The Twin Test

The Curve reentrancy exploit has the largest market-adjusted 30-day move in this library: -31.9 points. The typical absolute event gap is 6.2 points, and 4 of 16 comparisons are rarer than 90% of their untouched controls.

This page asks a simple question: how much did a coin move after an event, once the move that similar coins made anyway is removed? It also shows when the comparison is thin, ordinary or dominated by one example.

ACTUAL
SIMILAR COINS
DIFFERENCE
remove the market before crediting the event
Events tested16Fixed 30-day window
Coins in the price panel307Potential comparison pool
Daily records in the panel2,818Historical depth
Largest market-adjusted gapThe Curve reentrancy exploit-31.9%
Typical absolute gap6.2%Middle event
Gaps of at least five points11 of 16After removing the market
Rarer than 90% of controls4 of 16Untouched-coin comparison
Positive event gaps4 of 16Coin beat its twin
Raw move and market same direction14 of 16Headline could confuse cause
Largest raw coin moveThe Mango Markets manipulation-55.9%
Largest market moveSolana's five-hour outage+37.0%
Most comparison coinsThe Curve reentrancy exploit90 coins
Most untouched controlsThe Curve reentrancy exploit90 coins
Typical comparison basket90 coinsMedian across events
Event types covered10Hacks, policy, upgrades and more
Largest type-level gapProtocol exploit-31.2%
First price date2019-01-01Panel coverage
Latest price date2026-09-18Panel coverage

Ask this page.

Choose a useful question or type your own. The answer and chart use only the figures already checked for this page.

Choose a question to build a chart.

Current viewWaiting for a question

The chart will appear here.

What changed after the market was removed?

The actual coin move and the move in similar unaffected coins stay side by side. Their difference is never presented as certain proof.

EventCoinTypeDateActual moveSimilar coinsDifferenceControls as extremeComparison coins
The Curve reentrancy exploitCurveProtocol exploit2023-07-30-34.0%-2.0%-31.9%3.3%90
The Mango Markets manipulationSolanaProtocol exploit2022-10-11-55.9%-24.8%-31.2%7.7%90
Ripple's partial court winXRPCourt ruling2023-07-13+34.1%+9.1%+25.1%8.8%90
The SEC sues BinanceBNBEnforcement action2023-06-05-20.5%+4.1%-24.6%3.3%90
The Radiant Capital hackRadiant CapitalProtocol exploit2024-10-16-27.7%-9.0%-18.7%11.0%90
FTX collapsesSolanaExchange failure2022-11-08-54.5%-36.4%-18.0%12.1%90
The Bybit hackEthereumExchange hack2025-02-21-27.7%-21.0%-6.6%54.9%90
Binance settles with the DOJBNBEnforcement action2023-11-21+2.8%+9.4%-6.5%74.7%90
Staked Ether unlockedEthereumProtocol upgrade2023-04-12-5.1%-10.9%+5.9%40.7%90
USDC breaks its pegEthereumStablecoin stress2023-03-11+30.3%+24.9%+5.4%62.6%90
The WazirX hackEthereumExchange hack2024-07-18-23.5%-18.1%-5.4%50.5%90
Mt. Gox starts repayingBitcoinSupply event2024-07-05+6.4%+1.8%+4.6%58.2%90
Spot Bitcoin ETFs approvedBitcoinRegulatory approval2024-01-10-1.8%+1.8%-3.6%76.9%90
Spot Ether ETFs clearedEthereumRegulatory approval2024-05-23-5.9%-4.7%-1.2%86.8%90
The MergeEthereumProtocol upgrade2022-09-15-20.9%-20.3%-0.6%95.6%90
Solana's five-hour outageSolanaNetwork failure2024-02-06+36.7%+37.0%-0.2%98.9%90

The Curve reentrancy exploit has the largest market-adjusted 30-day move in this library: -31.9 points. The typical absolute event gap is 6.2 points, and 4 of 16 comparisons are rarer than 90% of their untouched controls.

Does the library survive its largest example?

The most extreme event is removed before the typical result is read again.

The Curve reentrancy exploit31.9%
Typical event6.2%
Typical without the largest5.9%

Research by for Charlie Quant Lab · Updated