HYPE in the Assistance Fund system address of 0xfefefefefefefefefefefefefefefefefefefefe has been formally recognized as burned.
The governance vote was based on stake-weighted consensus, with 85% of stake voting for burning, 7% against, and 8% abstaining.
i think hl + lighter are the two winning clobs
hl is ultra efficient perp infra platform
-i think reading stratechery's platform piece is the best way to understand the strategy.
-https://t.co/1G0uXOPgBD
-at many points (hip3, mobile, lending, spot) hl has chosen to be somewhat credibly neutral, and allow others to build on top of their platform
lighter is robinhood playbook, retail growth
-vertically build many of the components like spot
-two tier fee system attracts real retail users, who then attract real organic mm makers
-real zk allows for more composability/future upgrades
many exchange market structures have been top two + others:
ftx/binance + others
binance/bybit + others
hl/lighter + others
when looking at the other landscape of clobs, i find it hard to get excited about any of them, as in my view the set space of retail needs (0 fee, liquidity, execution) etc is now covered by these two, and any competitors must fight for the same non-incentivized retail taker and mm liquidity, which will be especially difficult on tails
thoughts on zero fees perp dexs - commoditize the field for traders’ good
I don’t think a zero-fee model is a bad idea. Hyperliquid generated ≈$1B in fees this year, paid by traders and mms.
those fees were distributed to the Assistance Fund, meaning value was effectively transferred from users to token holders. In a zero-fee setup, that transfer doesn’t happen: traders keep more of what they make, and the venue itself captures less (or nothing).
the tradeoff is that the business model stops being inherently profitable from fees, and the system becomes closer to zero-sum (ignoring other frictions like funding, slippage)
a simple analogy is a casino. Each game has a “return to player” (RTP), often around 99%. That 1% gap is the house edge.
even if outcomes feel close to 50/50 in any single round, over time that edge accumulates and players lose on average.
with a true 0-fee perp DEX (and setting aside slippage), there’s no built-in house edge from trading fees. So the baseline becomes closer to 50/50: in the long run, a trader with no informational edge and a symmetric strategy should expect roughly flat results, rather than a slow bleed to fees.
recent comparisons show that some 0-fee perp DEXs that are running incentives have better execution than Hyperliquid.
and the main driver often isn’t “0 fees” by itself. It’s incentives.
incentives effectively turn trading from a zero-sum game into a positive-sum game: the more you trade, the more you earn (points). That extra payout can outweigh spread/slippage for many users.
why incentives can improve execution
retail flow increases because it is incentivized, traders are being subsidized via points. Therefore, more (and healthier) flow attracts MMs, who also want points.
in that environment, MMs may optimize less for spread PnL and more for incentive yield.
as a result, MMs can quote much tighter spreads, potentially even “negative” effective spreads relative to what they’d normally require (i.e., tighter than the maker fee an exchange charges), because:
- tighter quotes → more fills
- more fills → more volume
- more volume → more points/rewards
so incentives can directly buy tighter markets and lower slippage.
what happens when incentives end
once incentives stop, the positive-sum subsidy disappears. The market tends to revert to zero-sum, or even negative-sum if market makers are charged fees.
to remain profitable, MMs generally have to widen spreads to cover their full cost stack:
1) Maker fee (e.g., 0.002% on Lighter)
2) Operational costs (infra, latency, engineering, capital)
3) Inventory/price risk (adverse selection, volatility)
4) MM profit margin
anyway, I still think this is better than charging fees.
MMs face the same underlying cost structure everywhere: infra, inventory risk, and operational costs. The main variable the venue controls is fees/rebates
on Hyperliquid, maker fees are close to zero for large-volume accounts, and can even turn negative via rebates. That can encourage tighter quoting and reduce spreads.
on Lighter, the “cost” is effectively expressed through the **spread** (since makers still need to cover their costs), but the venue isn’t also extracting an explicit trading fee from users.
If Lighter and Hyperliquid had equal retail flow, my base case is that Lighter would show better execution for retail because there’s no additional taker fee layer on top of spreads, even though if Lighter would have wider spreads.
put differently: Hyperliquid making ≈$1B/year in fees is effectively a much larger “house edge” paid by users. Lighter (without points) keeps that money to traders, instead of themselves / token holders.
this small read is not a $lit bullpost, moreover it is in opposite way, as it shows that the fees would stay with traders, not the house. While on Hyperliquid, fees come from traders to the house (holders)
If you ask me on which exchange would I trade, I’d choose Lighter. If you ask which token would I hold, I’d choose $HYPE.
thoughts on retail trading - play at the tables that dont exist yet
(this is going to be a longer post as i have a lot of things i want to say)
i have been a dumb retail trader, i have been a sharp retail trader, i have worked at the peak of tardfi/kwant
at a high level you absolutely can be successful as a retail trader, but it comes down to find your edge + the discipline to stick to it
what is your edge as a retail trader?
time + curiosity + breadth + flexibility + reactiveness
you dont have more information than the mms/flow desks/algo shops. you will not be able to predict price on basically any timeframe better than them. that is dumb retail trading. you can argue for trading cycle tops/bottoms, but its harder to time than it looks. predicting price at any ltf is literally called "uninformed retail flow"
but you can spend all your time obsessing over one phenomenon/project/company, unlike the tardfi analyst who covers a sector. the more retail the product the better - the institutions have better supply chain data, they can call up the CEO and ask how hes feeling, but they are not using it, their friends aren't using it, etc
even finding it takes time/curiosity/flexible mandate, which someone who can't afford to waste time may not pursue
that is real edge you can wield as retail. in 2021/2022 i understood nft trading better than any institution. i would argue i understand perp dexes now as good/better than any institution. and there are people that understand memecoins much better than me.
but critically these are specific niches. my point is you cannot have edge on eth as retail - you can think tom lee is retarded, or everyone else is retarded, but you dont have more information than the aggregate market and at the end of the day you're just coinflipping
we are so lucky to live in an era where companies are bootstrapping by giving early adopters "equity". even just a few years ago, before airdrops became quite so big, my edge - "play with good new things" meant finding emerging markets with high onboarding friction and positioning early. now you can simply use platforms organically and be rewarded, you just have to pick the right ones.
that is your edge as retail, being your own best user, through personal experience understanding which platforms are good and which platforms are not (and it wont always look like perp dexes), and investing your time and energy early to position yourself for upside, whether thats earning points, buying the token, buying stock, etc.
play with the good new things
including an old thread i wrote about game selection