Hyperliquid will never flip Binance in daily volume, here’s why:
Hyperliquid has done a great job at attracting both 1) retail flow and 2) maker liquidity. One of the ways they have done this include building a high performance state-layer with modifications that enshrine an order book.
One example of these modifications is how Hyperliquid handles order priority. Because Hyperliquid has consensus, each transaction has to buffer in the mempool — usually around 3 blocks on average. Hyperliquid prioritizing cancels helps maker liquidity because makers who run validators are able to see incoming profitable taker transactions against their maker orders and subsequently cancel and then amend their orders.
This is what we call a speed bump: a way for less sophisticated market makers to comfortably quote tight on a CLOB without worry of being picked off by some of the major HFTs (toxic flow).
Jeff’s recent tweets on transparency of the L1 increasing maker liquidity for whales makes sense if you understand that there is a speed bump in the exchange.
However, none of the major exchanges where price discovery lives for any asset, both Binance for crypto, and Nasdaq / CME for equities and futures implement any sort of maker-friendly speed bumps.
The reason for this is that this would ultimately constrain their volume, and price discovery would move elsewhere. A majority of Binance’s volume is not Market Maker <> Retail trader, it’s Market Maker vs Market Maker. If you make it so that the weaker market maker has the ability to get out of a losing position — then that will greatly reduce the ability for a market maker with actual alphas to make winning trades.
Let’s simplify this a bit: Hyperliquid has around $10-20b of daily volume (on a good day). On that same day, Binance has around 10x that ($200b). For Hyperliquid to match and surpass that volume, it would have to remove its speed bump. Period. Hyperliquid doesn’t want to do this (for a plethora of reasons, one such being that HLP is enshrined so heavily into the maker side of the CLOB).
This begs the question: Why is HL onchain and transparent in the first place? With their speed bump architecture, the answer is not for superior liqudity. The answer is because there are certain benefits of decentralization which we shall not name that they look to take advantage of.
When you start to look at these pieces from a Birds Eye view, you see how effective the HL team is at marketing but ultimately why their design decisions will constrain their liquidity (and hence make the flippening impossible).
At GTE we’re taking an opinionated approach that if we want to compete, we’re going to have to play a different game than what exists right now. Can we take the best of tradfi execution (this is not privacy, but rather a true price-time priority clob) and the best of decentralization (composable, permissionless, censorship resistant).
We call this the barbell strategy — HL took the middle path by owning the entirety of the stack (exchange, l1, matching, oracle) and implementing things like speed bumps, GTE is taking the other side of this by prioritizing two major groups, 1) retail and 2) sophisticated HFTs.
What will this look like in practice?
Retail will get a trading interface that will have heavy coverage of all the assets they want to trade while maintaing the same UI/UX as a CEX (it’s virtually the same hardware). HFTs will get a microstructure that they’re more familiar with (colocation, real-time, price-time priority, no speed bumps). If we’re able to nail this economic supply chain — GTE will surpass all exchanges in volume over the long term, both on and off chain.
Trade fast, win faster.
GTE.
Hyperliquid has done a great job at attracting both 1) retail flow and 2) maker liquidity. One of the ways they have done this include building a high performance state-layer with modifications that enshrine an order book.
One example of these modifications is how Hyperliquid handles order priority. Because Hyperliquid has consensus, each transaction has to buffer in the mempool — usually around 3 blocks on average. Hyperliquid prioritizing cancels helps maker liquidity because makers who run validators are able to see incoming profitable taker transactions against their maker orders and subsequently cancel and then amend their orders.
This is what we call a speed bump: a way for less sophisticated market makers to comfortably quote tight on a CLOB without worry of being picked off by some of the major HFTs (toxic flow).
Jeff’s recent tweets on transparency of the L1 increasing maker liquidity for whales makes sense if you understand that there is a speed bump in the exchange.
However, none of the major exchanges where price discovery lives for any asset, both Binance for crypto, and Nasdaq / CME for equities and futures implement any sort of maker-friendly speed bumps.
The reason for this is that this would ultimately constrain their volume, and price discovery would move elsewhere. A majority of Binance’s volume is not Market Maker <> Retail trader, it’s Market Maker vs Market Maker. If you make it so that the weaker market maker has the ability to get out of a losing position — then that will greatly reduce the ability for a market maker with actual alphas to make winning trades.
Let’s simplify this a bit: Hyperliquid has around $10-20b of daily volume (on a good day). On that same day, Binance has around 10x that ($200b). For Hyperliquid to match and surpass that volume, it would have to remove its speed bump. Period. Hyperliquid doesn’t want to do this (for a plethora of reasons, one such being that HLP is enshrined so heavily into the maker side of the CLOB).
This begs the question: Why is HL onchain and transparent in the first place? With their speed bump architecture, the answer is not for superior liqudity. The answer is because there are certain benefits of decentralization which we shall not name that they look to take advantage of.
When you start to look at these pieces from a Birds Eye view, you see how effective the HL team is at marketing but ultimately why their design decisions will constrain their liquidity (and hence make the flippening impossible).
At GTE we’re taking an opinionated approach that if we want to compete, we’re going to have to play a different game than what exists right now. Can we take the best of tradfi execution (this is not privacy, but rather a true price-time priority clob) and the best of decentralization (composable, permissionless, censorship resistant).
We call this the barbell strategy — HL took the middle path by owning the entirety of the stack (exchange, l1, matching, oracle) and implementing things like speed bumps, GTE is taking the other side of this by prioritizing two major groups, 1) retail and 2) sophisticated HFTs.
What will this look like in practice?
Retail will get a trading interface that will have heavy coverage of all the assets they want to trade while maintaing the same UI/UX as a CEX (it’s virtually the same hardware). HFTs will get a microstructure that they’re more familiar with (colocation, real-time, price-time priority, no speed bumps). If we’re able to nail this economic supply chain — GTE will surpass all exchanges in volume over the long term, both on and off chain.
Trade fast, win faster.
GTE.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
