<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Growi Finance (growi.fi) on Medium]]></title>
        <description><![CDATA[Stories by Growi Finance (growi.fi) on Medium]]></description>
        <link>https://medium.com/@growi.fi?source=rss-5d642f9e8e45------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*7RA4BBrQuwdo3-IkGRoFYg.png</url>
            <title>Stories by Growi Finance (growi.fi) on Medium</title>
            <link>https://medium.com/@growi.fi?source=rss-5d642f9e8e45------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 12:22:28 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@growi.fi/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Resolv crashed to $0.02 in 17 minutes. Morpho still said it was worth $1]]></title>
            <link>https://medium.com/@growi.fi/resolv-crashed-to-0-02-in-17-minutes-morpho-still-said-it-was-worth-1-fea0f9af4d76?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/fea0f9af4d76</guid>
            <category><![CDATA[lending]]></category>
            <category><![CDATA[risk-management]]></category>
            <category><![CDATA[defi]]></category>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Sat, 22 Aug 2026 10:15:25 GMT</pubDate>
            <atom:updated>2026-08-22T10:15:25.554Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Dcc4-NbpVYXNucJ3.jpg" /><figcaption>Resolv Labs suffered an exploit that allowed 50 million USR to be minted without backing, prompting the protocol to suspend operations.</figcaption></figure><p><em>Article 3 of the Growi lending risk series · ~6 min read · Growi Finance</em></p><p>On 22 March 2026, a little after 2:00 UTC, <strong>an attacker with a stolen key minted ~80 million USR against roughly $200K of collateral. Resolv’s stablecoin crashed 97% in seventeen minutes, trading near $0.025 on Curve. </strong>In the Morpho market that accepted its wrapped version (wstUSR) as collateral, nothing happened. No liquidation. No repricing. The oracle was not reading a market price. It was reporting the wrapper’s own internal exchange rate and assuming the underlying stablecoin was still worth exactly $1.</p><p>The previous article looked at the collateral side: an uncapped, untimelocked mint role controlled by a single key. But the mint alone does not explain what depositors lost. <strong>The attacker extracted ~$25M from Resolv itself. The rest of the damage came later, through the price feed.</strong></p><p>For hours after USR was trading at a few cents, anyone could buy it cheaply, wrap it, post it as collateral at ~$1.13, and borrow real USDC. Roughly 6.2M USDC left that Morpho market and was lost for good. The debt has since grown past $75M as interest accrues, because no liquidator can act. Chaos Labs later concluded that most of the bad debt was created after the exploit, not during it.</p><p><strong>This is what an oracle decides.</strong></p><ul><li>Collateral that fails is a loss shared between borrowers, liquidators, and lenders.</li><li>Collateral that fails while the market still prices it at par lets someone hand you a worthless token at yesterday’s price and keeps letting them until someone notices.So what does the price feed behind a Morpho market actually do, and how do you tell one that would have caught the fall from one that could not?</li></ul><h3>What an oracle is (and what Morpho asks of it)</h3><p><strong>A blockchain cannot see prices. A contract only knows what is written on its own chain. An oracle is the bridge: a contract that publishes a price on-chain so other contracts can read it.</strong></p><blockquote>A Morpho market asks its oracle one question, repeatedly:</blockquote><blockquote><strong>What is one unit of the collateral worth in the loan asset?</strong></blockquote><p><strong>What is one unit of the collateral worth in the loan asset? </strong>That single number drives every Health Factor and decides which positions can be liquidated. The oracle is fixed at market creation. Collateral, loan asset, liquidation threshold, and oracle cannot be changed later. Oracle updates follow two rules:</p><ul><li><strong>Heartbeat</strong> — the maximum time it can stay silent (often measured in hours).</li><li><strong>Deviation threshold </strong>— the price move that forces an immediate update (commonly 0.25–0.5% for stablecoins).</li></ul><p>Between updates the market sees only the last published snapshot, which is why oracle charts look like staircases.</p><p><strong>Many Morpho oracles are not a single feed. They are assemblies: one leg for the underlying, one for the loan currency, sometimes a rate taken straight from the collateral token’s own contract. </strong>Each leg can be a real market price — or a fixed assumption quietly standing in for one.</p><h3>What is the oracle actually measuring?</h3><p>This is the first and most important question. Almost every oracle we examine falls into one of three categories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xv_QxAnpfEWPKC5AlXTBgA.png" /></figure><p><strong>Hardcoded</strong> prices sit at the heart of most Morpho bad-debt events. The feed returns either a constant or a value produced entirely inside the protocol it is pricing. <strong>The classic form is exactly what happened with Resolv and Stream Finance: a yield-bearing wrapper priced by its own redemption rate, with the underlying assumed to be a dollar.</strong></p><p>A feed that tracks the market follows the collateral down. A hardcoded one does not. Liquidations never fire. Bad debt builds instead.</p><p>A partial market price is not automatically worse than a real one. It carries one extra risk: the collateral can drift from the token it is priced through. That imports the weaknesses of the underlying protocol (the subject of the previous article). A partial price over a robust, verifiable underlying can still be safer than a real price scraped from a single thin venue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vR9UcXBm0dJVYEOG5OyoRw.png" /></figure><h3>The other ways a price can fail you</h3><p>Beyond what it measures, three properties decide how much weight a feed can carry:</p><ol><li><strong>Who can change it?</strong><br>Who can update or re-point the feed? Is that power behind a timelock or a single key? The most dangerous pattern is an externally-owned account that can rewrite the oracle with no delay. We read this on-chain, not from docs.</li><li><strong>Can we cross-check it?</strong><br>Can the price be verified independently, from transparent public inputs or against other feeds that do not share the same source? A frozen oracle is much harder to spot if you have nothing to compare it to.</li><li><strong>Does it arrive on time?</strong><br>A correct price is useless if it arrives late. How often does the feed freeze? How far is the deviation threshold from the price drop that would create bad debt? This requires quantitative comparison of historical oracle outputs against spot prices and other feeds.</li></ol><p><strong>We grade each property by severity. The feed’s overall quality is set by its weakest link</strong>. An oracle can be well-sourced, timely, and carefully governed and still be useless if it is blind to the market it is supposed to watch. <a href="https://chaoslabs.xyz/posts/oracle-risk-and-security-standards-an-introduction">Chaos Labs’ series on Oracle Risk and Security Standards</a> is a place we recommend our most curious readers to go deeper.</p><h3><strong>How does Morpho look right now?</strong></h3><p>We scanned every market whitelisted by Morpho’s verified curators (all chains, V1 and V2) and read each collateral oracle on-chain</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0wvltctMBqTujVJOv3OQsw.png" /></figure><p><strong>About 62% of TVL sits in markets priced by a real market price and 27% by a partial one. </strong>That is broadly healthy. The remaining 11%, roughly $430M is priced by a hardcoded oracle. That share is not confined to a corner of the protocol: around 24% of the 322 listed vaults and 16 of the 26 verified curators have at least one hardcoded-oracle market.Of that $430M:</p><ul><li><strong>~$300M is in stablecoins priced at a flat dollar or yield-bearing stablecoins priced by redemption rate over an assumed $1 underlying.</strong></li><li><strong>~$110M is in Pendle PT products priced from an internal accrual rate with no external market check.</strong></li></ul><blockquote><strong><em>This is the exact Stream Finance / Resolv pattern, still live today.</em></strong></blockquote><p><strong>We are not saying these vaults are poorly managed</strong>. Curators may have mitigated the risk through supply caps, monitoring, or collateral quality that a scan cannot see. <strong>What we can report is what the price feeds actually do. The judgement is yours.</strong></p><p>Almost all of this is reproducible: the feed’s source, the market it rests on, the keys that can re-point it. We would rather you read them on-chain than take our word for it.</p><p>An oracle is one of the cheapest things in a lending market to check — and the most expensive to get wrong.<br>- <strong>Collateral quality tells you how likely the fall is.<br>- The oracle tells you whether anyone will see it in time.</strong></p><p>The next article puts the two together: how these assessments become a risk view of an entire market, and then of a vault.</p><h3><strong>Sources</strong></h3><ul><li>Resolv USR key compromise, unbacked mint and depeg — The Block; CoinDesk; Chainalysis post-incident analysis.</li><li>Bad debt forming after the exploit rather than during it — Chaos Labs.</li><li>Oracle Risk and Security Standards series — Chaos Labs.</li><li>Stream Finance / xUSD collapse — BlockEden; CCN.</li><li>Oracle classification and TVL shares: Growi’s own scan (June 2026) of every market whitelisted by Morpho’s verified curators with AUM above $1M, via the Morpho API and direct on-chain reads across 8 chains. Expired Pendle PTs and tokenised RWA are excluded, and the remainder normalised to 100%.</li><li>Figures for the wstUSR/USDC market — Morpho API and on-chain reads.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fea0f9af4d76" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Priced at $1.26, Trading at $0.10: How to Spot a Doomed Collateral Before It Collapses]]></title>
            <link>https://medium.com/@growi.fi/priced-at-1-26-trading-at-0-10-how-to-spot-a-doomed-collateral-before-it-collapses-55373ee64baf?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/55373ee64baf</guid>
            <category><![CDATA[lending]]></category>
            <category><![CDATA[defi]]></category>
            <category><![CDATA[collateral]]></category>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Sat, 22 Aug 2026 09:36:41 GMT</pubDate>
            <atom:updated>2026-08-22T09:36:41.197Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/784/1*cdwdljhUTODn8F7LcnqhoQ.png" /></figure><p><em>Stream Finance’s xUSD erased roughly $285M of exposure across DeFi lending. None of the warning signs were hidden. </em><strong><em>These are the three questions that would have caught it.</em></strong></p><p><em>Growi Finance · Lending risk series · ~5 min read</em></p><p>In November 2025, <strong>xUSD</strong>, a token built to trade as a yield-bearing stablecoin, fell from $1.26 to under $0.10 in a matter of days. <strong>The collapse froze about $160M of user deposits and left roughly $285M of exposure spread across DeFi lending.</strong></p><p>This was no obscure asset<strong>: xUSD was widely supplied and, more to the point, widely used as collateral</strong>. On Morpho, isolation and curation contained most of the damage; even so, around $50M of bad debt still sits in its markets, about $700K of it borne by depositors in curated vaults.</p><blockquote><strong>None of what made xUSD dangerous was hidden</strong></blockquote><p>Every weakness was visible in advance, to anyone who knew where to look. This article is about learning to see them; the flaws that sank Stream’s token and many others before it.</p><h3>The backing could not be verified</h3><p>xUSD rested on an opaque mix: part off-chain capital run by an external manager, part on-chain recursive looping, scattered across private markets. None of it could be traced. <strong>No proof-of-reserves, a published, checkable demonstration that the backing assets really exist.</strong> No independent attestation. Just a DeBank address bundle standing in for reserves, and a transparency page that read “Coming soon”.</p><p><strong>When the external manager reportedly lost around $93M, most of the dollar behind the token went with it.</strong></p><iframe src="https://cdn.embedly.com/widgets/media.html?type=text%2Fhtml&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;schema=twitter&amp;url=https%3A//x.com/PeckShieldAlert/status/1985562216721235999&amp;image=https%3A//i.embed.ly/1/image%3Furl%3Dhttps%253A%252F%252Fpbs.twimg.com%252Fmedia%252FG44iTsSaMAA-uxU.png%253Alarge%26key%3Da19fcc184b9711e1b4764040d3dc5c07" width="500" height="281" frameborder="0" scrolling="no"><a href="https://medium.com/media/47913a8f3eb6ee7f04aa1213478939a8/href">https://medium.com/media/47913a8f3eb6ee7f04aa1213478939a8/href</a></iframe><h3>The circle nobody drew</h3><p>The same episode exposed something stranger. Elixir’s deUSD had lent roughly two-thirds of its own backing to Stream, as its only borrower. deUSD was backed by xUSD. And xUSD was partly backed by borrowed deUSD.</p><p><strong>A stablecoin backed by a token backed by loans of the stablecoin itself. Dramatic.</strong></p><p>The circular dependency sat buried in private Morpho markets, and it all but guaranteed the two would fall together. When xUSD broke, deUSD dropped around 98%.</p><h3>The market could not see it fall</h3><p>The oracle behind those markets went on pricing xUSD near $1.26 while it changed hands below ten cents. With the collateral still reported as healthy, liquidators had no trigger to close positions before the shortfall hardened into bad debt. Oracles are a subject of their own; the next article in this series is given over to them.</p><p>Which raises the real question: can the weaknesses behind DeFi’s biggest crashes be identified in advance? We think so. Here is the checklist we run at Growi before any asset enters our markets.</p><h3>The three questions we ask of every collateral</h3><h4>1. Is the value actually there?</h4><p>Tracing an asset’s backing is the first step in judging it. On-chain assets such as WBTC or wstETH can be read straight from the chain. Off-chain ones leave you dependent on attestations and independent proof-of-reserves — and even those are no guarantee. <strong>Main Street’s msY carried a live proof-of-reserve and still collapsed much as Stream did, falling about 85% and freezing an $18M Morpho vault.</strong></p><blockquote><strong><em>Off-chain ones leave you dependent on attestations and independent proof-of-reserves — and even those are no guarantee</em></strong></blockquote><p>Proof of backing never brings off-chain risk to zero, and the kind of backing weighs just as heavily: treasury bills (USDtb) or fiat reserves (USDC) are not the same bet as private credit or opaque trading strategies (Stream, Main Street). Check how the backing is distributed, too: whether it leans on a handful of counterparties, and what risks travel with it. <strong>When a token’s reserves are themselves other off-chain-backed tokens with shaky backing of their own, it inherits every one of those weaknesses.</strong></p><h4>2. Can it be created from nothing?</h4><p>Minting is the creation of new units of a token, and <strong>the danger lives in units with nothing real behind them</strong>. The usual route is a mint role written into the token contract: an address allowed to issue new tokens, only ever as safe as whoever controls it. If a single key holds a role that can print uncapped, unbacked tokens, one successful hack can undermine the entire protocol, as with Resolv. <strong>The threat is sharper still when no timelock guards the mint, leaving nothing between the exploit and the loss</strong>. Where these roles exist, they should be timelocked and held by a multisig or a regulated issuer before the token can be called safe.</p><p>A close cousin of this risk lives in the bridge — the machinery that moves a token between blockchains. Some bridges mint on the destination chain only once a message confirms tokens were burned on the origin chain. Compromise the role behind that message, and again tokens appear with nothing behind them. The Kelp exploit grew from exactly this: the role doing the verifying sat behind a single key.</p><blockquote><strong><em>If a single key holds a role that can print uncapped, unbacked tokens, one successful hack can undermine the entire protocol</em></strong></blockquote><h4>3. Could liquidators sell it in time?</h4><p>For a Morpho lender, this question carries real weight. What happens when liquidators cannot act because the collateral is too illiquid to sell? We went looking for the answer in our own data. We analysed 212 Morpho markets liquidated on Ethereum across the 14 heaviest stress days since 2024, August 2024 and October 2025 among them. The median market shed just 1.9% of its collateral. <strong>But the tail is brutal: roughly 1 in 4 liquidated markets lost more than 25% of their collateral in a single episode.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/974/0*b9qPmRDIazKTtyRA" /></figure><p><strong>A safe token, then, should hold enough liquidity to absorb a meaningful share of its markets being liquidated together.</strong> Measuring that means tracking the collateral’s true depth across DEX pools and aggregators (1inch, Odos, and the rest), estimating slippage under stress, and weighing how much debt a stress event is likely to throw up at once. This is the hardest of the three to judge, because it asks for quantitative work rather than a simple lookup, and it ultimately rests on each curator’s risk judgment.</p><h3>How we score it</h3><p>We grade each question by the severity of what we find, from minor caveats to critical flaws. <strong>The collateral’s overall quality is then set by its weakest dimension rather than its average:</strong> a token can pass two questions comfortably and still be rejected if the third is broken, because a crash finds the weak point, not the mean.</p><p>On top of the three questions,<strong> code quality acts as a penalty on the score: a good audit cannot rescue a bad collateral, while a bad one makes every other weakness more dangerous.</strong> And we treat every unverified area as a high risk in its own right; a less conservative curator might not. To us, an asset we cannot see cannot score well.</p><blockquote><strong>A fragile collateral is survivable when the market can see it fall, and lethal when it cannot.</strong></blockquote><p>Reading the collateral tells you how large the danger is, and how likely. xUSD showed the other half of the picture too: priced at $1.26 while it was worth cents. <strong>A fragile collateral is survivable when the market can see it fall, and lethal when it cannot.</strong> The next article turns to oracles, and to how Morpho markets set their prices.</p><p>This is how Growi curates lending vaults: every collateral in our markets has passed this framework before a single dollar is exposed to it. Explore our vaults at <a href="https://growi.fi/">growi.fi</a>, and follow us on X for the next piece in the series, oracles, and how xUSD stayed “healthy” all the way to zero.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1G2V4HBSZZcbxD1J" /><figcaption>Growi Lending Official Vault | HyperEVM | Built on Morpho</figcaption></figure><h3>Sources</h3><p><em>• Stream Finance / xUSD collapse — BlockEden; CCN.</em></p><p><em>• Kelp rsETH bridge exploit — CoinDesk.</em></p><p><em>• Resolv USR key compromise — DeFiPrime.</em></p><p><em>• Collateral figures on Morpho: Growi’s own on-chain analysis (Morpho API), cross-checked with public post-mortems. The 14 stress events are the 14 days with the highest number of liquidations registered by the Morpho API.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=55373ee64baf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The team behind Hypercore best-Sharpe vault is launching in DeFi lending. Risk is the core product.]]></title>
            <link>https://medium.com/@growi.fi/the-team-behind-hyperliquids-best-sharpe-vault-is-launching-in-defi-lending-97e75e59d3c9?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/97e75e59d3c9</guid>
            <category><![CDATA[quant]]></category>
            <category><![CDATA[lending]]></category>
            <category><![CDATA[defi]]></category>
            <category><![CDATA[vault]]></category>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Tue, 28 Jul 2026 09:42:54 GMT</pubDate>
            <atom:updated>2026-07-28T10:17:07.731Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FMhsljsCSt2eL7DDzAHpng.png" /></figure><blockquote>Article 1 of the Growi Lending risk series · Beyond APY: how we read risk in DeFi lending · ~6 min read</blockquote><p>Two years ago we launched a trading vault on Hyperliquid with a simple thesis: in DeFi, risk management is the real alpha. Today Growi HF is one of the largest vaults on the platform, with the best Sharpe ratio among the top vaults and around $15M under management across Hyperliquid, Hibachi, Pacifica and institutional accounts.</p><p>Next week, we open the next chapter: <strong>Growi Lending goes live</strong>.</p><p>Our first vault, <strong>Growi USDC Core</strong>, opens for public deposits on HyperEVM, built on Morpho — the same infrastructure trusted with over $6B in deposits. You deposit USDC; we do the rest. <strong>We expect it to deliver an average 7% yield on your stablecoins</strong>. And we are not launching on a promise alone. We have been running the vault quietly since early June, with our own capital, to prove the engine in production before opening the doors. The result: over its first six weeks, Growi USDC Core returned <strong>8.38%</strong> APY — ahead of every comparable USDC vault on HyperEVM. Six weeks at early scale is a data point, not a track record — but it is exactly the start we wanted, and this week you can join it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*Ypwu1YaJp78WZt06cYveUA.png" /></figure><h3>Why lending, and why us:</h3><p>DeFi lending is one of the largest sectors on-chain — Aave and Morpho alone rank #2 and #3 on DeFiLlama, with tens of billions supplied to earn yield. It is also a sector where the easy money is gone: the top stablecoin markets that paid 7–8% in 2024 now pay around 2–3%. As Morpho’s co-founder Paul Frambot put it:</p><blockquote><strong>“Undifferentiated lending converges toward risk-free rates.”</strong></blockquote><p>That word — <strong><em>undifferentiated</em></strong> — is why we are here.</p><p>Earning more than the risk-free rate means going where rates are higher, and higher rates always come attached to risks that most dashboards never show you. This is precisely the game Growi knows how to play. We spent two years managing leveraged futures, where a single unmanaged risk can wipe you as unfortunately happened to many teams &amp; individuals on 10th Octobre 2025 or 5th February 2026; <strong>that discipline — quantitative, automated, obsessive about downside — is what we are now bringing to lending</strong>. <strong>Think of it as lending+: the yield comes from the market, the edge comes from our risk engine.</strong></p><p>That edge has two layers:</p><ul><li><strong>A risk framework that never sleeps. </strong>We audit every collateral asset and every oracle before allocating a single dollar, rank each market accordingly, and monitor depeg, bad-debt and liquidity trap risk in real time.</li><li><strong>A dynamic allocation algorithm.</strong> Capital moves between approved markets as rates and liquidity shift, instead of sitting passively in whatever paid best last month.</li></ul><h3>Our vision abount Lending market:</h3><h4>The vault era has a concentration problem</h4><p>In roughly three years, DeFi lending has been rebuilt around two layers:</p><ul><li><strong>Neutral protocols like Morpho </strong>that provide the plumbing and the accounting.</li><li><strong>Layer of curators </strong>— risk teams who package overcollateralized loans into one-click portfolios and decide where your capital actually goes.</li></ul><p>That curator layer is now the dominant way to earn yield on-chain, and where serious money is arriving: Coinbase routes lending through it, Kraken plugs its exchange into it, and asset managers from Bitwise to Apollo have moved in. It is also where the real risk now lives. Independent research from Blockworks makes the point cleanly: <strong>when curators choose the markets, the oracles and the leverage, the meaningful credit decisions migrate off the protocol and onto them.</strong></p><blockquote>Competition among curators should be a race to the top on risk vs return not a race on APY.</blockquote><p>And that layer is narrow. <strong>A small handful of curators intermediates the majority of curated deposits, and the largest players tend to hold overlapping collateral and route through the same underlying markets — so their liquidity stress moves together. When one big curator is hit, others are prone to withdraw/evacuate alongside it.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dASatV2sZ5nCuvLWrY90dQ.png" /></figure><p>We intend to earn our place, and we want the teams around us to raise their game too: <strong>a lending market underwritten by a dozen strong, transparent risk shops is safer for everyone than one underwritten by three.</strong> That is why the first thing we are publishing, before the vault even opens, is not a yield number but our risk framework, in full, article by article. <strong>In a permissionless market with no regulator, transparency is the discipline that keeps everyone honest — and the standard professional allocators expect.</strong></p><h3>The numbers nobody puts on the deposit screen</h3><p>Why so much emphasis on risk in a product launch? Because in lending, the downside is not theoretical, and we would rather show you the data than sell you a dream.</p><p>To date, roughly <strong>$94M in bad debt, </strong>has built up in Morpho markets: about $50M from the Stream Finance exploit (October 2025) and $44M from the Resolv exploit (March 2026). It is still happening: around $42M is currently stuck in the msY (Main Street) and AVLT (Altura) markets, where depositors cannot withdraw. Aave carries roughly <strong>$86M in bad</strong> debt of its own, mainly 3 from the Kelp exploit (April 2026) — an event that also left about $5.4B of USDC and USDT frozen for days. Some of these losses hit large, well-regarded curators, drawn in by above-market yields.</p><p>Here is the pattern that matters: <strong>in every one of these failures, the root cause traces back to the collateral or to the oracle pricing it — and both were assessable before deposit.</strong> That is not hindsight; it is the core claim of the framework this series will walk you through, and the framework Growi USDC Core has been running every day since June.</p><h3><strong>Where we build: </strong>Morpho in sixty seconds</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ghtwOBZNO2pCoF-qEg_FFg.png" /></figure><p>We chose to build Growi Lending on <strong>Morpho</strong>, one of the largest lending protocols in DeFi. Unlike a shared-pool protocol such as Aave — where all collateral and loans sit together and one bad asset can shake the whole system, as the Kelp exploit showed — <strong>Morpho is built from isolated markets. Each market pairs one collateral asset with one loan asset, so a loss in one market stays inside it.</strong></p><p>There is a catch, and it is where we earn our keep: <strong>Morpho markets are permissionless.</strong> Anyone can create one, on any collateral, with any price feed, and the protocol makes no promise about quality. Loans are overcollateralized and protected by liquidations, but that protection is only as good as the collateral behind the loan and the oracle pricing it. Deciding which markets deserve your deposit is the curator’s job — and that is the role Growi plays in its vaults: your funds are lent only to markets that have passed our collateral and oracle audits.</p><p>(For the full mechanics, Morpho’s documentation is excellent; articles 2 and 3 of this series will go deep on the two risk pillars.)</p><h3><strong>What this series will cover</strong></h3><p>We are publishing our risk framework openly, article by article:</p><ol><li><strong>This article</strong> — the launch, and why risk is the product.</li></ol><p>2. <strong>Evaluating collateral</strong> — what actually backs the loan? Can it hold its value? Can the protocol behind it be exploited?</p><p>3. <strong>Evaluating oracles</strong> — how does the market price the collateral, and is it blind to the real market?</p><p>4. <strong>Market and vault risk </strong>— turning the collateral and oracle assessments into a score for any market or vault.</p><p>5. <strong>Real market evaluations</strong> — applying the framework to live markets, including our own vault.</p><p>6. <strong>Active risk management </strong>— what we keep monitoring after the deposit, and when we pull the trigger to exit.</p><p>By the end, the goal is simple: you will be able to look at a lending market and see roughly what we see.</p><p>Some depositors see only an APY, a market name and a Deposit button. Everything that determines whether that number is safe sits off the screen. If you are the kind of investor — or allocator — who would rather understand the risk than take it blind, <strong>this series is for you, and we would genuinely like to hear your views, questions and criticism in the comments.</strong></p><h3><strong>Be there when Growi Lending opens</strong></h3><p><strong>Growi USDC Core opens for deposits this week on HyperEVM, targeting 7% APY on USDC.</strong></p><p>Visit <a href="https://hf.growi.fi/">growi.fi</a> to be first in line when deposits open.</p><p>Follow the series, the launch announcement and our performance updates on Medium, Twitter/X (@<a href="https://x.com/GrowiFinance">GrowiFinance</a>, @<a href="https://x.com/GrowiHF">GrowiHF</a>) and <a href="https://www.linkedin.com/company/growi-fi/?viewAsMember=true">LinkedIn</a>.</p><p><em>Next up — Article 2. Evaluating collateral: what actually backs the loan?</em></p><p><em>Disclosure: Growi Lending and the Growi USDC Core vault are developed and operated by Growi Finance. Yields are variable and not guaranteed; figures shown are historical or target ranges, not a promise of future returns. Nothing here is financial advice.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wguUC_B0zpbwqnoaxleUiw.png" /></figure><h3>Sources</h3><p>DeFi yields falling below traditional savings, and the Frambot quote — CoinDesk, 7 Apr 2026.</p><ul><li><em>Kelp DAO bridge exploit and Aave bad debt — CoinDesk, 20 Apr 2026; DL News.</em></li><li><em>Resolv exploit and bad-debt cleanup — DeFiPrime.</em></li><li><em>Stream Finance exploit — BlockEden.</em></li><li><em>Bad-debt totals obtained via the Morpho API (Morpho markets) and on-chain for Aave (getReserveDeficit).</em></li><li><em>The $29M msY/AVLT freeze can be checked via the Morpho API.</em></li><li><em>5 Jun — 20 Jul 2026 APY comparison: on-chain convertToAssets() data, HyperEVM USDC vaults (Figure 1).</em></li><li><em>Curator-layer concentration, upward risk migration and tail co-movement — Zbandut &amp; Goldstein, Institutionalizing risk curation in decentralized credit, Blockworks Research, Dec 2025 (arXiv:2512.11976); Blockworks Research “Morpho Vaults” dashboard.</em></li><li><em>Institutional entry into curated vaults (Coinbase, Kraken DeFi Earn, Bitwise, Apollo/Morpho) — DIA Vault Map; DeFiPrime,</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=97e75e59d3c9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Surviving the storm: How 85% of HyperLiquid vaults failed, and what the survivors have in common]]></title>
            <link>https://medium.com/@growi.fi/surviving-the-storm-how-85-of-hyperliquid-vaults-failed-and-what-the-survivors-have-in-common-d3ed82b410f4?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/d3ed82b410f4</guid>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Fri, 15 May 2026 08:35:51 GMT</pubDate>
            <atom:updated>2026-05-15T08:37:36.874Z</atom:updated>
            <content:encoded><![CDATA[<p><em>By Growi Finance, May 2026</em></p><p>In September 2025, we published an in-depth analysis of HyperLiquid Vaults, ranking them by risk-adjusted performance and showing that headline APR doesn’t tell the full story. Eight months later, the ecosystem has grown dramatically, but two major market events have reshaped the landscape, leaving a trail of broken vaults behind. This is the story of what happened, who survived, and what separates the winners from the rekt.</p><h3>Part 1, the HyperLiquid vault ecosystem: growth and carnage</h3><p>HyperLiquid has cemented itself as a leading DEX, with total TVL reaching $5.04 billion as of May 2026, more than 3x the $1.6 billion we reported in September 2025. The vault ecosystem has grown alongside it: 9,444 user vaults have been created to date, and vaults hold $734 million in total value (of which $80 million belongs to external, user-operated vaults).</p><p><strong>But behind the growth numbers lies a harsh reality.</strong></p><h4>The numbers don’t lie</h4><p>We analyzed every single user vault on HyperLiquid (excluding HyperLiquid’s own vaults like HLP). Here’s what we found:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/661/1*qdFRodJXmqV5JKH786zw2g.png" /></figure><p>The net result: user vault investors have collectively lost $82 million more than they’ve earned. Nearly two-thirds of all vaults ever created are now closed, and only 23% have ever turned a profit.</p><p>Among “serious” vaults, those that handled more than $10,000 in cumulative PnL, the picture is even bleaker: 85% are now dead or abandoned.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*S7AxM9q89hGJVNMJpfrEuw.png" /></figure><h4>Survival by vintage</h4><p>Breaking down survival rates by when vaults were created reveals a clear pattern: older vaults have been through more storms, and most didn’t make it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/671/1*jNnU6EmdmyhliMJ-DzzoSw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*xwkRMJsoY7c4CZAnkdvqpw.png" /></figure><p>Three out of four vaults created before 2025 are now closed. Even among the most recent cohort, only about a third are profitable. The consistent message: building a sustainable vault strategy is extremely hard.</p><h3>Part 2, two storms: October 10 and February 5</h3><p>Since our last analysis, two major market events shook the HyperLiquid vault ecosystem. Each caused a different kind of damage, and together, they reshaped the competitive landscape entirely.</p><h4>The October 10 crash: The vault killer</h4><p>The market event around October 10, 2025 was devastating for the vault ecosystem. Using HyperLiquid’s public API data (which provides PnL snapshots roughly every 46 days, covering approximately mid-September through early November), we estimate the aggregate impact across all user vaults:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/662/1*2Af62kzAk-mwP_gFlx822Q.png" /></figure><p>$25 million wiped out in a single event window. The biggest casualties:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/677/1*9bVBAPfagP2nK7GjpOV93Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*JYVgeigBcGiAcDDHSBVeCQ.png" /></figure><h4>Community favorites that got rekt</h4><p>Perhaps the most telling evidence comes from vault-analyser.com, a community platform where users bookmark their favorite vaults. Tracking the “Top Favorites” leaderboard over time shows how beloved strategies were wiped out:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/658/1*W2o7VgLDjbLacXg3LaQQ6A.png" /></figure><p>Tedybe550, Stabilizer, and Spitfire were among the most followed vaults on the platform. All three were destroyed by the October 10 event. Not every disappearance is a rekt, Makrochronios and Amber Ridge closed with profits, but the pattern is clear: October 10 was a turning point for the community’s most-watched strategies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*0CASJnJeuOwfgd3Kc_NCZQ.png" /></figure><h4>The February 5 shock: Stress-resting the survivors</h4><p>If October 10 was the vault killer, February 5 was the stress test for whoever was left standing. This time, the aggregate numbers look smaller, but that’s because many large vaults were already dead. The real story is in the individual impact on the vaults that had survived October and grown since:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/658/1*ozUD-A6m1Ev49GDqs7u5Zw.png" /></figure><p>Most of today’s top vaults suffered drawdowns between 20% and 42% during the February 5 event. The key difference from October: these vaults recovered. They took the hit, absorbed the loss, and came back, which is ultimately what separates a surviving strategy from a rekt one.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*8pizKQuBBVGpW8xqpFq63A.png" /></figure><h4>Two events, two kinds of damage</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*r8PYMvwjAKwAM6siiE-ONw.png" /></figure><p>A note on transparency: Growi HF was profitable through the October 10 event. February 5 was a different story, it was a tough day for the entire ecosystem. Our drawdown was significantly larger than what HyperLiquid’s aggregated data shows, as is the case for all vaults (the API loses temporal resolution when aggregating data points). We recovered, but it was a reminder that no strategy is immune to market shocks.</p><h4>Part 3, the survivors: Risk-adjusted performance</h4><p>So who’s left standing? We filtered for user vaults that are at least 120 days old, currently profitable, and still actively managing capital. We added HyperLiquid’s own HLP vault as a benchmark. Here are the results, ranked by Sharpe Ratio, the gold standard for risk-adjusted performance:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/662/1*Mf_otj0g_bz9knKn0Pk-nQ.png" /></figure><h4>Risk vs. return: The picture is clear</h4><p>The scatter plot below maps annualized yield against annualized volatility for each vault. The dashed lines represent constant Sharpe Ratios, the higher and further left a vault sits, the better its risk-adjusted performance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*eXKH2m48KIEQLc7YgZwtMg.png" /></figure><p>Growi HF sits in the upper-left corner, delivering 64% annualized yield with only 25% volatility. Most other vaults cluster on the right side of the chart, with volatilities ranging from 130% to 250%. This isn’t a small difference: Growi HF’s volatility is roughly 6x lower than the average competitor.</p><h4>Sharpe ratio: The definitive ranking</h4><p>The Sharpe Ratio divides return by volatility, higher means more return per unit of risk. A Sharpe above 1.0 is considered strong in traditional finance. Above 2.0 is exceptional.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*fp_P41Z7u5PgVjczx_JZKw.png" /></figure><p>Growi HF leads with a Sharpe of 2.55, more than 2.5x the second-place vault (Orbit Value Strategies at 0.96). This isn’t just a marginal edge; it’s a fundamentally different risk profile.</p><h4>Cumulative performance over time</h4><p>The Time-Weighted Return (TWR) chart shows the cumulative growth of $1 invested in each vault since inception. TWR adjusts for deposits and withdrawals, giving a fair comparison of pure strategy performance regardless of vault size.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*ZuYkkpyNDpvri4Kr_FJTWg.png" /></figure><p>Notice the difference in trajectory smoothness. Some vaults show explosive growth followed by violent drops, classic high-volatility behavior. Growi HF (green line) shows the most consistent upward path, reflecting its low-volatility approach. After 670 days, $1 invested at inception would have grown to approximately $2.40, a 140% cumulative return.</p><h4>Drawdown profiles: Where the pain lives</h4><p>Drawdown charts show the percentage decline from each vault’s all-time high at any given point. They reveal the investor experience in a way that return charts don’t, this is what it feels like to hold through the bad times.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*o2iO91i-sfncVGmb0xuqxw.png" /></figure><p>The contrast is stark. Most vaults show deep, prolonged drawdowns, some dropping 30% to 50% from their peaks. Growi HF’s drawdown profile is visibly shallower, reflecting a strategy designed to limit downside exposure.</p><h4>Growi HF vs HLP: Independent strategy vs the house</h4><p>HLP (HyperLiquidity Provider) is HyperLiquid’s own vault, the largest on the platform. It serves as a natural benchmark. How does an independent strategy compare to the house?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*KRTjyjYWbCWKpUNNBPhojA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/657/1*8dwEED_0QaiYnXZx4TPulQ.png" /></figure><p>Growi HF delivers higher yield (64% vs. 42%) with dramatically lower volatility (25% vs. 152%), resulting in a Sharpe Ratio over 9x higher than HLP. This demonstrates that independent, specialized strategies can outperform the platform’s own vault on a risk-adjusted basis.</p><h4>Methodology &amp; limitations</h4><p>All data in this analysis comes from publicly available APIs: DeFiLlama for global TVL data, and HyperLiquid’s own public endpoints for vault-level data. No private or proprietary data was used.</p><p><strong>Important limitations to keep in mind:</strong></p><p>MDD is underestimated: HyperLiquid’s API does not use traditional financial candlesticks (open/high/low/close). Instead, it provides periodic snapshots of account value. When data is aggregated, intra-period peaks and troughs are lost. This means the Maximum Drawdown (MDD) reported by the API is lower than the real MDD for all vaults, including Growi HF and all competitors. We recommend treating MDD values in this analysis as lower-bound estimates.</p><p>Aggregate event data is approximate: The global PnL series from HyperLiquid has a resolution of approximately 46 days. When we report the impact of the October 10 and February 5 events using aggregate data, the “before” and “after” snapshots capture a wider window (roughly 6 weeks) rather than a precise date. For individual vault analysis, we used HyperLiquid’s detailed vault API which provides higher-resolution data.</p><p>Selection criteria: For the risk-adjusted analysis (Part 3), we selected user vaults that are at least 120 days old, currently profitable (positive lifetime PnL), and sorted by TVL. We then computed TWR, annualized yield, annualized volatility, Sharpe Ratio, MDD, and RoMaD using the detailed vault API. Hyperrr and pmalt were excluded due to extreme volatility / data inconsistencies that distort comparisons.</p><p><em>Disclosure: Growi HF is developed and operated by Growi Finance, the author of this analysis. While we have made every effort to present objective, data-driven results, readers should be aware of this potential conflict of interest. All data sources are public and the methodology is described above.</em></p><h3>Conclusion</h3><p>The HyperLiquid vault ecosystem is growing fast, but the numbers paint a sobering picture: $82 million in net losses, 85% of serious vaults dead or abandoned, and two major events that reshaped the competitive landscape entirely.</p><p>October 10 was the vault killer, it wiped out community favorites like Stabilizer, Spitfire, and Tedybe550 permanently. February 5 was the stress test, it hit the survivors hard, with most top vaults dropping 20% to 40%, but the strong ones recovered.</p><p>For investors choosing a vault strategy, the lesson is clear: don’t chase APR. Look at the Sharpe Ratio, look at volatility, look at track record length, and ask yourself, can this strategy survive the next storm?</p><p>At Growi Finance, we believe transparency in DeFi investments benefits everyone. We’ve been through both storms, October 10 was good to us, February 5 was not, and after 670 days and counting, we’re still here. That track record, combined with a Sharpe Ratio more than 2.5x the next best vault, speaks for itself.</p><h3>Invest in Growi HF</h3><p>Visit our website: <a href="https://hf.growi.fi/">https://hf.growi.fi/</a></p><p>Direct access to the protocol: <a href="https://app.hf.growi.fi/">https://app.hf.growi.fi/</a></p><h3>Connect with us</h3><p>Meet the team behind Growi on LinkedIn. Stay updated on protocol developments by following Growi Finance and Growi.HF on Twitter.</p><p><em>Previous analysis: Analysis of the Top 10 HyperLiquid Vaults and What Quality Returns Mean (September 2025).</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d3ed82b410f4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Analysis of the Top 10 HyperLiquid vaults, and what quality returns mean]]></title>
            <link>https://medium.com/@growi.fi/analysis-of-the-top-10-hyperliquid-vaults-and-what-quality-returns-mean-245fc1f5d2be?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/245fc1f5d2be</guid>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Mon, 08 Sep 2025 19:29:39 GMT</pubDate>
            <atom:updated>2025-09-08T21:44:48.010Z</atom:updated>
            <content:encoded><![CDATA[<h4>5-Sept-2025 — Growi Finance Research</h4><p>Hyperliquid’s vaults are growing fast, but headline APR doesn’t tell the full story. Beyond raw performance, we compare the <strong>Top-10 vaults by TVL</strong> and focus on <strong>risk-adjusted quality</strong> rather than raw yield.</p><p>In this article we take a closer look at the leading vaults, filtering for those with enough history to draw meaningful conclusions, and highlight which strategies truly stand out when risk and stability are considered.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/697/1*5tyXE1W-yy1MZBCYIjnNnQ.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>)</em></figcaption></figure><p>To make the comparison meaningful, we focus on vaults with at least four months of history. Normally we prefer six months as a minimum, but in this case lowering the threshold allows us to include more strategies in the analysis.</p><p>One exception is the Sifu vault. Its performance profile is very atypical, showing a persistent loss of USDC over time, which makes it less representative for a study centered on sustainable strategies.</p><p>So this is the group under review:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/637/1*PGpT-SI_5C-BbnToY9DqEg.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>)</em></figcaption></figure><p>If we look at the main performance metrics, the picture becomes clearer.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/706/1*2Y3oWnrZ0_YwUv5Bu3Qv0Q.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>)</em></figcaption></figure><p><em>Brief chart´s legend:</em></p><ul><li><em>Sharpe = Return ÷ Volatility</em></li><li><em>RoMaD (Return over Maximum Drawdown) = Annualized Return ÷ Maximum Drawdown. It measures how much return is achieved per unit of downside risk.</em></li><li><em>MDD (Maximum Drawdown) = The largest drop in value from a portfolio’s highest point to its lowest point, before reaching a new high.</em></li></ul><p>Notice how Growi HF stands out not only in terms of Sharpe but also in RoMaD, mainly because its drawdowns are so small compared to the others. In fact, the second vault with the lowest risk, <em>Stabilizer</em>, shows a maximum drawdown almost <strong>four times larger</strong> than Growi HF.</p><p>Leaving aside <em>Hyperrr</em> — whose volatility reaches levels that distort the comparison (around 30,000%) — this annualized view allows us to evaluate the risk-return trade-off. The steeper the slope, the more efficient the strategy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*osYZgTOA2hfaP2-9n8Ai8g.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>)</em></figcaption></figure><p>So Growi HF leads the pack, delivering the highest return per unit of risk.</p><p>But what does having a high Sharpe and RoMaD really mean in practice?</p><p>To illustrate this, let’s look at the drawdown curves of the different strategies:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/983/1*yjN1DDgZF3JEiT1nFQAJbw.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>)</em></figcaption></figure><p>Yes, Growi HF appears as only small orange blips — barely noticeable compared with other strategies. This becomes even clearer when analyzing the drawdowns in detail:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/943/1*8hBmA1HeinNgtGWdG-ip7A.png" /><figcaption><em>Source: HyperLiquid’s public API (</em><a href="https://api-ui.hyperliquid.xyz/info"><em>https://api-ui.hyperliquid.xyz/info</em></a><em>). </em>[There were some data issue with “Long Hype” and it wasn’t possible to calculate the DD parameters]</figcaption></figure><p>In this context, an episode refers to a period spent in drawdown — the time between a peak and the recovery of that peak. The metric “% Time Underwater” captures how often a strategy remains below its previous high.</p><p>Looking at these measures, Growi HF shows three clear advantages:</p><ol><li>Shorter drawdown episodes: The longest drawdown episode lasted only 19 days, and in half the cases drawdowns lasted less than a single day (median = 0).</li><li>Less time underwater: Growi HF spent only 29% of the time below its peak, compared with 40%, the second-best number.</li><li>Shallower losses: Its maximum drawdown was just -4.7%, far lower than any of the other vaults (most between -18% and -41%).</li></ol><p>In practical terms, this means that even in the worst case an investor in Growi HF would need to wait <strong>only 19 days before returning to profit</strong>. This translates into real peace of mind: there is no need to worry about the precise timing of entry, and withdrawals can typically be made quickly and with positive returns.</p><h3><strong>Conclusion</strong></h3><p>Overall, all the leading Hyperliquid vaults deliver very attractive returns. What sets Growi HF apart is its ability to achieve these results with far lower risk: its maximum drawdown is four times smaller than that of the next best strategy. This combination of strong returns and exceptionally low risk results in the best risk-adjusted profile across Hyperliquid.</p><p><em>By Growi Finance Research Team</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=245fc1f5d2be" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[HyperLiquid Top Vaults: Why Investors Are Shifting to Growi HF]]></title>
            <link>https://medium.com/@growi.fi/hyperliquid-top-vaults-why-investors-are-shifting-to-growi-hf-bf595f7ae45f?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf595f7ae45f</guid>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Thu, 22 May 2025 08:31:35 GMT</pubDate>
            <atom:updated>2025-05-22T08:31:35.652Z</atom:updated>
            <content:encoded><![CDATA[<p>21st May 2025 — Growi Research Team</p><p>When investors choose a vault on HyperLiquid, they often look first at size and reputation. For a long time, HyperLiquid’s own HLP vault has dominated the scene, boasting a large community and significant assets. But lately, more investors are starting to notice Growi HF — smaller, newer, yet quietly outperforming. What makes Growi HF attractive to these investors? Let’s compare them to understand better.</p><p>At first glance, the two vaults differ significantly in strategy. HLP earns profits from market-making and liquidation fees, keeping investors’ funds locked for four days. Growi HF, in contrast, employs automated, data-driven strategies, specifically designed for futures trading with higher leverage, aiming consistently for higher returns through precise real-time risk management.</p><p>Imagine you invested $1,000 in each vault back in July 2024. By today, your investment in HLP would have grown about 13%. However, your Growi HF investment would have increased by around 70%. That’s a striking difference. But raw returns only tell part of the story.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/622/1*sP7VWQ8-C93Slhhdua8WWQ.png" /><figcaption>Cumulative Performance (TWR) of Growi HF and HyperLiquid’s HLP</figcaption></figure><p>Looking closer, financial metrics suggest Growi HF offers appealing advantages. Its annual yield is notably higher at 86% compared to HLP’s 51%. While Growi HF has slightly higher volatility (18% compared to HLP’s 14%), it rewards investors more effectively for the risk they take, reflected in a stronger Sharpe ratio. Both vaults, however, boast impressive Sharpe ratios, significantly higher than what you typically find in the best hedge funds within traditional finance (TradFi) — so point for DeFi! ;)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/955/1*Ux9SkUrO2UJ2PEW-wHtBbg.png" /></figure><p>Beyond just the Sharpe ratio, when comparing other key metrics like the Sortino ratio and Maximum Drawdown (the maximum possible loss from a peak), Growi HF matches or slightly edges HLP. But the true standout is Growi HF’s Return over Maximum Drawdown (RoMaD), highlighting how effectively our strategy recovers and generates profit after downturns.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/605/1*fTkCDIdSh8E0KruiI5spow.png" /></figure><p>Increasingly, our clients allocate portions of their funds to both vaults, yet Growi HF consistently generates a higher share of profits:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*oaJlWgtLQ3MzqgPQWBiQzw.png" /><figcaption>Growi HF represents 20% of the investment and 30% of the profits</figcaption></figure><p>Some investors, encouraged by this reliability, now trust Growi HF exclusively, and we genuinely appreciate their support.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*N79VxLlhAGpTcdSReAGLEg.png" /></figure><h4>A Quick Note for Investors: Vault vs Protocol, What’s Live and What’s Next</h4><p>Growi HF is quickly approaching its maximum capacity on HyperLiquid, with only limited space left for new deposits. Due to platform rules, once the threshold is hit, it’s simply not possible to invest — so if you’re thinking about it, keep in mind that spots may run out fast. We deeply appreciate the community’s confidence, bringing us close to this threshold.</p><p>But there’s good news — you’re not limited to the HyperLiquid vault. Through our protocol, <a href="https://app.growi.fi/">Growi.fi</a>, you can already access Growi HF directly.</p><p>We’re also currently testing a second strategy, Growi HF 2 (<a href="https://app.hyperliquid.xyz/vaults/0x9b5b1931fa5838a891973f9877c7648c93ee80cc">link</a>), which is showing promising early results. In the near future, our protocol will include smart allocation across multiple vaults — Growi HF, Growi HF 2, and HLP — to optimize yields and balance risk. This integration is still in development, but will go live in just a few weeks — bringing smarter, more diversified investing to the protocol.</p><p>Ultimately, smart investing isn’t about picking the largest or oldest vault. It’s about finding the best combination of returns and manageable risk, something investors are increasingly discovering in Growi HF.</p><p>Explore Growi HF and see why thoughtful investors look beyond size to true performance.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf595f7ae45f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[We Analyzed the Top HyperLiquid Vaults — Which Ones Are Really Profitable?]]></title>
            <link>https://medium.com/@growi.fi/we-analyzed-the-top-hyperliquid-vaults-which-ones-are-really-profitable-b8fb43e1c6ee?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/b8fb43e1c6ee</guid>
            <category><![CDATA[defi]]></category>
            <category><![CDATA[hyperliquid]]></category>
            <category><![CDATA[investment]]></category>
            <category><![CDATA[yield-farming]]></category>
            <category><![CDATA[hedge-fund-investment]]></category>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Thu, 08 May 2025 15:18:16 GMT</pubDate>
            <atom:updated>2025-05-08T15:19:00.365Z</atom:updated>
            <content:encoded><![CDATA[<h3>We Analyzed the Top HyperLiquid Vaults — Which Ones Are Really Profitable?</h3><p><em>1st May, Growi Finance Reseach Team</em></p><p>As the HyperLiquid ecosystem continues to mature, users are increasingly drawn to vaults that deliver not just strong performance — but consistent profitability for individual followers. In this article, we break down an in-depth analysis of follower data across top-performing vaults, revealing how Growi HF stands out as one of the most effective strategies on the platform.</p><h3><strong>Methodology: How we measured vault performance</strong></h3><p>To assess which vaults are delivering real value to their users, we analyzed follower behavior and performance across the top vaults on HyperLiquid. The data was obtained through HyperLiquid’s public API (<a href="https://api-ui.hyperliquid.xyz/info">https://api-ui.hyperliquid.xyz/info</a>), limited to 100 clients per vault. We focused on the top 50 vaults that:</p><ul><li>Are older than 90 days</li><li>Allow new deposits (with two high-TVL exceptions included for reference)</li><li>Have complete and extractable data (27 vaults were excluded due to processing issues)</li></ul><p>The key metrics analyzed include:</p><ul><li><strong>Active followers</strong>: Users with more than $0.10 in TVL</li><li><strong>Winning followers</strong>: Active users with positive PnL</li><li><strong>VaultEquity 50%</strong>: <em>Median </em>user TVL</li><li><strong>VaultEquity average</strong>: <em>Mean </em>user TVL</li><li><strong>DaysFollowing</strong>: Duration of user engagement, with breakdowns by median, mean, and deviation</li></ul><p>By focusing on both central tendencies (median, average) and variability (standard deviation / average), we aimed to understand not just raw performance but <em>how consistently</em> vaults deliver that performance across their user base.</p><h3>Key Findings: Which Vaults Are Delivering the Best Results for Users?</h3><p>The results discussed in this section are also summarized in the chart just below this text, and are available in full in this published Google Sheets document [<a href="https://docs.google.com/spreadsheets/d/e/2PACX-1vSj8vOEcWzwPqAF_NaKvqH6HvWOrkRuApfsWXarQtjGUG8MnagYEJswTUCs3IoIVDsRtqUQQo8htGb7/pubhtml">link</a>].</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*4EExZ1KonG92xON85dUKCQ.png" /><figcaption>HyperLiquid vaults considered in this study</figcaption></figure><h4><strong>1. Growi HF has one of the highest win rates</strong></h4><p>Out of the vaults that allow deposits, Growi HF reports a 96% win rate among active users — a figure that matches the Hyperliquidity Provider (HLP) vault and is only surpassed by Stabilizer, which posts a 98% win rate.</p><p>This is remarkable when you consider that Growi HF is still a relatively young vault (296 days) compared to HLP (706 days), yet it has already attracted a loyal following of 100 users, all with positive track records.</p><p>For comparison, the chart below shows the PnL of both Growi HF and Stabilizer:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/427/1*KXsU_S9aWhjzFOFzROpMJw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*6pGMnl8jeieB6PfMOPXqLQ.png" /></figure><h4><strong>2. Client engagement and stickiness</strong></h4><p>Another interesting metric is the daysFollowing ratio, which measures how Vaults vary in how long users remain engaged. A very high average follow time — especially if much higher than the median — may suggest that only long-term users are staying, while new clients are not joining.</p><p>Elsewhere shows the highest average user retention, largely because it no longer accepts new deposits — so only long-term users remain.</p><p>Growi HF has a median follow time of 42 days and an average of 49. For a vault that’s been live for 10 months, this indicates steady interest from new users and a strategy that still inspires trust.</p><h4>3. Distribution of Client Capital</h4><p>Vaults on HyperLiquid differ not only in performance, but also in the profile of their user base. By analyzing how client capital is distributed, we gain insight into whether a vault appeals mainly to large investors, smaller retail users, or a balanced mix of both.</p><p>Martyrbit, for example, shows a highly concentrated user base, with both the median ($174,053) and average ($196,207) TVL per user indicating a preference for large-capital participants. Stabilizer, in contrast, leans heavily toward small investors, with a median of just $281 and an average of $5,646, despite its high win rate. Elsewhere sits in between, with a median of $829 and a much higher average of $56,574, suggesting some very large users skew the distribution.</p><p>In this context, Growi HF exhibits a healthy and balanced capital distribution. The median TVL per user is $974, while the average is $24,995 — a sign that it attracts both smaller and larger investors. This balance, combined with strong user outcomes, positions Growi HF as a strategy accessible to a broad audience across the HyperLiquid ecosystem.</p><h3>Conclusion</h3><p>The follower-level data shows that some vaults on HyperLiquid not only achieve strong user outcomes but do so with consistency.</p><p>Among them, Growi HF delivers one of the most reliable and profitable user experiences, with 96% of its active users reporting positive PnL, while remaining accessible to both smaller and larger investors.</p><p>Matching the percentage of profitable users seen in the flagship HLP vault, and outperforming it in both overall yield and Sharpe ratio, Growi HF is becoming a go-to strategy for users seeking dependable, steady returns.</p><p>As the ecosystem grows and more users explore vault strategies, consistently high-performing vaults like Growi HF — where clients not only succeed but continue to see positive results — are likely to play a key role in shaping the future of decentralized asset management.</p><h3><strong>How to Invest in Growi HF</strong></h3><p>If you want to invest in <strong>Growi HF</strong>, visit our website:<br> ➡️ <a href="https://hf.growi.fi/">https://hf.growi.fi/</a></p><p>For direct access to the protocol, visit our web app:<br> ➡️ <a href="https://app.hf.growi.fi/">https://app.hf.growi.fi/</a></p><h3><strong>Want to Learn More? Connect With Us!</strong></h3><ul><li>Meet the team behind Growi on <a href="https://www.linkedin.com/company/growi-fi/people">Linkedin</a>.</li><li>Stay updated on protocol developments and our new upcoming strategy by following: <a href="https://x.com/GrowiFinance">Growi Finance</a> and <a href="https://x.com/GrowiHF">Growi.HF</a> on Twitter</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b8fb43e1c6ee" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Breaking Down the Financial Performance of HyperLiquid Vaults]]></title>
            <link>https://medium.com/@growi.fi/breaking-down-the-financial-performance-of-hyperliquid-vaults-6c9b86ac466f?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/6c9b86ac466f</guid>
            <category><![CDATA[hyperliquid]]></category>
            <category><![CDATA[hedge-funds]]></category>
            <category><![CDATA[crypto]]></category>
            <category><![CDATA[vault]]></category>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Tue, 11 Feb 2025 13:59:19 GMT</pubDate>
            <atom:updated>2025-02-18T11:51:19.972Z</atom:updated>
            <content:encoded><![CDATA[<p>HyperLiquid <strong>Vaults</strong> represent a <strong>growing opportunity </strong>within the ecosystem, yet their overall performance, risk, and potential remain largely unexplored. While most attention within the HyperLiquid protocol has been on <strong>trading volume and fees</strong>, we took a <strong>broader </strong>approach, analyzing the <strong>entire Vault landscape</strong> — examining <strong>TVL</strong>, overall <strong>profitability</strong>, and the <strong>age </strong>distribution of Vaults. At the <strong>strategy level</strong>, we assessed <strong>annualized returns, volatility, and Sharpe ratio</strong> to identify the best-performing Vaults and highlight key risks. Our goal is to provide valuable insights to the community, helping investors navigate this landscape while also showcasing how <strong>Growi HF stands out</strong>. Below is a <strong>TL;DR summary</strong> of our findings — read on for the full analysis.”</p><p><strong><em>Table of contents:</em></strong></p><ul><li>TL;DR (Summary)</li><li>Overview of HyperLiquid Vaults: Basic Statistics and Key Insights<br> — TVL, Age, APR &amp; Profitability</li><li>Analysis of the financial performance (including risks) of selected vaults<br> — Selection criteria<br> — Risk-adjusted financial performance</li><li>How to Invest in Growi HF</li><li>Contact us &amp; feedback</li></ul><h3>TL;DR (Summary)</h3><p>The <strong>HyperLiquid Vaults</strong> ecosystem provides a way for investors to access automated strategies, but <strong>each vault has different risk and return profiles</strong>. To help the community <strong>better understand performance</strong>, we analyzed key financial metrics like <strong>annual returns, volatility, and Sharpe ratio</strong>.</p><h4>🌍 HyperLiquid Vaults: A Growing Ecosystem</h4><p>HyperLiquid has gained strong traction, further boosted by its recent airdrop that distributed billions of dollars. While most attention has been on trading volume and fees, <strong>Vaults are emerging as a key part of the ecosystem</strong>.</p><p>📈 The <strong>total TVL in HyperLiquid is $1.6B</strong>, with <strong>Vaults holding $442.8M</strong> (27% of the total).</p><p>💰 However, <strong>most of this belongs to HyperLiquid’s own Vaults</strong> — <strong>external Vaults hold just $35.4M</strong>, highlighting major growth potential.</p><p>🛠️ Vault creation is active, with <strong>2,396 open Vaults</strong>, but <strong>only 16.2%</strong> have been profitable in the last 30 days. Only <strong>20.1% of Vaults have a lifetime profit</strong>, showing that careful selection is crucial.</p><p>🔍 <strong>Vaults provide transparency and flexibility</strong>, enabling users to monitor performance within the HyperLiquid ecosystem. Enhancements like <strong>customizable fees, scheduled fee collection, and personalized lock-in periods </strong>could drive further adoption.</p><h4>📊 Understanding Vault Performance</h4><p>To help the community compare different Vaults, we analyzed their <strong>risk-adjusted performance</strong>, focusing on:</p><p>✅ Only <strong>24 strategies met the defined criteria</strong> of being at least 6 months old, having $20k+ TVL, and showing profitability since inception.</p><p>📊 <strong>Returns vs. risk</strong>: Some vaults showed high returns but extreme volatility, with some exceeding 500% annualized volatility, making consistency a key challenge.</p><p>💡 <strong>Sharpe ratio analysis </strong>helped identify the most consistent performers, where <strong>Growi HF had the highest Sharpe among all Vaults analyzed</strong>.</p><p>📊 <strong>Compared to HyperLiquid’s own Vault (HLP)</strong>, <strong>Growi HF delivered higher yield </strong>(over 100%) <strong>with slightly lower risk</strong> (19%), based on direct data from the HL Vault API. This shows that independent Vaults can provide strong opportunities beyond HL’s strategies.</p><p>At Growi.fi, we believe transparency in DeFi investments benefits everyone. Our analysis shows that <strong>HyperLiquid Vaults hold great potential</strong>, and careful strategy selection is key to maximizing returns.</p><p>🚀 <strong>Growi HF stands as a top performer, balancing strong yield with controlled risk</strong>.</p><h3>Breaking Down the Financial Performance of HyperLiquid Vaults</h3><p><strong>February 07, 2025</strong> — by <a href="https://hf.growi.fi/">Growi Team</a> (growi.fi)</p><p>HyperLiquid has been gaining popularity for some time, and its recent airdrop has <strong>further increased its visibility</strong>. This event gave away hundreds of millions of dollars to participants. As the token price increased, the total value of the airdrop grew to <strong>several billion dollars</strong>, benefiting more than 94,000 addresses.</p><p>Since then, people have been closely watching HyperLiquid’s <strong>trading volume and fees</strong>. These numbers help estimate how much the HYPE token might be worth. However, one important part of HyperLiquid that many have overlooked is<strong> its Vaults</strong>.</p><p>While the DEX holds most of the total value locked (TVL), we believe <strong>Vaults could play a big role in HyperLiquid’s future growth</strong>. More businesses are starting to build on HyperLiquid — just like we have at Growi.fi — which means Vaults could become much more important.</p><p>In this analysis, we will explore<strong> HyperLiquid Vaults from an investor perspective</strong>. Our goal is to provide <strong>clear insights </strong>that help compare different Vaults, identify the best-performing ones, and understand broader trends.</p><p>The steps in this analysis are: 1. Present <strong>basic statistics</strong> about HyperLiquid and its Vaults. 2. <strong>Analyze all existing Vaults</strong> and summarize key numbers. 3. <strong>Select Vaults</strong> based on key criteria such as age, activity status, and TVL. 4. <strong>Evaluate the financial performance</strong> of these selected Vaults and highlight the top performers.</p><h3>Overview of HyperLiquid Vaults: Basic Statistics and Key Insights</h3><h4>TVL</h4><blockquote><em>As of 7th of February of 2025, </em><strong><em>the total value locked (TVL) </em></strong><em>in the HyperLiquid protocol stands at </em><strong><em>1.3 billion</em></strong><em>, with</em><strong><em> Vaults collectively holding 556.3 million, representing 42.1%</em></strong><em> of the total TVL. However, since most of these funds belong to HyperLiquid’s own Vaults, which we will analyze separately, the actual proportion of externally managed Vaults is significantly smaller.</em></blockquote><blockquote><em>There are </em><strong><em>6688</em></strong><em> Vaults in total, of which </em><strong><em>2392 </em></strong><em>are currently open, representing </em><strong><em>35.8% </em></strong><em>of all Vaults. This relatively low percentage suggests that many Vaults may have been created for testing purposes. Closed Vaults have zero TVL and will not be included in our analysis, so our focus will be on the 2392 open Vaults.</em></blockquote><p>To better understand the distribution of TVL across Vaults, we begin by ranking the top 10 Vaults by TVL:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/335/1*0SbRJEDmuG7A2ClDO58QTg.png" /></figure><p>The <strong>largest Vaults</strong>, with a significant difference from the rest, <strong>belong to HyperLiquid itself</strong>. These include <strong>HLP</strong>, which acts as the parent Vault for other strategies such as A, B, and Liquidator. The total TVL of HyperLiquid Vaults is represented by HLP, which includes the TVL of its child strategies.</p><p>By <strong>removing HyperLiquid’s own Vaults</strong>, we get a different perspective on the remaining Vaults as a group.</p><blockquote><em>The combined value of all Vaults, excluding the Vaults owned by HyperLiquid, is </em><strong><em>32.8 million</em></strong><em>. </em><strong><em>These Vaults represent just 2.5% of HyperLiquid’s total TVL</em></strong><em>, highlighting a significant potential for growth.</em></blockquote><p>Vaults offer a <strong>clear and transparent way for other protocols to build on HyperLiquid</strong>, as they allow easy tracking of strategies. However, <strong>more flexibility could encourage broader adoption</strong>. Features like <strong>customizable fees, scheduled fee collection, and personalized lock-in</strong> periods could make Vaults even more attractive. While <strong>Accounts </strong>can also be used for building on HyperLiquid, they lack the visibility and auditability of Vaults.</p><p>The following graph shows the <strong>distribution of Vaults by TVL</strong> (excluding HyperLiquid’s own Vaults). As expected, the distribution follows a Pareto-like pattern, with a few large Vaults and many smaller ones.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/714/1*CK6eMw7cerZQJDxvtlOmvQ.png" /></figure><p>Let us illustrate this further with a <strong>Pareto Chart</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/631/1*VzI_2fgZE6SNVVFBkxd97g.png" /></figure><h4>Age</h4><p>Next, we examine the <strong>age of the Vaults</strong>:</p><blockquote><em>Oldest vault: 712 days (~2.0 years)</em></blockquote><blockquote><em>Newest one: 0 days</em></blockquote><blockquote><em>Average age: 185 days (~6.2 months)</em></blockquote><blockquote><em>Median age: 124 days (~4.1 months)</em></blockquote><p><strong>Most Vaults are relatively new</strong>, though there are some exceptions. The histogram below shows the distribution of Vault ages:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/712/1*ExxGAKWweJbFMKSxYRjFSQ.png" /></figure><p>Interestingly, most Vaults are only 1–3 months old, but there is also a <strong>peak around the one-year mark</strong>.</p><p>Below is a list of the Vaults that have been active <strong>the longest</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/771/1*l_J_xm_iDKl0urXIWbhXaQ.png" /></figure><h4>APR &amp; Profitability</h4><p>HyperLiquid provides <strong>APR data</strong> for all Vaults. However, it is important to note that APR is calculated based only on the last 30 days, which does not accurately reflect long-term performance. Still, let us take a quick look at the <strong>number of Vaults currently showing a positive APR</strong>.</p><blockquote><em>Out of 2392 open Vaults, </em><strong><em>only 288 have a positive APR</em></strong><em>, meaning just </em><strong><em>12.0%</em></strong><em> were profitable in the last month. This suggests that recent market conditions have been challenging.<br>The highest APR recorded this month was </em><strong><em>7789%</em></strong><em>, while the lowest was </em><strong><em>-100%</em></strong><em>. The </em><strong><em>average APR is 22%</em></strong><em>, but the </em><strong><em>median is 0.00%</em></strong><em>.</em></blockquote><p>A more useful metric is the <strong>total profit and loss (PnL) accumulated by each Vault over its lifetime</strong>. This gives a clearer picture of actual earnings.</p><p>First, let us determine if HyperLiquid Vaults (excluding HyperLiquid’s own Vaults) have generated <strong>net positive wealth:</strong></p><blockquote><em>The</em><strong><em> total PnL across these Vaults is $-19.2 million</em></strong><em>, highlighting the </em><strong><em>challenge </em></strong><em>of achieving consistent </em><strong><em>profitability </em></strong><em>in this environment.</em></blockquote><blockquote><em>· Largest single Vault loss: $-6.2 million million<br>· Largest single Vault profit:</em><strong><em> $2.0 million</em></strong><em><br>· Average PnL: </em><strong><em>$-8042</em></strong><em><br>· Median PnL: </em><strong><em>$-99.5</em></strong></blockquote><blockquote><em>These numbers suggest that most Vaults hover around the break-even point. In terms of profitability,</em><strong><em> only 21.1% (504 Vaults) have a positive PnL</em></strong><em>, reinforcing the importance of selecting well-performing Vaults to maximize returns..</em></blockquote><p>The following histogram shows the <strong>distribution of PnL across all Vaults</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/713/1*Vnzpu-seDVX469QVBaVLag.png" /></figure><p>The PnL distribution resembles a <strong>normal curve</strong> with a slight skew toward negative results.</p><p>Measuring yield since inception would provide better insights, but using aggregated data from HyperLiquid makes this challenging. <strong>Deposits and withdrawals significantly impact TVL and PnL</strong>, making general statistics less reliable.</p><p>To gain a more accurate understanding of each Vault’s actual performance, we will now conduct a <strong>detailed analysis</strong> using direct API calls to individual Vaults within HyperLiquid. The next section will focus on this <strong>in-depth evaluation</strong>.</p><h3>Analysis of the financial performance (including risks) of selected vaults</h3><h4>Selection criteria</h4><p>In this section, we analyze the <strong>financial performance</strong> of Vaults that meet specific criteria. We have defined what we consider reasonable for assessing a Vault’s viability:</p><ul><li>At least<strong> 6 months old</strong>, ensuring enough time to gather meaningful performance statistics.</li><li>At least<strong> $20,000 in TVL</strong> at the time of the study, assuming this represents a serious level of investment.</li></ul><blockquote><em>Only 39 strategies meet these criteria</em></blockquote><p>We then apply an additional filter: the vault must also have been <strong>profitable over its</strong> (minimum six-month) <strong>history</strong>.</p><blockquote><em>With this condition, only 24 vaults remain</em></blockquote><p>Since the number is small, we list them here:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/590/1*SO5fHB4bxqasWbk-YAltqw.png" /></figure><p>Among them, <strong>Growi.HF </strong>(<a href="https://hf.growi.fi/">https://hf.growi.fi/</a>) is our <strong>own developed strategy</strong>.</p><p>To compare performance, <strong>we also include HyperLiquid’s own Vault</strong> (HLP) alongside the selected strategies.</p><p>To extract relevant financial parameters for our analysis, we processed the data. However, some <strong>Vaults encountered issues</strong> in data retrieval and processing, and as a result, they will not be included in our analysis:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/355/1*BWfFeAoR5_krg3XA9UHtzg.png" /><figcaption><em>(Funny name if you speak Spanish!)</em></figcaption></figure><h4>Risk-adjusted financial performance</h4><p>Sorting by annualized yield, we see some staggering returns! Here are the top five:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/290/1*-yVa-DMu6T5NSfc_BJ9CbA.png" /></figure><p>However, we need to put these returns in context by <strong>including volatility</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/564/1*oJLm1VRsGBsdujS1SdNk5g.png" /></figure><p>These Vaults also exhibit <strong>extreme volatilities</strong>. A volatility of 800% with an average return of 7,000% represents the standard deviation of returns over a year. Theoretically, this means there is a 32% probability that the return in a year falls outside the range of 7,000% ± 800%.</p><p>At first glance, this may seem like a great deal — after all, you always win, right? But such large swings usually indicate <strong>inconsistency</strong>, with <strong>extreme ups and downs</strong>.</p><p><strong><em>Is Volatility Underestimated?</em></strong>: In reality, volatility in all strategies is likely even higher. The HyperLiquid API (and website) does not use traditional financial “candles” (open, high, low, close) but instead samples historical PnL or TVL. This means the actual highs and lows are likely greater than the reported figures.</p><p>Below, we visualize <strong>annual yield vs. annualized volatility:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/806/1*54olfnwinSkNJRvwSKQlUg.png" /></figure><p>For most investors, choosing a strategy with volatility above 50% might not be ideal. A strong strategy should be consistent, with returns behaving <strong>predictably </strong>rather than wildly fluctuating.</p><p>The graph clearly shows that <strong>higher returns come with higher volatility</strong> (note the <strong>log scale</strong> axes). A strategy is <strong>superior </strong>when, for the <strong>same level of risk</strong>, it achieves a <strong>higher yield</strong> (i.e., for a fixed x-axis value, the higher y-axis points are better).</p><p>This leads us to the <strong>Sharpe Ratio</strong>, a widely used measure that compares returns relative to risk:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/318/1*5e_Yf8AW6Jmwoc-I-eoZaQ.png" /></figure><p>Since <strong>quantitative </strong>strategies typically <strong>ignore the risk-free rate</strong>, we simplify the formula to:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/78/1*Yw1lDzTzJ5R4rQRoKBo54Q.png" /></figure><p>This ratio compares average returns to standard deviation, where <strong>higher values indicate a more stable and reliable strategy</strong>.</p><p>Other risk-adjusted metrics, such as <strong>Sortino </strong>Ratio, <strong>Calmar </strong>Ratio, or <strong>RoMaD</strong>, also exist. For this analysis, we focus on Sharpe Ratio, but future studies could incorporate these alternatives.</p><p>Now, let’s zoom in on strategies with <strong>volatilities below 100%</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/835/1*wkiwpZTglPKwArHuWhFeFQ.png" /></figure><p>Using the Sharpe Ratio, the <strong>best-performing strategy</strong> corresponds to the point with the <strong>highest slope</strong> from (0,0) in our chart.</p><p>Let’s visualize this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/845/1*EH5aNyBziuqSIRRJfDgw_g.png" /></figure><p>As shown, <strong>Growi HF has the highest Sharpe Ratio</strong>, but several strong Vaults on HyperLiquid also exhibit competitive risk-adjusted performance.</p><ul><li>Growi HF exhibits<strong> slightly less risk</strong> than HyperLiquid’s own Vault, yet delivers <strong>significantly higher returns</strong></li><li>While Growi HF manages <strong>fewer funds</strong>, it likely cannot scale to HyperLiquid’s level due to its controlled leverage — but its volatility remains well-managed, demonstrating that great opportunities exist beyond HyperLiquid’s own Vaults.</li></ul><p>If we examine the tweet below, <strong>many HyperLiquid Vaults outperform major hedge funds in TradFi</strong>. As DeFi traders know, the space is full of high-yield opportunities!</p><iframe src="https://cdn.embedly.com/widgets/media.html?type=text%2Fhtml&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;schema=twitter&amp;url=https%3A//x.com/TheShortBear/status/1877116819862892983%3Fref_src%3Dtwsrc%255Etfw&amp;image=" width="500" height="281" frameborder="0" scrolling="no"><a href="https://medium.com/media/613cd9d0afba8b606632d9318f55eeea/href">https://medium.com/media/613cd9d0afba8b606632d9318f55eeea/href</a></iframe><h3>How to Invest in Growi HF</h3><p>If you want to invest in <strong>Growi HF</strong>, visit our website: <br>➡️ <a href="https://hf.growi.fi/">https://hf.growi.fi/</a></p><p>For direct access to the protocol, visit our web app: <br>➡️ <a href="https://app.hf.growi.fi/">https://app.hf.growi.fi/</a></p><p>⚠️ Note: We <strong>do not recommend investing directly through HyperLiquid</strong>. As mentioned earlier, due to factors such as customizable fees, scheduled fee collection, and lock-in periods, we plan to restrict deposits through HyperLiquid in the short term, allowing deposits only through our protocol.</p><h3>Want to Learn More? Connect With Us!</h3><ul><li>Meet the team behind Growi on <a href="https://www.linkedin.com/company/growi-fi/people">Linkedin</a>.</li><li>Stay updated on protocol developments and our new upcoming strategy by following: <a href="https://x.com/GrowiFinance">Growi Finance</a> and <a href="https://x.com/GrowiHF">Growi.HF</a> on Twitter</li></ul><h3>Your Feedback Matters!</h3><p>Did you find this study useful?<br>Would you like additional analyses or insights?<br>Let us know in the comments!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/824/1*2NB2nuJ6r2OMAR_Tp3KYYw.png" /></figure><p><em>This article was originally </em><a href="https://deepnote.com/app/growifi/Analysis-of-HyperLiquid-Vaults-by-Growifi-team-068ab19f-5133-4a52-89c1-e42ecc39f1f7?utm_content=068ab19f-5133-4a52-89c1-e42ecc39f1f7"><em>published in DeepNote</em></a><em>, where some tables provide additional data</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6c9b86ac466f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Slashing in proof-of-stake: What happens when validators fail]]></title>
            <link>https://medium.com/@growi.fi/slashing-in-proof-of-stake-what-happens-when-validators-fail-828ff144b400?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/828ff144b400</guid>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Fri, 29 Nov 2024 22:36:09 GMT</pubDate>
            <atom:updated>2024-11-29T22:36:09.665Z</atom:updated>
            <content:encoded><![CDATA[<p>In proof-of-stake (PoS) blockchains, validators play a vital role in maintaining network security and processing transactions. However, with this responsibility comes the risk of misconduct or negligence.</p><p>To ensure validators act honestly and reliably, PoS networks use a penalty system called slashing, which enforces accountability by reducing the staked cryptocurrency of validators who break the rules.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/704/1*XQnKawfY81vRXkPNjr4lDw.png" /></figure><p><strong>What is slashing and how does it work?</strong></p><p>Slashing is designed to deter bad behavior by imposing financial penalties on validators who act against the network’s interests. When a validator violates the rules, a portion — or, in severe cases, all — of their staked cryptocurrency is deducted. The primary aim of slashing is to protect the network from threats such as malicious attacks, negligence, or coordinated dishonesty.</p><p><strong>What triggers slashing?</strong></p><ol><li>Double signing: When a validator attempts to produce multiple conflicting blocks for the same slot, their stake is slashed to prevent fraud, such as double-spending.</li><li>Inactivity: Validators are required to remain online and perform their duties. Prolonged downtime or failure to validate blocks may result in penalties.</li><li>Surround voting: Validators colluding to confirm blocks that contradict the network’s consensus can be penalized for attempting to destabilize the system.</li></ol><p><strong>Implications of slashing</strong></p><ol><li>Network security: Slashing ensures validators are financially motivated to act honestly, reinforcing the network’s integrity and resilience against attacks.</li><li>Decentralization concerns: High penalties could deter smaller validators from participating, leading to potential centralization of power among larger, well-resourced entities.</li><li>Trust and accountability: Slashing reassures users that the system is designed to penalize bad actors, fostering confidence in PoS networks.</li></ol><p>By implementing slashing, PoS networks align validator incentives with network health, ensuring both accountability and security. However, designing balanced slashing mechanisms remains critical to maintaining decentralization and fairness.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=828ff144b400" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Gasless transactions: Making DeFi easier]]></title>
            <link>https://medium.com/@growi.fi/gasless-transactions-making-defi-easier-40a0bfdea01f?source=rss-5d642f9e8e45------2</link>
            <guid isPermaLink="false">https://medium.com/p/40a0bfdea01f</guid>
            <dc:creator><![CDATA[Growi Finance (growi.fi)]]></dc:creator>
            <pubDate>Fri, 22 Nov 2024 05:01:21 GMT</pubDate>
            <atom:updated>2024-11-22T05:01:21.223Z</atom:updated>
            <content:encoded><![CDATA[<p>One of the biggest barriers to adopting DeFi is the need for gas fees, transaction costs paid in the blockchain’s native token, like ETH.</p><p>For new users, this can be confusing and inconvenient, as they often must first purchase and manage these tokens before even interacting with a dApp.</p><p>Gasless transactions are emerging as a solution to this problem, simplifying blockchain interactions and making DeFi more accessible to everyone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*RqGoUZAj7N8AB7sTocpt9A.png" /></figure><p><strong>How gasless transactions work</strong></p><p>Gasless transactions use technologies like meta-transactions, where a third-party relayer submits the transaction on behalf of the user.</p><p>Instead of requiring users to hold native tokens, the gas fees are paid by the protocol, a dApp sponsor, or integrated into the application’s fee structure.</p><p>This abstraction eliminates the need for users to worry about managing native tokens, allowing for a seamless experience similar to traditional web applications.</p><p><strong>Implications for DeFi adoption</strong></p><ol><li><strong>User onboarding simplified</strong>: Gasless transactions remove a significant entry barrier for new users. Without needing native tokens, users can directly interact with dApps, significantly improving their first experience with DeFi.</li><li><strong>Enhanced user experience</strong>: For existing DeFi users, gasless models streamline interactions, making transactions faster and less cumbersome. This improvement is particularly valuable in applications like gaming, NFTs, or micro-transactions, where high gas fees could outweigh the value of the activity itself.</li><li><strong>Centralization</strong>: While gasless models improve accessibility, they often rely on third-party relayers or fee sponsors, which can introduce elements of centralization. Striking a balance between user experience and decentralization will be key to ensuring the long-term success of this innovation.</li></ol><p>Gasless transactions represent a critical step toward making DeFi truly user-friendly. By eliminating the complexities of gas fees, they are helping to bridge the gap between decentralization and mass adoption.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=40a0bfdea01f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>