Understanding perpetual futures

Introduction to perpetual futures as a crypto financial innovation that serves casino-like exchanges. Perps represent 6-8x the volume of spot trading and primarily exist to collateralize positions. The core problem they solve is reducing capital requirements for exchanges that need to maintain trust while running profitable betting operations.

• •$300B in stablecoins outstanding, with ~25% on exchanges primarily collateralizing perp positions
• •Perp volume is 6-8x larger than spot market trading, similar to traditional derivatives markets
• •Exchanges are capital-intensive due to trust requirements, not technology - they need assets exceeding customer balances
• •Perps reduce total capital requirements shared between exchanges and market makers while maintaining gambling functionality

Deep dive into how crypto exchanges trade against their own customers through affiliated entities like Merit Peak (Binance) and Alameda Research (FTX). Explains historical front-running practices and why this behavior persists in crypto despite being discouraged in traditional finance.

• •Merit Peak Limited is Binance's captive trading entity, also accused of money laundering by US agencies
• •Alameda Research (FTX's affiliate) was criminally convicted of money laundering and fraud
• •Front-running historically meant brokers choosing post-facto whether to assign profitable trades to themselves or customers
• •Billions in customer funds can be disguised to banks as 'proprietary trading' and 'OTC desk' operations

Technical explanation of perp mechanics, including multiple daily settlements via funding rates. Covers how Robert Shiller's 1992-93 proposal for perpetual cash-settled futures found its home in crypto through BitMEX's implementation.

• •Perps settle multiple times per day (e.g., every 4 hours) instead of having specific expiration dates
• •Winners get paid by losers at each settlement through the funding rate mechanism
• •Example: 0.1 BTC perp at $86k rises to $90k = $400 profit credited, position continues automatically
• •Funding rates include embedded interest (e.g., 3 basis points/day), but at 10x leverage that's actually 30 bps/day
• •Some exchanges cap funding rates per settlement period to dampen volatility, unlike traditional circuit breakers

Explains how sophisticated market makers profit from price divergences between perps and spot markets through delta-neutral basis trades. This mechanism keeps perp prices anchored to reality while providing negative carry income to market makers.

• •When markets are exuberant, perps trade above spot - market makers short perp and buy spot in equal size
• •Basis trade is delta-neutral: no exposure to Bitcoin's fundamentals, just collecting the spread
• •Market makers get paid via funding rates while waiting for convergence (negative carry)
• •Perp prices can diverge 10%+ from spot before gamblers stop participating
• •Critical risk: Market makers must keep collateral on exchanges, which becomes bankruptcy claims if exchange fails

Why market makers care intensely about capital efficiency when operating across multiple crypto exchanges. Includes the cautionary tale of Galos Capital losing 40% of assets on FTX.

• •Galos Capital had 40% of assets on FTX, wound down fund selling bankruptcy claim for 16¢ on dollar
• •Market makers function across many exchanges - fully collateralizing all venues would be prohibitively expensive
• •Most crypto exchange trades require immediate cash settlement, unlike traditional T+1 or T+2 settlement
• •Capital efficiency is critical because missing profitable opportunities costs money, but pre-deploying everywhere is impossible

Contrasts traditional finance's Regulation T (2x initial, 4x maintenance margin) with crypto's extreme leverage offerings of 20x, 50x, 100x+. Details how exchanges profit from liquidating overleveraged traders and the role of insurance funds.

• •Traditional US brokerages limited to 2x leverage at opening, 4x maintenance by Regulation T (nearly 100 years old)
• •Crypto routinely offers 20x, 50x, 100x+ leverage - FTX 'responsibly' self-limited to 20x
• •Liquidation fees (e.g., 0.5% at Binance) calculated from amount at risk, can be large portion of losses
• •If liquidation fee exceeds negative balance, exchange/insurance fund keeps the difference as profit
• •Insurance funds pay winners when liquidations go negative, but only at management's discretion

Explains the shocking mechanism of automatic deleveraging where exchanges retroactively reduce winning positions when losers can't pay. This is buried in contracts and particularly devastating for basis traders who thought they were hedged.

• •ADL activates when insurance fund is insufficient - exchanges automatically reduce winners' positions retroactively
• •Theory: ADL hits highly leveraged, high-profit accounts first. Practice: opacity around actual implementation
• •Example: You thought you were 4x leveraged and earned 80%, but were retroactively only 2x, so you earned 40%
• •Devastating for basis traders: 100 units short perp + 100 long spot = zero delta, but ADL to 50 short = +50 delta exposure
• •October 2025: $19B+ in liquidations with widespread ADLing, usually happens on sharp price decreases not increases

Real examples of perp market failures including Alameda's $100M+ loss during Terra collapse and their role as FTX's backstop liquidity provider using customer funds.

• •Alameda lost $100M+ during Terra stablecoin collapse in 2022 as backstop liquidity provider
• •FTX's insurance fund was 'made up by a random number generator' per testimony
• •Alameda took up liquidated positions - lucrative most of the time, catastrophic during Terra blowup
• •Alameda paid winners with 'borrowed' customer assets, leading to criminal convictions
• •Contrast with traditional markets: margin calls give days to post collateral, and negative balances must be paid

Compares traditional finance's layered risk waterfall (brokerage capital, clearing pools, guarantee funds, mutualized insurance) with crypto's approach. Explains why traditional finance won't adopt perps despite crypto advocates' expectations.

• •Traditional waterfall: brokerage reserves → clearing pools → guarantee funds → mutualized insurance → loss allocation
• •E*TRADE nearly went bankrupt during 2008 GFC from being long subprime mortgages, not from customer margin
• •Traditional finance also not fully capitalized against all events, but relies on underwriting participant risk
• •Perps won't be adopted in traditional markets: existing derivatives work well, liquidations aren't the business model
• •Learning you're either hedged or bankrupt on a 20% down day doesn't appeal to traditional finance professionals