Increasing Correlation Challenges Asset-Level Diversification’s Risk-Management Power

Traditional long-only portfolio management theories, pioneered by Nobel laureate Harry Markowitz in his Modern Portfolio Theory (MPT), postulated diversification as the main method to manage risk. Risk is commonly defined as volatility or standard deviation of an asset’s historical return. MPT involves finding the optimal point alongside the efficient frontier, i.e., given an expected return, finding the portfolio asset weighting matrix that bears the lowest risk. Although holding anything that is not perfectly correlated with each other would help reduce portfolio risk, increasingly positive correlations between previously negatively-correlated assets have caused problems to MPT practitioners on the asset level as the diversification effect waned. For example, the stock-bond correlation has shifted into a less negative regime since COVID-19, as both assets have been lock-stepped with global central banks/interest rate policies.

The above chart shows the 90D rolling correlation between the S&P 500 and the Vanguard Total Bond Market Index Fund (BND).

As for crypto, the market capitalization of digital assets has become highly correlated with equities and almost completely lost its own cycles. Even within crypto, diversifying into different themes (L1s vs. NFTs), chains (EVM vs. non-EVM), and factors (value vs. growth) only provided a limited reduction in portfolio volatility as inter-crypto correlations are high as shown below.

Importance of Strategy-Level Diversification

Asset-level diversification is under an inherent premise of investors being asset owners – a classic “long-only” mindset. However, owning assets is just one strategy of investing; adopting more strategies than simply owning assets would be what we call strategy-level diversification.

Drawing an analogy to TradFi investing, strategy-level investing is what the largest sovereign wealth funds (SWF), pension funds, and insurance funds have been doing for decades. A typical SWF would likely deploy capital into both asset-owning strategies (primary/secondary market securities, real estate, etc.) and alternative strategies, which can be outsourced to specialized hedge funds. For example, some hedge funds specialize in running “black-swan strategies”, which is essentially a rolling bet on tail events. Such strategy-level diversification significantly reduces the expected volatility of the overall portfolio, as non-asset-owning strategies naturally have lower correlations with asset-owning ones given they are designed to earn returns from different risk premia.

Diversify and Hedge Portfolios With Delta Neutral Structures

Among all the strategies available for diversification, delta neutral ones are perhaps the most hyped, debated, and questioned in TradFi, CeFi, and DeFi. In a delta neutral strategy, the investment return is supposedly independent of the asset’s price return, compared to an asset-owning strategy without hedges being “delta-one”.

Delta neutral strategies are no stranger to the world of TradFi. For example, arbitraging FX pairs or shares of the same company listed on different exchanges is a classic – arbitrageurs are usually net flat on the underlying asset they are “arbing”, but are exposed to other hurdles such as different settling times, stamp duties, and other costs, mispricing due to technical glitches, business day count differentials, counterparty risks, and so on. 

Some strategies assumed to be delta neutral have failed before, as Long Term Capital Management demonstrated to the world when its “statistical arbitrage” trades got hit by mathematically-impossible tail events and eventually collapsed.

Drama aside, this paper aims to cover the fundamentals of delta and provide a comprehensive overview of the various delta neutral strategies that traders and investors can employ in the crypto markets.

What Is Delta

Readers who are already familiar with delta can skip the next two sections.

Delta is simply how much a position’s value would change if the underlying asset moved $1. Mathematically, delta equals dP/dS, where P is the investor’s position value and S is the spot price of the underlying asset. The concept of delta originates from derivatives, specifically options, but it can be applied to almost all financial instruments.

Delta can be interpreted as the directional exposure on a chosen underlying asset’s price return. Directional exposure is simply the risk of loss resulting from exposure to the asset’s direction. As mentioned briefly above, long-only asset-owning strategies without hedges are delta-one, while a short position in linear instruments, e.g., futures, has a delta of -1. 

Here is a closer look at delta with an example:

  1. Alice’s portfolio holds 1 Ethereum (ETH) which she bought at $2,000
  2. Alice’s portfolio has an ETH delta of 1
  3. This means that if ETH falls by $1, Alice will incur an unrealized loss of $1
  4. Vice versa, if ETH increases by $1, Alice will gain an unrealized profit of $1

A portfolio manager with multiple assets can compute a single figure to represent how much the portfolio will move if the benchmark, such as S&P 500, moves by $1. In this case, the portfolio manager can calculate the beta-weighted delta by computing the individual beta of each stock to the benchmark, and solving for the weighted sum of the portfolio delta to the benchmark.

Beta Weighted Delta = β × Δ × (Stock Price/Index Price)

In the above example, a simple sum of all the delta (10+5+5) is not as meaningful as 1 AMZN delta is not the same as 1 MSFT delta from a portfolio standpoint. By adjusting the deltas to the SPY (SPDR S&P 500 ETF Trust), we can see that for every $1 increase in the SPY, our portfolio is expected to gain $66, and vice versa. Investors can use this information to hedge S&P delta by selling an equivalent amount of SPY to become SPY delta neutral. 

If one chooses to run this SPY delta neutral portfolio, then he is inherently betting that the portfolio stocks will outperform SPY (even if AMZN/TSLA/MSFT falls less than SPY, the portfolio still profits), instead of betting on an absolute price increase in an unhedged portfolio. It is important to note that beta-weighted deltas are constantly changing.

What Is Delta Neutral?

Delta neutral essentially means that one’s portfolio has zero deltas to a chosen underlying asset or benchmark of one’s choice (dP/dS = 0). Note that delta neutral is always a relative concept depending on the S chosen in dP/dS. For example, a portfolio with zero ETH delta doesn’t necessarily mean it has zero BTC delta too. 

Here is a simple example:

  1. Imagine a portfolio that contains 1 spot ETH and 1 short ETH perps position:
    • ETH delta for 1 spot ETH is 1, and -1 for 1 short ETH perp
  2. Ignoring funding costs and price deviations, the overall ETH delta for the portfolio is 0

One question arises: if a portfolio is delta neutral, how does it make money? This is where our recent risk premia article comes in handy, as it demonstrates in more detail how a delta neutral portfolio generates return from risk premia other than market risk.

R = rf + Risk Premia1 + ... + Risk Premian

Popular Delta Neutral Strategies

This section covers some common delta neutral strategies that investors can implement.

Spot Arbitrage

1. CEX Spot Arbitrage

As previously stated, cross-exchange arbitrage is one of the most basic ways to generate delta neutral profits because it involves simultaneously buying and selling on different exchanges.

CEX spot arbitrage entails identifying price differences on a specific cryptocurrency between CEXs such as Binance and Bybit and trading the difference until it disappears or becomes so small that the cost of trading exceeds the potential profits. Because some CEXs are larger than others and have higher trading volumes, arbitrage opportunities exist. As a result, demand and supply on one CEX may significantly differ from another, thus influencing price.

In practice, arbitrageurs may encounter friction that can erode arbitrage profits, particularly when attempting to transfer tokens purchased on one exchange to another for sale, i.e., execution risk. One way to mitigate this is to be long on one exchange while simultaneously opening a short position on the other, then, waiting for the two exchanges to converge. However, because two positions are opened, the arbitrageur must maintain double the collateral/capital.

Spot arbitrage opportunities have become increasingly difficult to find in recent years as more sophisticated participants with faster technologies enter the market. Investors might still find arbitrage opportunities by looking at less popular cryptocurrencies, which typically have a wider price spread due to their low volume and depth of order book. On top of that, arbitrage opportunities typically arise during high market volatility as prices and volume across exchanges may become temporarily out of balance (i.e., during Terra’s implosion in May).

2. MEV

Maximal extractable value (MEV) refers to a strategy that involves profiting from pending transactions in a mempool by including, excluding or reordering transactions in a block. The three most common examples of MEV are: DEX arbitrage, liquidations, and sandwich trading. We will cover more on DEX arbitrage in a separate section.

Liquidations are prime MEV strategies. Whenever a borrower is liquidated, there is a liquidation fee associated with it. Searchers can determine which borrowers can be liquidated and compete to be the first to execute a liquidation transaction to collect the fee for themselves.

Sandwich trading, on the other hand, is similar to high-frequency trading (HFT) in TradFi, which involves front-running transactions that are waiting for block confirmation. For example, when you submit a buy order of ETH to a DEX, the MEV bot places the same buy order before your transaction (by paying a higher gas fee to be prioritized first). Then, the MEV bot submits a sell order right after your buy order is executed, essentially creating a “sandwich” trade.

MEV could be argued as a form of negative externality as it brings a whole suite of problems such as network and chain congestion. There are several up-and-coming protocols that are designed to mitigate these existential risks by either reducing the total MEV available at the consensus layer or by offering a transparent and fair set of rules to order transactions within each block. Some of these mentions include Rook, Eden, and Flashbots.

3. DEX Spot Arbitrage

DEX spot arbitrage is similar to CEX spot arbitrage where one captures the price difference between DEXs. The same steps discussed in CEX spot arbitrage can also be applied to DEXs. However, unlike CEXs, DEX arbitrageurs can operate without execution risk by creating a smart contract or an MEV bot that atomically executes both legs of the trade. Atomicity ensures that all actions within a transaction are valid and possible for the transaction to be accepted. This significantly reduces execution risk as there is no time lag between buying, transferring, and selling.

Furthermore, arbitrageurs do not need to worry about laying fiber-optic cables across the country which is what sophisticated arbitrage funds do to reduce latency. This is due to the fact that all transactions are ultimately subject to the blockchain’s block time, which puts all traders on the same starting line.

However, DEX arbitrageurs are still bound by the rules of the miners and validators as these blocks can be re-arranged and re-ordered in the priority of being processed. In addition, DEX arbitrageurs are also vulnerable to front-running because the transactions are visible to the parties who validate the blocks. When the transactions are submitted to the blockchain, miners and validators can use the information to execute the trade before the arbitrageur.

Funding Arbitrage

1. Spot vs. Futures Spread (Cash and Carry Basis)

The Cash and Carry trade is a proven strategy in both TradFi and crypto markets. The trade takes advantage of the term structure of futures and involves a long/short pair between spot and futures contracts. As the futures approach expiry, their prices converge to spot price, and the spot-future price difference can be pocketed as delta neutral profit. 

Futures contracts obligate involved parties to buy or sell the underlying asset at a predetermined price (forward price) for delivery at a specified date in the future. For example, there are multiple BTC futures available with different expiry dates. Prices between spot and futures, and among differently-dated futures, may not be, and are usually not, equal at a given time. Such pricing differences arise from costs and benefits between holding futures and spot assets.

The general formula for futures pricing can be derived from the non-arbitrage principle:

F = S(1 + r + c - d - v)t

Where, 

F = Futures price

S = Spot Price

r = Opportunity cost (risk-free interest rate)

c = Carrying costs (e.g., storage costs)

d = Dividend yield (for equities)

v = Convenience yield

t = Time

The futures term structure is a plot of prices of different futures contracts against their corresponding expiration dates. An upward-sloping curve is called contango, and the opposite is called backwardation. Term structure is a reflection of the pricing factors in the above formula plus supply and demand dynamics. 

Both crypto and equities futures typically exhibit a contango term structure. On futures expiry, the futures price must equal the spot price by design. Therefore, a long/short between spot and futures contracts held until futures expiry will generate a delta neutral profit of the differential between the forward price and the spot price at the time of entry.

Although the premium, which is the positive price differential between futures and spot, fluctuates quite a bit before expiry, there is a clear trend of premium converging to zero as expiry approaches. If an investor entered in long BTC spot / short BTC Sept 30 futures around the end of March, he would have made ~3% before fees if he held the structure until September 30.   

Pros:

  • Easy to implement with only spots and linear derivatives involved
  • Leverage is available on the futures leg, depending on exchanges
  • P&L is path-dependent, i.e., traders do not necessarily have to wait for futures to expire to realize a profit if spot-futures differentials converged in favor of the trade before expiry

Cons:

  • There might be settlement costs at futures expiry depending on exchanges
  • Futures leg can be liquidated if high leverage is used
  • Futures term structure can move against the trader and result in unrealized mark-to-market loss before expiry
  • Contango term structure is friendlier to Cash and Carry trade as long spot / short futures is operationally simple. Backwardation term structure requires short spot / long futures, which could incur extra borrowing costs on the spot leg and might be hard to implement due to lack of borrowing on certain assets

2. Spot vs. Perps Funding Basis

Another variation of the Cash & Carry trade is implemented through perpetuals. Perpetuals are basically futures without an expiry date. Unlike futures that inherently reflect in-kind funding costs via term structures, perpetuals impose an actual funding cash flow because they do not have an expiry date that forces spot and futures prices to converge. Funding cash flow exchanges hand between long and short open interests (OI) either hourly or 8-hourly depending on the exchange.

Perpetuals’ funding rate is mainly determined by the differential between traded price and “mark” price, although other factors such as financing rate for leverage play a factor too. A positive funding rate, where long OI pays short OI, means that the perpetual is trading above its “mark” price determined by the exchange, and usually implies that long OI overpowered short OI in the previous funding cycle.

With that in mind, traders can effectively capture the funding rate in a delta neutral way.

Similar to the spot and futures trade:

  1. The funding rate on BTC-PERP is 0.01% every 8 hours:
    • Longs have to pay the shorts to keep their position open
  2. A trader can short BTC-PERP and long the corresponding amount of spot BTC-USD 
  3. The trader can keep the position open as long as the funding rate remains above the desired threshold for the profit to be meaningful

The chart above shows the theoretical value of a portfolio with a $10K starting principal that runs the carry strategy for three years non-stop. The trade involves shorting perp and longing spot when the funding rate is more than 0%, and the opposite is true for the reverse. Across a three-year time period, the strategy would have yielded 29.54% annualized returns. Note that this strategy would return much less in the real world, because: 

  1. Frequent turnover (flipping from long spot/short perp to short spot/long perp in such strategy would incur significant transaction costs, especially if all the trades are executed as liquidity takers; 
  2. The majority of the period between 2020-2022 was in BTC bull market and likely associated with persistent positive funding costs, which will not be the case in a bear or sideway market;
  3. The strategy did NOT take into account short spot borrowing costs when perp funding rate is negative. Traders need to find the optimal balance in this strategy between trading costs and funding calls.

DeFi vaults have been built to automate spot-perp funding arbitrage. Friktion features a vault that helps users automatically capture the basis yield from funding premiums on SOL tokens.

Pros:

  • Same ones as spot-futures Cash and Carry trade
  • No need to roll futures as perpetuals never expire

Cons:

  • Same ones as spot-futures Cash and Carry trade
  • Transaction costs can be high in a market where funding rates keep flipping between positive and negative (can be partially offset by setting a threshold before flipping positions to reduce turnover)

Delta Neutral Volatility Strategies

Since options utilize greeks other than delta, they have become natural instruments to express views on non-delta risk premia. The common greeks of options include:

Delta: dP/dS, where P = options price and S = spot price

Gamma: d(Delta)/dS

Vega: dP/d(IV), where IV = implied volatility

Theta: dP/dT, where T = time

Rho: dP/dR, where R = interest/discount rate

Volatility strategies are not necessarily delta neutral if one chooses to keep residual delta. In our discussion below, we focus on dynamically delta-hedged structures which are pure volatility bets. Betting on volatility usually means having exposures to gamma, vega, and theta.  

One should note the difference between realized volatility (RV) and implied volatility (IV). RV are backward-looking statistics defined simply as the standard deviation of the spot price of a given historical time period. As the name suggests, its value cannot be known until the time period has passed. IV is the market’s pricing of expected forward-looking volatility, which is backed out from option prices via option pricing models, typically Back-Scholes. 

Empirically, gamma is a bet on RV and vega is a bet on IV. The reason for the latter is clear as vega is, by design, the derivative of option price on IV change. However, the reason for the former is less intuitive, and it will be shown below.

1. Short Volatility

Short vol is a popular income-generating strategy. The income earned can be attributed to a few risk premia. 

Firstly, one can generate yield from variance risk premium (VRP), which refers to the differential between IV and RV. Historically, IV had been higher than RV because traditional long-only investors are often price-insensitive when buying options to hedge their portfolio. The clean implementation of the VRP harvesting strategy is via delta-hedged calendar spread, which produces long gamma (RV) / short vega (IV) exposures. 

Secondly, outright short vol strategies earn income from time decay or positive theta. As an option approaches expiry, the underlying has less time, thus, less chance to move in-the-money (which produces intrinsic value), causing options to lose price as time goes by, ceteris paribus. Since short vol means selling options, the strategy benefits from a reduction of time value embedded.   

Delta neutral short-vol option structures can be implemented in multiple forms. The most straightforward is an at-the-money (ATM) straddle, which involves selling both ATM calls and puts at the same time. 

ATM options tend to have a delta of ~0.5 as it has an equal probability of ending up either ITM or out-of-the-money (OTM) at the expiry, according to normal distribution. However, in the real world, asset returns do not behave according to normal distribution. For example, in the early days of crypto, returns are highly asymmetrical to the upside which skews the real-world probability. Thus, an ATM call may not have a delta of 0.5 (or -0.5 for puts).

It is important to note the delta on the option leg when entering into a short straddle. Most order book style option exchanges such as Deribit will list the delta for each option strike. The figure below shows that the $1,600 strike price happens to be the strike that has a delta of 0.50 for both calls and puts, despite the spot price being nowhere near this strike.

The second structure to consider is the short strangle, which involves both selling OTM calls and puts. It follows a similar logic to the first structure (straddle), but instead of having both legs struck at the same price, the strikes will be further away from the spot price.

In this case, the trader can enter into a delta-30 OTM strangle by selling the $1,950 call strike, and the $1,300 put strike.

Although the above structures are delta neutral at inception, delta will drift away from zero, given the non-zero gamma exposure once spot moves from the price at inception. Hence, this strategy requires a dynamic delta hedging strategy to maintain delta neutrality. However, for a short vol structure, dynamic hedging is a money-losing feat: because of the negative gamma associated with any short options positions, delta will increase as spot goes down and vice versa, forcing hedgers to buy high and sell low. For this reason, short vol investors usually avoid dynamic hedging and instead, choose to buy further OTM calls and put to “knock-out” or cap their losses should spot prices move too much. These structures are called butterflies and iron condors for their payoff profile’s shape.

2. Long Volatility

Contrary to short vol which is a natural income-generating strategy due to VRP and theta decay, long vol in itself is a negative carry structure. However, the positive gamma exposure associated with long options positions gives a benefit to dynamic hedgers: since delta increases as spot increases and vice versa, dynamic hedgers naturally buy low and sell high on the underlying asset in order to maintain delta neutrality. As long as the realized dynamic hedging P&L is more than theta decay paid by the long-vol position, the overall structure is profitable. 

The possibility of profiting from the underlying asset’s movement, i.e., volatility, is the reason why gamma is deemed as a bet for RV – long gamma is essentially a bet that the RV (realized through dynamic hedging) is more than theta decay.

To implement a delta neutral long-vol strategy, Instead of selling both call and put options, we now have to buy both calls and put options, which leads to long straddle or strangle positions. As long as the deltas on each leg cancel out each other, we are considered delta neutral on inception.

The aforementioned volatility strategies can be easily executed on options exchanges such as Deribit and Bybit for centralized exchanges, or Zeta Markets and Lyra for decentralized exchanges.

Executing multi-leg option structures can be cumbersome and expensive. Some tools, such as Deribit’s strategy function, allow traders to buy or sell combo structures (i.e., call spreads, ratios, strangles, straddles, etc.) in a single transaction which saves fees and mitigates the risk of one leg being left unfilled.

If executing options outright is too complex for some, there are several DeFi protocols that automate long and short volatility strategies for the user. On the long volatility side, Dopex features an “Atlantic Straddle Vault” that allows users to buy straddles on the underlying. Similarly, Polynomial protocol has a “Gamma Vault”, that allows users to enter into a delta neutral strategy that combines both options and futures powered by Lyra and Synthetix Futures.

For short-vol vaults, Opyn features a “crab” strategy vault that holds short oSQTH against ETH collateral. oSQTH is a power perpetual swap that tracks the index of ETH2; it has an ETH delta of d(eth2)/d(eth) = 2eth and a constant gamma of d2(eth2)/d(eth2) = 2. To obtain the gamma exposure, long oSQTH open interests must pay in-kind funding, which is similar to option buyers paying theta to sellers. Therefore, by shorting oSQTH and hedging deltas with spot ETH, the crab strategy is essentially a short volatility strategy with dynamic hedge, collecting funding (theta) but losing money in the buy-high-sell-low hedging. As long as the funding profit is larger than the hedging loss, the vault is profitable.

Delta Neutral Liquidity Provision (LP)

Providing liquidity into a classic constant product AMM is akin to shorting volatility as impermanent loss leaves liquidity providers with more assets that are being sold more. For example, in an ETHUSDT constant product AMM pool, when ETH spot moves lower, the liquidity provider ends up with more ETH delta, and vice versa – this is similar to a short gamma exposure.

While option sellers are entitled to theta decay income, the liquidity provider’s compensation for shorting vol is not automatic but depends on trading volume – if a pool doesn’t attract volume, the LP position will not make money even if vol drops to zero, i.e., price doesn’t move at all. 

Similar to delta neutral short-vol strategies, one can structure an LP position in a delta neutral way as well. 

1. Pegged Assets LP 

Leaving collateralization methods and de-peg risks aside, pegged assets’ prices are supposed to move in tandem, i.e., P(pegged asset) = P(native asset), which means native asset delta of pegged asset = dP(pegged asset)/dP(native asset) = 0. Hence, LP between two pegged assets on the same native assets, or pegged and native assets are, by definition, delta neutral.

Two main categories of pegged assets exist, namely stablecoin and liquid-staked tokens. Overcollateralized stablecoins issued by trustworthy backers can be deemed as fully backed by fiat, and the LP between them has zero delta as well. 

More nuances surface when it comes to liquid-staked tokens: just being fully collateralized (which all liquid-staked tokens so far are) is not enough to be delta neutral to native tokens. The liquid-staked tokens also need to be redeemable for native tokens after a known period of time. For example, sAVAX can be slow-burnt on BenQi and becomes redeemable for AVAX in 15 days, which makes sAVAX similar to a 15d future contract of native AVAX. On the other hand, stETH is not redeemable for native ETH until an uncertain point in the future, hence cannot be regarded as pegged and trades at a spread from ETH depending on the market’s pricing of that indefinite period before it becomes freely redeemable. 

Stablecoin liquidity providers earn yield income from trading fees and would be taking on the risk of de-peg on one of the stablecoins, smart contract risks, and protocol risks associated with the LP platform. A temporary imbalance of flow among stablecoins could also cause unrealized mark-to-market loss of a stable LP position.

Liquid-staked/native tokens liquidity providers earn yield from providing a secondary market to liquid stakers who don’t want to wait through the redemption period and slow burn. The liquidity provider takes on the same set of risks borne by stablecoin liquidity providers, in addition to protocol risk of the liquid-staking dApp. Note that liquid-staked/native tokens LP positions are only delta neutral in the token’s perspective (e.g., stkBNB has no BNB delta, but still has USD delta). For those who start from USD and buy a token before liquid-staking half of it and forming an LP, they can hedge token-USD delta in perps to achieve USD delta neutral.

2. Pseudo Delta Neutral Leveraged Yield Farming

In a levered farm, liquidity providers rely on the borrowed assets to achieve USD delta neutral when at the inception of the trade. Note that impermanent loss will tip the delta neutrality as the spot moves away from price at inception. 

Assuming a starting capital of $1K USD to invest in a leveraged ETHUSDT leveraged farm, there are generally two ways to arrive at a delta neutral farming position at inception depending, on the amount of leverage involved. 

The first structure involves:

  1. Swapping USD to $1K worth of ETH
  2. In a leveraged farming protocol, select 2x leverage and borrow 100% in ETH, then supply $1K worth of ETH 
  3. Based on settings selected in 2), the protocol will automatically borrow $1K worth of ETH and sell that into $1K USDT and enter an LP position of [$1K worth of ETH – $1K USDT]

If there is no IL and zero borrowing costs, the position’s asset ($1K worth of ETH) exactly matches the liability ($1K worth of ETH). If ETHUSDT sells off 10%, the position would only have $0.9K worth of ETH left, but the borrowed ETH is only worth $0.9K; after unwinding the position and repaying the $0.9K worth of ETH debt, the position is left with exactly $1K USD stablecoin. Considering IL and non-zero borrowing costs, the position will be left with more ETH than it had borrowed when ETHUSDT sells off, leaving the position with positive ETH delta. Vice versa when ETHUSDT rallies. A dynamic hedge is required to keep the position delta neutral.

The second structure involves:

  1. In a leveraged farming protocol, select 3x leverage (total borrow $2K worth), borrow 25% ($0.5K worth) in USDT and 75% in ETH ($1.5K worth)
  2. Given the settings selected above, the position will be executed in essentially two parts:
    • For $0.25K USDT principal, the protocol borrows $0.5K USDT and converts half of the principal + borrow into $0.375K worth of ETH to form ETHUSDT LP; this position is effectively leveraged long $0.375K worth of ETH because it borrowed stable to buy ETH
    • For the rest of $0.75k, the protocol borrows $1.5K worth of ETH and sells $1.125K worth of ETH into USDT to form ETHUSDT LP; this leaves the position with $1.5-1.125=$0.375K worth of borrowed ETH, essentially short ETH
  3. Now the position is delta neutral at inception with $0.375K worth of leveraged long ETH and $0.375K worth of borrowed/short ETH

Again, with IL and borrowing costs, the position will have delta exposure as spot moves, which can be hedged dynamically with perps. Given larger leverage than the first 2x method, this strategy will have a faster drift away from delta neutrality and could be more easily liquidated.

Ending Thoughts

In a nutshell, delta neutral strategies can help investors reduce the correlation between traditional asset-owning investments and deliver returns without exposure to asset price movements. It allows investors to gain exposure to alternative risk premia that provides returns with higher Sharpe ratios.

Treehouse’s flagship portfolio analytics app, Harvest, provides historical NAV, P&L calculation, and performance attribution for users to gain insights into their delta neural positions in leveraged farming, on-chain spot/perpetual pairs, and automated vaults. For those who want to test the waters in delta neutral investing, Harvest helps them confidently navigate DeFi; for the more sophisticated veterans, Harvest adds value by aggregating, automating, decomposing, and visualizing what they already know.

Disclaimer

This publication is provided for informational and entertainment purposes only. Nothing contained in this publication constitutes financial advice, trading advice, or any other advice, nor does it constitute an offer to buy or sell securities or any other assets or participate in any particular trading strategy. This publication does not take into account your personal investment objectives, financial situation, or needs. Treehouse does not warrant that the information provided in this publication is up to date or accurate.