performance of different DEXs under ever-shifting market conditions.

Only looking at top-line fee yield is akin to ignoring impermanent loss, and liquidity providers who fall into this trap risk making misguided, unprofitable LP decisions. And while it is somewhat useful to compare LP returns against the hypothetical value of your assets if you simply HODL’d, this comparison does not sufficiently consider alpha in your decisions.

At the end of the day, the only way to determine which DeFi projects warrant attention, and which ones should be left to die, is to find a better way to benchmark different LP returns across different DEXs. The only way to do so is for liquidity providers to focus on bottom-line profit yields under dynamic market conditions, and the best way to accomplish this is to benchmark your DEX LP returns against a daily rebalancing portfolio.

Written by

Mark Lurie

Published on

November 28, 2022

March 29, 2024

What are stablecoins used for?, Clipper expands to Base, the latest AdmiralDAO proposals, and more.

February 28, 2024

A positive omen for DeFi derivatives, Clipper's new facelift, the latest AdmiralDAO proposals, and more.

Keep up with DeFi by subscribing to Shipyard’s industry newsletter

We’ll send you the latest news about DEXs, stats on our products, interesting articles, and CEO Mark Lurie’s thoughts on it all.

Thank you! Your submission has been received!
Oops! Something went wrong. :-( Please try again.