Pinned post
Why looping in crypto is broken, why we stopped, why you're being sold repackaged credit risk as a savings account, and why we don't need it anymore. Thanks to @chameleon_jeff and @HyperliquidX.
Before founding D2, we spent a decade in institutional derivatives.Started at Bloomberg, then moved to global multi-billion-dollar hedge funds, and finished with a few years at PAG, a $55B Asian multi-strategy fund, trading volatility as a PM responsible for a multi-billion notional book.
Across that span we sat next to some of the best credit teams in the business. So when we say what the market is being sold as "stablecoin yield" looks like a leveraged credit strategy in a wrapper, this isn't theory. It's pattern recognition from watching credit professionals price this class of risk for a living.
Nothing inherently wrong with leveraged credit.
There is something wrong with selling it as "yield" without pricing the tail.
@ethena has executed beautifully on a favorable market setup.
Funding has been positive, basis has been wide, delta-hedged shorts have earned. Mechanically it remains a leveraged basis trade that pays until funding flips, liquidity thins, or the short leg breaks. That is not a savings account. That is a credit hedge fund wrapped in USDe.
The same logic applies to every looping stablecoin variant: sUSDe loops on Aave, syrup loops on Maple, avUSD, others. Same structure. Borrow cheaper stable, lend richer yield-bearing stable, pocket the spread, repeat 5x.
Works until it doesn't.
This week we documented two of these positions at -93% and -135% net APY, HF 1.01 and 1.02, ~$1B combined gross exposure.
D2's on-chain auditable track record (attached):
+3.8% (Dec 2023)
+115% (2024)
+31% (2025)
+3% (2026 YTD)
2.6 Sharpe. 28 of 28 positive months. 200% cumulative.
We looped actively in February 2024. Long ETH looped into USDC on @aave , short ETH on @GMX_IO , at funding rates that peaked above 100% APR.
One of the trades that made 2024 our best year.
Nothing against looping when the risk is properly compensated.
In 2025 we did lighter looping on the @fundstrat ‘ ETH run from $1,500.
We exited when the setup started to look like a bubble, and rolled the delta exposure into call options on @DeriveXYZ. July expiry paid materially. August expired mostly worthless.
Net result: we captured the DAT mania with defined downside.
We did not need a crystal ball.We needed convex payoffs.Since then we have not found a looping setup worth taking. Not ideology, the risk-reward no longer clears.
The change that matters: Hyperliquid liquidity has made gamma replication, delta hedging, and structured payoffs executable at spreads that do not exist anywhere else on-chain. We can now replicate most of the payoff profiles we like, with defined downside and substantially better risk-reward, without taking liquidation risk at HF 1.01.
Do not let zero-fee gimmicks mislead you. Execution happens at the spread, not the fee. Hyperliquid spread is why this works. That is @chameleon_jeff and the team's real contribution to institutional DeFi.
For fund managers: if the risk profile of your product is "leveraged credit, short convexity, earn carry until you cannot," that is a legitimate strategy. Label it correctly. Do not package it as a stablecoin and compare it to money market yields.
For everyone else: if a product offers you double-digit "stable yield," ask what happens in tail conditions. "We have Umbrella coverage" or "the collateral is fully backed" is not an answer. That is a marketing line. Ask for the funding chart. The liquidity depth in stress. The concentration of the borrower base. The unwind mechanics.
We learned looping by running it when it paid. We stopped when it didn't.Looping pays carry until it doesn't. Options pay when it matters.
Thanks to @chameleon_jeff and the Hyperliquid team for making this possible.
Fade D2 at your own risk. 🫡
Before founding D2, we spent a decade in institutional derivatives.Started at Bloomberg, then moved to global multi-billion-dollar hedge funds, and finished with a few years at PAG, a $55B Asian multi-strategy fund, trading volatility as a PM responsible for a multi-billion notional book.
Across that span we sat next to some of the best credit teams in the business. So when we say what the market is being sold as "stablecoin yield" looks like a leveraged credit strategy in a wrapper, this isn't theory. It's pattern recognition from watching credit professionals price this class of risk for a living.
Nothing inherently wrong with leveraged credit.
There is something wrong with selling it as "yield" without pricing the tail.
@ethena has executed beautifully on a favorable market setup.
Funding has been positive, basis has been wide, delta-hedged shorts have earned. Mechanically it remains a leveraged basis trade that pays until funding flips, liquidity thins, or the short leg breaks. That is not a savings account. That is a credit hedge fund wrapped in USDe.
The same logic applies to every looping stablecoin variant: sUSDe loops on Aave, syrup loops on Maple, avUSD, others. Same structure. Borrow cheaper stable, lend richer yield-bearing stable, pocket the spread, repeat 5x.
Works until it doesn't.
This week we documented two of these positions at -93% and -135% net APY, HF 1.01 and 1.02, ~$1B combined gross exposure.
D2's on-chain auditable track record (attached):
+3.8% (Dec 2023)
+115% (2024)
+31% (2025)
+3% (2026 YTD)
2.6 Sharpe. 28 of 28 positive months. 200% cumulative.
We looped actively in February 2024. Long ETH looped into USDC on @aave , short ETH on @GMX_IO , at funding rates that peaked above 100% APR.
One of the trades that made 2024 our best year.
Nothing against looping when the risk is properly compensated.
In 2025 we did lighter looping on the @fundstrat ‘ ETH run from $1,500.
We exited when the setup started to look like a bubble, and rolled the delta exposure into call options on @DeriveXYZ. July expiry paid materially. August expired mostly worthless.
Net result: we captured the DAT mania with defined downside.
We did not need a crystal ball.We needed convex payoffs.Since then we have not found a looping setup worth taking. Not ideology, the risk-reward no longer clears.
The change that matters: Hyperliquid liquidity has made gamma replication, delta hedging, and structured payoffs executable at spreads that do not exist anywhere else on-chain. We can now replicate most of the payoff profiles we like, with defined downside and substantially better risk-reward, without taking liquidation risk at HF 1.01.
Do not let zero-fee gimmicks mislead you. Execution happens at the spread, not the fee. Hyperliquid spread is why this works. That is @chameleon_jeff and the team's real contribution to institutional DeFi.
For fund managers: if the risk profile of your product is "leveraged credit, short convexity, earn carry until you cannot," that is a legitimate strategy. Label it correctly. Do not package it as a stablecoin and compare it to money market yields.
For everyone else: if a product offers you double-digit "stable yield," ask what happens in tail conditions. "We have Umbrella coverage" or "the collateral is fully backed" is not an answer. That is a marketing line. Ask for the funding chart. The liquidity depth in stress. The concentration of the borrower base. The unwind mechanics.
We learned looping by running it when it paid. We stopped when it didn't.Looping pays carry until it doesn't. Options pay when it matters.
Thanks to @chameleon_jeff and the Hyperliquid team for making this possible.
Fade D2 at your own risk. 🫡
