Introducing InfinityPools: Or how I learned to stop worrying and love leverage
Perpetual futures & bad leverage
Over the years, many crypto twitter influencers have repeatedly tried to convince their followers not to use leverage. Turns out, they were partially right. For a “correct” perpetual futures trade, the expected return sharply drops off after ~10x leverage. Above 15x leverage, traders can expect to lose money, even when they are directionally accurate.
Despite this, the truth is that leverage is necessary. Investing is only worth a person’s time if the return upside is high enough. Without leverage, for most retail users, the only investments that could potentially offer high enough returns to justify time spent are volatile small market cap assets (ie. shitcoins). This limits investment opportunities and the types of ventures that can receive funding.
Unfortunately, most forms of leverage currently available are constrained and often lead to undesirable outcomes (eg. liquidations on short term, large price moves). Think of the numerous funds that lost money on their Luna shorts when it was death spiraling because they got liquidated on a small bounce in price. Using perpetual futures today is a Faustian bargain — you gain the capital requirements to trade with size, but offer up something just as important in exchange.
Liquidations as the root of all problems
In order to understand why perpetual futures are limited in their capabilities and can cause undesirable outcomes for its users, we first need to understand their risk constraints.
Perpetual futures markets are only functional as long as the liquidation of traders is being done efficiently enough. Of course, this isn’t easy, and as a result, liquidators need an operating margin (usually under the form of a ~1–3% liquidation penalty for the user) to ensure the underlying exchange does not incur any bad debt. However, even in very liquid markets sometimes the liquidation penalty isn’t big enough: as users increase their notional trade size, the leverage offered by exchanges gets reduced.
Liquidating positions for liquid assets like BTC and ETH is already hard enough, so what happens when one has to liquidate a position for a volatile, illiquid asset? The exchange must either increase the liquidation penalty size or it needs to put a tighter cap on trade size. Neither solution is particularly attractive.
If these measures fail (and they do) and an exchange accumulates enough bad debt on a given perpetual futures market, it will have a forced settlement event. Forced settlement means, amongst other things, that some traders may not get paid out on their profitable position.
To recap, to mitigate the risks involved in liquidating positions, exchanges impose hefty penalties on their users, limit the leverage & assets available, and have significant counterparty risk via the forced settlement of markets. Unfortunately, in some cases these problems are exacerbated, as one of the main sources of income for exchanges comes from liquidations. As anyone that works with blockchains knows, stories with unaligned incentives never have happy endings.
InfinityPools & good leverage
What if you could somehow guarantee that all positions on your protocol closed perfectly? The problems listed above would no longer be applicable. Traders would be able to have unlimited leverage for any asset and even avoid liquidations altogether. All without any counterparty risk.
This is the breakthrough InfinityPools has achieved. InfinityPools is the first protocol to make “good” leverage accessible to everyone. Thorough backtesting supports this claim and as you can see below, expected returns for traders using leverage are improved by orders of magnitude, especially as they get towards higher multiples.
High level overview
InfinityPools is a two sided marketplace, with traders on one side and liquidity providers on the other. InfinityPools’ liquidity provisioning is similar to lending out Uniswap v3 tokens, but with no credit risk.
InfinityPools liquidity providers can either deposit assets directly into the protocol or can lend out their Uniswap v3 LP tokens. The yield generated is superior (above the gains from delta hedging, while AMMs typically pay less than delta hedging), continuous (you generate yield even when your liquidity is not in range) and JIT proof.
Let’s imagine a scenario where a ETH/USDC pool exists and the current market price of ETH is 1000 USDC. A liquidity provider comes in and provides 1000 USDC worth of liquidity in a tight liquidity range centered at 900 USDC. They get back an LP token representing that liquidity range.
A trader, wanting to go long ETH with leverage, could then borrow that LP token and redeem it for the underlying 1000 USDC. The trader could then swap the 1000 USDC for 1 ETH, minus fees and slippage (using a spot DEX of their choice). There are now three possible scenarios:
- If the price of ETH goes up, the trader makes money and can swap back a part of the 1 ETH for 1000 USDC to pay back the LP token.
- If the price of ETH goes down to right above 900 USDC, the 1 ETH the trader holds is worth 900 USDC and the liquidity provider is expecting back an LP token worth 1000 USDC. The trader therefore needs to start out with 100 USDC (or ~0.11 ETH) in collateral in order to pay back the full LP token.
- If the price of ETH goes below 900 USDC, the liquidity provider is expecting back an LP token worth 1.11 ETH. The trader already holds 1 ETH and the remaining 0.11 ETH is worth, at most, ~100 USDC. Therefore, same as with scenario B, the trader needs to start out with 100 USDC or 0.11 ETH in collateral in order to pay back the full LP token.
The minimum collateral requirement for a trader borrowing an LP token worth 1000 USDC at the 900 USDC liquidity range is therefore ~100 USDC (~10x leverage). It follows that the closer the liquidity range borrowed is to the current market price of ETH, the more leverage it enables. For example, if the liquidity range borrowed was deployed at 999 USDC, the maximum loss / initial collateral required would be 1 USDC, which corresponds to 1000x leverage.
Of course, traders pay an ongoing “funding rate” to liquidity providers in exchange for borrowing their LP tokens. As long as this rate is paid, traders can keep their position open, no matter what the price of the asset is.
The full whitepaper is available on our Discord. Join to learn more!
Testnet out late February / early March. Follow us on Twitter, and sign up on our website for updates.
Data for charts: closed form solution of expected return for each derivative trade. The price of the underlying asset follows a geometric Brownian motion with drift (resulting in a Sharpe ratio of 1).
