Hyperliquid Market Maker Wallets: Why Wintermute's $115.76M Short Is Not a Bear Call
Hyperliquid market maker wallets made $25.44M since the midterms and Wintermute holds $115.76M of shorts. How to read MM flow without mistaking hedges for bets.
$ Stop reading delayed data. Read live order book depth on the 100 most liquid Hyperliquid pairs right now.
Launch Free Terminal →A Wintermute address on Hyperliquid has accumulated about $222 million of profit across spot and perpetual trading and holds $115.76 million of short open interest, according to an OnchainLens report carried by ChainCatcher on November 7. A day earlier, 71 Hyperliquid addresses labeled to six market making groups had added about $25.44 million of combined perp PnL since the US midterm elections, per TokenPost.
Those addresses show up in the same leaderboards traders use to gauge whale sentiment. Hyperliquid market maker wallets are visible in a way they are not on any centralized exchange, and that visibility is useful only if you read them for what they are. A $115 million short held by a liquidity provider is a very different signal from a $115 million short held by a directional fund.
Hyperliquid Market Maker Wallets Are Public Data
On Hyperliquid, every order, trade and liquidation runs on-chain, and every order is publicly verifiable, as CoinShares puts it. That turns market making from a black box into a dataset. On a CEX you infer liquidity providers from the shape of the book. On Hyperliquid, once a tracker labels an address, you can watch its positions and PnL directly.
The label is the weak link. TokenPost attributes the 71 addresses to Wintermute, Auros, Selini, Jump, Cumberland and Flow Traders, but the article does not name its data provider or explain how the addresses were labeled. Treat these as tracker attributions, not confirmed ownership.
The size of the activity is the strong part. In the 24 hours covered by a November 6 update, those addresses recorded about $5.101 billion of perp volume and about $13.59 million of net realized trading profit, per TokenPost. Their trading spanned digital assets, stocks and indexes, and some positions were reduced while large new exposure was added elsewhere.
Why a Market Maker Short Is Usually a Hedge
A market maker earns the spread by filling whoever crosses it. When aggressive buyers lift offers all day, the market maker ends up short by construction. That short can be a residual of filled flow, a hedge against spot inventory held elsewhere, or a funding carry trade. None of those is a view that price will fall.
The press has flagged this before. When OnchainLens showed Wintermute with $190.8 million of Hyperliquid shorts in August, split across $53 million of ETH, $30.7 million of BTC, $22.6 million of SOL, $11.4 million of HYPE and $10.2 million of XRP, NameCoinNews noted that large shorts may reflect hedging of inventory or funding exposure rather than a bearish bet. The same report put Wintermute's lifetime PnL at $203.6 million at the time, with $5.85 million of unrealized losses on the shorts.
Compare that with November. The accumulated profit figure is now about $222 million and the short open interest is $115.76 million, per ChainCatcher. The two reports may not measure PnL the same way, so the gap is indicative rather than exact. The direction is clear enough: the short book shrank by roughly $75 million while the profit line kept climbing, which is what a hedged liquidity provider looks like over a volatile quarter.
The other tell is turnover. Roughly $5.1 billion of volume in a single day against positions measured in the hundreds of millions means the inventory turns over many times. A directional whale holds; a market maker recycles.
When Market Makers Lose: The July Drawdown
Market makers do not always win, and the losing periods are the informative ones. Four Hyperliquid market makers, Wintermute, Auros Global, Cumberland and Flow Traders, posted a combined 30-day loss of $23.515 million on about $17.254 billion of volume, according to TradingBeats data reported by KuCoin on August 11. Since early July they had given back $27.075 million.
The losses clustered in mid-July volatility. Maximum 24-hour drawdowns were $5.651 million for Wintermute, $4.041 million for Cumberland, $1.937 million for Auros Global and $1.127 million for Flow Traders. Cumberland and Flow Traders had not shown recent two-sided quotes in that report.
This is adverse selection in its plainest form. A market maker loses when the flow it fills is informed, when the side crossing the spread knows where price is going. That is what VPIN tries to estimate from volume buckets. When VPIN rises and labeled market maker PnL turns sharply negative at the same time, liquidity providers tend to widen or pull quotes, and the book thins right when you need it.
What the Post-Midterm Profits Say About Flow
The November picture is the opposite. $13.59 million of realized profit on $5.101 billion of volume works out to about 27 basis points of volume in one day. That figure mixes spread capture with gains on hedges and inventory, so it is not a clean spread number. It does say the flow these accounts traded against was, on balance, not informed in that window.
After the midterm selloff and the liquidation waves around it, that fits. Forced liquidations are the least informed flow in the market: they cross the spread because a margin engine tells them to, not because they have a view. A market maker sitting on the other side of a liquidation cascade gets paid for absorbing it.
For a trader, that is the useful inversion. Rising market maker PnL during a selloff suggests the selling was mechanical. Falling market maker PnL during a selloff suggests someone knew something.
Reading the Whale Board With Market Makers in It
Whale snapshots rarely say whether they strip out liquidity providers. The CoinGlass snapshot carried by ChainCatcher on November 7 showed $7.573 billion of whale positions, 54.49% of it short, and the report does not state whether labeled market maker addresses are excluded. If one labeled address alone holds $115.76 million of short open interest across markets, the headline short percentage can carry a meaningful share of hedging flow.
A practical filter for any large Hyperliquid wallet: check holding period, check whether it trades both sides in the same market, check whether its volume is many multiples of its open position, and check whether its PnL is smooth or lumpy. Market makers score high turnover, two-sided activity and steady PnL. Directional whales score the reverse.
The whale tracker on Buildix follows large Hyperliquid wallets with their entries and size changes, which makes this kind of sorting a matter of minutes. Pairing that with VPIN and CVD on a pair page such as HYPE shows whether the flow those wallets are absorbing looks toxic or mechanical.
Hyperliquid's transparency is an edge only for traders who separate the people taking risk from the people renting out their balance sheet. Count the market makers as whales and the short side looks more crowded than it is. Watch their PnL instead, and they tell you how informed the other side of the market really is.