Dragonfly's Qureshi: Hyperliquid's real rival is ICE, not Coinbase

Perps still have room to grow because the real market isn't crypto

Qureshi argues that perps onchain were a "missing market" for years until Hyperliquid "cracked the code" on UX, latency and performance, setting a pace competitors like Lighter and Variational are now following. He says Dragonfly backed dYdX and derivative projects as far back as 2018-2019 before those attempts failed on execution.

The bigger story, he says, is that crypto trading itself is not the total addressable market for these venues.

"Now already more than 50% of their volume is RWAs... So the bigger story is that okay, I'm not trying to be Coinbase. I'm trying to be ICE. I'm trying to be the New York Stock Exchange. I'm trying to be the NASDAQ."

He notes Hyperliquid is now a meaningful share of Binance's total volume, and that RWA trading on these platforms (oil, gold, indices) is still tiny relative to global CFD and regulated markets, which he sees as the real room to grow.

Asked how to underwrite valuations, he frames it as layered comparisons: Variational and Lighter versus Hyperliquid, Hyperliquid versus Binance and OKX, and the whole onchain perp complex versus Nasdaq, CME and Interactive Brokers. He puts Variational's fully diluted value around $1-2 billion and Lighter's around $4-5 billion against Hyperliquid's roughly $80-90 billion, implying they've captured only 1-5% of Hyperliquid's scale so far.

Vaults are the weakest link in Dragonfly's four-to-five category framework

Asked how Dragonfly allocates its roughly $4 billion in assets (following a $650 million Fund IV raise) across perps, prediction markets, tokenization, stablecoins and vaults, Qureshi says vaults are his least favorite.

"It's not obvious to me that they're going to have a lot of defensibility... this might be a very commodified product at the end of the day."

He argues issuers, not vault platforms, may capture most of the value, comparing it to how commodified financial products are often loss leaders for large institutions.

He also stresses that as a VC he tries not to overweight what's hot right now, pointing to Dragonfly's early bets on Lighter before Hyperliquid's rise and on Polymarket before the 2024 election as examples of positioning ahead of category growth.

Perps beat tokenization for the next year, but that could flip

Qureshi distinguishes tokenization from vaults: vault users are indifferent between competing products offering the same yield, but tokenization users care deeply about trading the most liquid version of an asset, which he says gives tokenization more economies of scale over time.

Still, he favors perps over tokenization on a one-year horizon.

"If you're talking about like one year from now, I would still bet on per over tokenization. But five years from now, that might change."

His reasoning: perps suit short-term directional trading with leverage and cash settlement, while tokenization suits long-term saving, and most crypto-native users today are trading, not saving. He expects saving behavior to migrate onchain eventually, especially via fintechs distributing diversified portfolios the way stablecoins are distributed today, but says that will take longer than perp growth.

Prediction markets face a wash-trading probe and a coming Supreme Court fight

On Kalshi's wash-trading allegations, Qureshi says the open-interest-to-volume ratio is "a little bit lopsided" and calls it "embarrassingly bad," though he adds it isn't central to how Kalshi makes money and that the data was publicly accessible via API rather than hidden.

On New York's attorney general targeting Polymarket as an unlicensed gambling operation, he says this isn't really about legality but about taxes:

"It's not that you can't do this, that you can do this, but you have to pay taxes to the states."

He explains there's now a circuit split, with some appellate courts ruling for prediction markets and others for states, which he expects the Supreme Court to take up, likely next year, possibly "splitting the baby" between federal and state jurisdiction. He expects Kalshi to be the likely defendant given it faces the most lawsuits.

Near and Zcash represent crypto's privacy "reawakening"

Qureshi, an investor in both Near and Zcash (and personally in the Zashi wallet), frames both as part of a broader privacy resurgence running alongside Bitcoin's institutionalization. He describes Near as building account abstraction across chains and venues, enabling activity like shielded Zcash transactions that apps such as Phantom don't support, contrasting the "show everyone my P&L" ethos with the desire to "disappear into the spectre of privacy."

He also downplays a link between Kalshi's wash-trading scrutiny and any acceleration of CFTC support for onchain perps, calling the two issues "pretty orthogonal," and expects Kalshi to face a fine rather than a major disruption. On prediction market advertising saturation, he expects public backlash to fade as the platforms mature and marketing shifts from acquisition to brand awareness.