Kise

Researcher and analyst. Token Engener. Crypto Hunters
Quantitative researcher. Polymarket

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 22/ And one trap worth knowing.

The LIT ticker on Binance is NOT @Lighter_xyz. Different project. $0.74 vs $2.21.

Sum books by ticker without checking and Lighter's liquidity comes out inflated.

I measured on Gate, Bybit, KuCoin, MEXC. Verified via exchange metadata.

[Kise @Kise911_]

The #3 perp DEX refuses to take its own token as collateral.
The #1 and #2 don't refuse.
@Lighter_xyz accepts USDC, ETH, tokenized Robinhood stock. Not LIT.
@HyperliquidX and @Aster_DEX accept theirs.
One of these designs survives the next Oct 11. 🧵

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 21/ Caveats, because they matter.

No market caps for edgeX, GRVT, Drift — absolute depth only, no normalization.

Paradex, Extended, Backpack, Pacifica, Ostium: no exchange-listed token to measure.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 20/ That side: @Lighter_xyz @edgeX_exchange @grvt_io @dYdX @GMX_IO

The newest venues chose differently. That isn't an accident. It's the lesson of 2024 and 2025, priced into the architecture.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 19/ The takeaway.

The dividing line isn't size. Isn't volume. Isn't liquidity.

It's one design decision: does the protocol take its own token as collateral?

This side: @HyperliquidX @Aster_DEX @JupiterExchange @DriftProtocol

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 18/ Tier list, 2 of 2

🟡 HYPE — worst combo on paper. In practice the only one that survived −56% and held better than outside venues on Oct 10.

🟢 LIT, EDGEX, GRVT — token isn't collateral
🟢 DYDX, GMX — 89% and 100% float, dilution done

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 17/ Tier list, 1 of 2

🔴 JUP — own lending + own stable, untested
🔴 DRIFT — lending by design, no liquidity

🟠 ASTER — own token as margin
🟠 APEX — 15% float

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 16/ Size does not protect you.

HYPE is 22x bigger than LIT by market cap.

Relatively thinner book. Far more dependent on its own token as collateral.

Bigger is not sturdier.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 15/ Why float beats depth as a signal.

Low float means the supply is still coming. And it lands in the same thin book a liquidator has to dump collateral into.

@dYdX and @GMX_IO don't have that problem anymore. They already lived through it.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 14/ Here's what actually predicts blowups. It isn't book depth.

It's float.

HYPE 22%
APEX 15%
LIT 25%
ASTER 34%

vs @dYdX at 89% and @GMX_IO at 100%.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 13/ @DriftProtocol wires a lending market directly into the exchange. Spot borrow against perp collateral.

But book depth there is $31k.

Mostly an academic question.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 12/ That is the exact @CurveFinance architecture from 2024.

The one that liquidated the founder for $140M, dropped CRV 30% in two days, and left bad debt STILL uncleared as of April 2026.

Same shape. Younger. Untested.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 11/ Who's most exposed here? Not Hyperliquid.

@JupiterExchange.

Own lending protocol — Jupiter Lend, $873M.
Own stablecoin — JupUSD, accepted as collateral since January.

Own token. Own lending. Own stable.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 10/ @avantisfi has the most honest design in the set.

AVNT staked in the Security Module can be slashed up to 20% in a shortfall.

The token isn't collateral. It's a declared shock absorber.

You know you're paying for the failure. You don't discover it via liquidation.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 9/ @grvt_io went furthest.

First to wire external yield straight into trading collateral. ONE Balance: margin earns up to 11% via Aave while still backing positions.

Their collateral is external yield-bearing assets. Not their token.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 8/ @edgeX_exchange — same logic.

Collateral: native USDC, liquid staking tokens, stablecoins.

EDGE handles governance and validator delegation.

Not collateral.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 7/ So what does LIT actually do?

Governance, staking for LP access, fee discounts, buybacks, burns.

$17.9M bought back since launch, $14.6M of it in Q1 2026.
15.5M LIT burned — 6.3% of circulating.

Real money. Just not load-bearing collateral.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 6/ @Lighter_xyz: 6.4 bps.

Twice as deep, relative to size. At a $553M cap with 25% float.

A venue 3x smaller than Aster runs a book twice as deep — and refuses to build its collateral system on its own token.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 5/ Now the number that surprised me.

I summed every bid within 2% of mid across venues, then divided by market cap.

HYPE: 3.0 bps
ASTER: 3.0 bps

Hyperliquid is 7.6x larger. Identical relative book thinness.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 4/ @Aster_DEX — #2. ~20% share. $41.7B in 30 days.

ASTER cap: $1.61B. Float 34%.

ASTER is accepted as margin on Aster itself.

Collateral and traded asset: the same token again.

[Kise @Kise911_]

@Lighter_xyz @HyperliquidX @Aster_DEX 3/ @HyperliquidX — #1. Roughly a third of all perp DEX volume. ~$196B in 30 days.

HYPE cap: $12.3B.
Float: 22%. Lowest in the set.

And 83% of every deposit in HyperLend is HYPE in three different wrappers.