Every Market Crash Liquidates Long Positions Worth Billions. Those Losses Are Our Gains.

We have a proven business plan that is already profitable: With our funds combined, we short excessively-hyped coins on Binance and other exchanges, forcing over-leveraged traders into liquidation. Each liquidation pushes prices further down, in turn causing more liquidations in a cascade that results in an avalanche of profits for our token-holders. In a perfect world, where market liquidity was infinite and futures markets were perfectly insulated from spot markets, the Liquidator DAO’s strategy would not work.

But this is not a perfect world.

The value of the Liquidator token does not depend on marketing, celebrity endorsements, or banner ads. We don’t rely on trendy NFTs, Reddit spam, forced memes, or aggressive Telegram bots with poor grammar. Instead, we harvest value from futures liquidations, using the same business model that makes Binance and its ilk so tremendously profitable. This value is extracted in the form of blue-chip cryptocurrencies (BTC, ETH, BNB, and USDC), which we then use to repurchase Liquidator tokens.

Liquidator is fundamentally different from other tokens in two important ways:

(1) Our token’s value rises without requiring others to purchase it;
(2) As liquidations tend to increase in number and value during broad downturns in the cryptocurrency market, our token serves as a hedge against market crashes and generates profit during times of widespread decline.
(3) Liquidator operates as a DAO, offering token holders unprecedented security and control over the project.

Read our whitepaper for more information. We do the math.

How To Trade Liquidator Token

Step 1: Install MetaMask

MetaMask is an open-source, industry-standard wallet suitable for trading Liquidator and other tokens on Decentralized Finance (DeFi) platforms. No other wallet has the same degree of independence, and MetaMask is continually audited by independent researchers.

MetaMask is a browser extension and is compatible with Chrome, FireFox, Brave, and Microsoft Edge. It operates on Desktop, Mobile, and Tablet platforms.

Be certain to back up your seed phrase in order to secure your Liquidator tokens and any other virtual assets you may store on MetaMask.


Step 2: Configure MetaMask for Binance Smart Chain (BSC)

By default, MetaMask operates on the Ethereum main network (“Beacon Chain”). In order to trade Liquidator tokens, you will need to adjust its configuration to support the Binance Smart Chain (BSC). Once so configured, you can easily switch between Ethereum and BSC, and securely store and trade assets on both networks. Follow these steps:

A. Open MetaMask, and select “Settings” >> “Add New Network”.
B. Set Network Name to “Binance Smart Chain”.
C. Set RPC URL to “https://bsc-dataseed.binance.org/”.
D. Set Chain ID to “56” (in base-10) or “0x38” (in hexadecimal or base-16).
E. Set Symbol to “BNB”.
F. Set Block Explorer to “https://bscscan.com”.

Once these steps are completed, select “Save”. You can now switch to the Binance Smart Chain in the drop-down menu at the top of MetaMask’s main screen. You may also view instructions on Binance’s website:


Step 3: Obtain BNB (Binance Coin)

Binance Coin (hereafter referred to as “BNB”) is issued by Binance, by far the world’s largest cryptocurrency exchange. It is also the native asset of Binance Smart Chain, serving the same functionality on BSC as Ethereum does on the Ethereum Virtual Machine (EVM). With a market capitalization on the order of $100 billion, BNB is a blue-chip cryptocurrency by any measure.

Liquidator tokens are priced in terms of BNB, and therefore BNB is required to purchase such tokens. You can purchase BNB directly on many exchanges. Alternatively, you may purchase other cryptocurrencies (such as Bitcoin or Ethereum) on exchanges such as CoinBase, send them to BNB-supporting exchanges, and purchase BNB.

Once you have BNB, transfer it to your MetaMask wallet. You are now ready to exchange these BNB for Liquidator tokens, and vice versa.


Step 4: Trade BNB for LIQUIDATOR on PancakeSwap

MetaMask is an open-source, industry-standard wallet suitable for trading Liquidator and other tokens on Decentralized Finance (DeFi) platforms. No other wallet has the same degree of independence, and MetaMask is continually audited by independent researchers.

MetaMask is a browser extension and is compatible with Chrome, FireFox, Brave, and Microsoft Edge. It operates on Desktop, Mobile, and Tablet platforms.

Be certain to back up your seed phrase in order to secure your Liquidator tokens and any other virtual assets you may store on MetaMask.

Tokenomics & Locks

Total Liquidator Token Supply:
1,000,000 [One Million]

Total Reserved Tokens:
250,000 [25%]

– for Developers: 50,000 [5%]
– for Moderators: 10,000 [1%]
– for Liquidity Providers: 100,000 [10%]
– for CEX Listing Fund: 50,000 [5%]
– for Marketing Fund: 40,000 [4%]

Available on DeFi: 750,000 [75%]

The Liquidator project uses straightforward tokenomics, without reflection, “taxes”, or other gimmicky features. These tokenomics are the product of careful consideration. To learn why, please read our mini-FAQ below:


Question 1:

“Why is the supply of Liquidator token only one million? Why didn’t you mint a quadrillion billion zillion Brazilian tokens? Even in 2022, the concept of ‘market cap’ is not widely understood, and buyers will perceive a token as ‘cheaper’ if they can trade a dollar for a number of tokens so large that expressing it requires scientific notation. Don’t you know that a high token supply encourages buying?”

Answer 1:

We are well-aware of the trend of projects minting more tokens than atoms in the observable Universe. The Liquidator team doesn’t follow trends, however. We are a fundamentally different project, and we set ourselves apart from the memecoins and moonshots in many ways – one of which is avoiding the gimmick of minting absurdly large numbers of tokens. In terms of buyers, we believe in attracting quality over quantity, and the sort of person who buys Liquidator tokens isn’t going to be fooled by high token supply. There are also three major practical reasons for keeping the number of tokens sane.

First, high token supply means that users will be dealing with enormous numbers of tokens in their wallets, and typing out long strings of zeroes is time-consuming and error-prone, especially in the heat of the moment when trading volumes are high and every second counts. Likewise, minting quintillions of tokens ensures that charts and price quotes will forever be a nightmare to read. There is a reason why petrol stations don’t list the price of gasoline in thousandths of a cent per tablespoon; there is a reason why grocery stores don’t list the price of rice in millionths of a dollar per rice grain. Serious traders know that their time is valuable, and we refuse to make them squint at charts because they’re uncertain if there are thirteen or fourteen zeroes after the decimal point on the Y-axis. Long strings of identical digits are difficult to read, and this can result in errors. There is, in fact, a biological analogy to this: even DNA replicase enzymes tend to make mistakes when the same nucleotide is repeated hundreds of times in a row.

Second, inflated token counts scare away serious traders. Not all shitcoins have high supply, and not all tokens with high supply are shitcoins, but the correlation is undeniable. Avoiding this practice is a long-term investment in the reputation and respectability of the Liquidator project.

Third, and perhaps most importantly, the price of a token influences what pair(s) it is listed against on centralized exchanges. Due to technical constraints, exchanges typically will not list tokens with very low value against major stablecoins (USDC, USDT, etc.). Instead, they will insist on listing them only against proprietary stablecoins (ex. one-cent stablecoins that are exchange-specific), and we want to avoid this for obvious reasons.


Question 2:

“Why isn’t the Liquidator token reflective? That is – why don’t you have a mechanism in place to discourage selling by ‘taxing’ transactions and sending a portion of each transaction to a burn address, or to a marketing wallet, or to a redistribution wallet, or to the liquidity pool, or to Vitalik?”

Answer 2:

The decision to avoid reflection, automatic liquidity, and other “gimmicky” tokenomics is based on the combined experience of Liquidator’s developers and (equally important) our moderator team, the latter of whose members has spent years “in the trenches” of Telegram and Discord channels for other crypto projects.

Firstly, reflective tokenomics send a clear message that the token’s only value (or, at least, a substantial part of its value) is derived from an ever-increasing flow of new buyers. In a sense, reflection is the programmatic and mathematical incarnation of a Ponzi scheme: newer purchasers “pay off” earlier purchasers, and when the influx of new buyers ends, the house of cards comes falling down. Liquidator is different: we create value for holders by profiting from liquidations on major exchanges, and we want to make that crystal clear.

Secondly, complex tokenomics ensure that a project’s Telegram channel will be perpetually full of confused users who are panicking because their transactions haven’t gone through due to insufficient slippage. This has the potential to disrupt the conversation around a token, and wastes moderation and community-management resources. Ask any moderator of any cryptocurrency project with reflection, and you’ll hear the same thing: no matter how many messages are pinned, and no matter how many times the need for slippage is emphasized, someone will still be confused. By avoiding such features, the Liquidator project ensures that trading in our token is simple, straightforward, and user-friendly.

Thirdly, reflection and similar features are a major impediment to Centralized Exchange (“CEX”) listings. Representatives of the Liquidator project have been in contact with centralized exchanges since before our token contract was even written, and we consistently hear that exchanges strongly prefer to list tokens without these features. The reason for this is simple: CEX users tend to become displeased when reflection mechanisms cause tokens to “disappear” during withdrawals, and these upset users will complain to their exchanges about this, even though tokenomics are of course outside of an exchange’s control. From the standpoint of an exchange operator, reflective tokens present a high reputational and customer-service risk. This doesn’t mean that centralized exchanges will never list assets with complex tokenomics – certain such tokens have been listed. However, exchanges demand far higher established trading volume and much larger liquidity pools (not to mention much higher listing fees) when dealing with reflectionary tokens. The simple tokenomics of Liquidator’s crypto-asset allow us to list on more exchanges, sooner, and with less cost than would otherwise be possible.

Telegram Channel

TEXT

.

Liquidator Whitepaper

I. ABSTRACT:

“Liquidate” is a curious verb.

In one sense, liquidation describes the forced closure of a financial position. Consider a Binance trader: he might choose to open a leveraged long position in futures, effectively borrowing money to purchase more units of a cryptocurrency than he could otherwise afford. These tokens are in fact purchased on his behalf by Binance. If the tokens rise in price, he will reap profits far in excess of those he would earn from conventional (i.e., unleveraged) trading. However, if the tokens fall in price to the point where his financial losses are poised to exceed the value of his collateral, Binance will close his position at a loss – typically, at a total loss. We call this “liquidation” because the tokens representing his position are sold in exchange for some liquid asset, typically a fiat currency or a stablecoin representing the same.

In another sense, when the core of the Chernobyl Nuclear Power Plant detonated, the “Liquidators” (Ликвидаторы) were the men who labored in perilous conditions and gave their lives to contain the catastrophic radiation leaks consequent to the blast.

We see one obvious parallel between these uses of the word “liquidate”: in both cases, liquidation is a sort of “cleaning up”, and a way of resolving a dangerous situation. In the financial case, liquidation involves an exchange protecting itself from financial losses by eliminating the risk of “holding a customer’s bags”, i.e., the risk of absorbing the losses of a customer’s unprofitable leveraged Futures position. In the radiological case, liquidation involves a nation protecting its civilians from irradiation by eliminating the risk of further isotope leaks.

There exists a second parallel – a deeper, more meaningful parallel. The explosion at Chernobyl was, at its core (both metaphorically and literally), caused by a chain reaction. This reaction started small: a single neutron collided with an atom of Uranium-235, releasing energy and splitting the destabilized nucleus into fission products, among which were three additional neutrons. Each of these three neutrons split another atom of U-235, and one neutron became three, then nine, then twenty-seven, and so-on. The core had reached criticality, and the rest, of course, is history. Binance, like Chernobyl, is susceptible to chain reactions. Just as a nuclear meltdown can begin with a single neutron, so can a market meltdown begin with a single liquidation: as an insolvent trader is liquidated, Binance must sell off his cryptocurrency holdings before they drop in value further. However, this sell-off itself decreases the price of the cryptocurrency in question, and this price movement can force other traders with leveraged long positions into liquidation. Further liquidations cause additional selling by Binance, which causes further price drops, which in turn cause ever-more liquidations. This process can result in billions of dollars in liquidations, as occurred on April 21st, 2021 (over $5B liquidated) and on May 17th, 2021 (over $2B liquidated).

Money lost in liquidations disappears from the accounts of reckless Futures traders, but it does not disappear in an absolute sense; rather, it accrues to market-makers, holders of short positions, and to an exchange’s “insurance fund”. Money on an exchange, like mass-energy in an atomic nucleus, is conserved. One man’s short position is another man’s long position; one man’s loss is another man’s gain.

Liquidation cascades are hazardous, but just as Man may someday harness the power of atomic chain reactions to conquer the stars, so we at Liquidator.Finance harness the power of liquidation “chain reactions” to extract value from cryptocurrency exchanges and their customers. If a particular trading pair has large amounts of positive leverage (i.e., the dollar-values and leverage factors of long positions are large relative to total liquidity), it is “fissile”: Just like highly-enriched uranium, it is unstable and can be provoked into undergoing a violent chain reaction.

When Liquidator.Finance identifies such currency pairs, we use a portion of the liquidity of our token as collateral to open massive short Futures positions, causing Binance or other exchanges to sell tokens on our behalf, thereby increasing effective supply and dropping prices. As prices drop, traders with long positions are liquidated en masse, each liquidation triggering further price drops and ever-more liquidations, until all or nearly all long positions have been liquidated. The value of these liquidations – potentially in the millions of dollars – is transferred to our short position. We then close the shorts and use the profits both to add liquidity to the Liquidator token (increasing price stability) and to purchase our own tokens at market value (boosting prices). As the value of the Liquidator token and its liquidity pools grows, we take on ever-more-ambitious projects.

Here, we speak primarily of long-position Futures liquidations. However, short positions can also be liquidated. This is rarer, given that most Binance users are broadly bullish on cryptocurrencies (if they weren’t broadly bullish, they likely wouldn’t be on Binance); hence long positions tend to be dominant over shorts for most asset pairs at most times, and squeezing long positions is often easier and more profitable than squeezing short positions. Regardless, the fact that any leveraged position can be liquidated for fun & profit allows Liquidator.Finance to generate revenue for token-holders no matter the market conditions.

The largest cryptocurrency whales use the same strategy we employ, forcing smaller traders into liquidation, consuming their margin accounts like so many krill filtered from the twenty-thousand-league depths of the Seven Seas. The average trader stands little chance against these whales on his own. Individually, we may be minnows, but through Liquidator.Finance, it is we who become the whale.

You wake up in the morning, reach for your phone, and see the headline:

“Crypto Markets Crash In Worst Sell-Off On Record – $20 Billion In Futures Liquidated In Less Than An Hour”

Will you cry into your morning coffee? Or, will you smile, knowing that you’re on the other side of these liquidations, and that these losses are your gains?


II. SYMBOLISM:

The Liquidator.Finance project includes Atomic Age symbolism from the USSR in honor of the sacrifices made by the Chernobyl liquidators. As the Liquidators served the Soviet Union, so we serve the purchasers of our tokens.


III. THE ECONOMICS OF FUTURES MARKETS

Futures are a subset of financial options, and to understand them, we begin with a treatment of options more generally.

An options contract represents the right to buy or sell some financial asset or commodity in the future at a pre-arranged price. This price, known as the strike price, is agreed upon at the time the contract is bought. The moment an options contract is bought (by a trader) is also the moment when it is sold (by another trader, market-maker, exchange, or other counterparty). Critically, the strike price is intrinsic to the contract and will not fluctuate with market conditions, having the potential to be greater (or less) than the free-market price of the underlying asset at any given time. It is the ability to buy (or sell) units of an underlying asset at a price other than market price that gives options their potential financial value.

An enormous variety of options are available, and this variety is not exclusively due to the range of underlying assets referenced. Even for a single asset (such as Bitcoin, or platinum, or shares of AMC), a wide range of contracts exist, defined in part by the following characteristics:

1. Polarity (a right-to-buy or “long” contract versus a right-to-sell or “short” contract);

2. Strike Price (the price at which one may sell or buy);

3. Expiration (options may expire at any pre-defined time after they are purchased, or alternatively, they may be “perpetual”, i.e. non-expiring, which is equivalent to having an expiration date infinitely far in the future);

4. Exercise Rights (certain contracts may be exercised at any time after purchase, while others may be exercised only at or close to their expiration dates – the former are often called “European-style” and the latter “American-style”, though in reality both types are widely available in all markets);

5. Obligation to Exercise (conventionally, exercising an options contract is optional (hence the term “option”); in some cases, however, a purchaser of a contract is financially obligated to exercise it, and this obligation is secured by collateral in order to force exercise even if at a loss);

6. Counterparty (in some cases, an exchange is merely a market-marker and doesn’t directly sell options, rather allowing its own users to mint and secure and sell the contracts bought by others; in some cases, an exchange is the counterparty for some or all contracts).

This whitepaper is not a primer on financial instruments generally. We will not be presenting a comprehensive review of every nuance of every iteration of every options contract available in every nation and in every epoch since the 17th century Dutch Golden Age. Rather, we will focus on futures contracts offered by Binance, which are typical of futures contracts in the cryptocurrency space. Importantly, cryptocurrency futures behave differently from “normal” stock market options in several respects. RobinHood and Binance may have significant overlap in terms of user base, but the financial instruments they offer are dissimilar in critical aspects. Binance’s futures contracts – which are typical of cryptocurrency futures overall – have the following characteristics:

1. The strike price is equal to the prevailing market price (also called the “spot price”) of the asset in question at the instant a futures counteract is purchased. For example, if the price of BTC is $69,420 at the moment a future is opened, then this will be the strike price for that future – regardless of whether it is long or short. The situation becomes more complicated in that opening futures contracts (and closing them, for that matter) itself influences the market price of an asset if the value of the futures is sufficiently large relative to liquidity in the spot market. Large orders may therefore be filled at a range of prices, and the contract may be treated as having a strike price at the share-weighted average price of purchased contracts. The net result is that large long futures contracts, when opened, will tend to have strike prices slightly above the market price before the futures were opened. Large short futures contracts, in turn, will have strike prices slightly below the pre-opening market price.

2. Cryptocurrency futures are perpetual; they do not expire.

3. In order to keep futures prices close to the actual underlying price (spot price; market price) of an asset, users holding futures contracts in the “prevailing” direction (that is, in the direction corresponding to the majority of futures) are charged periodic fees, which are paid to users holding contracts in the opposite direction. For example, if the majority (by value) of Ethereum futures are long, then users holding long futures will be periodically charged a percentage of the notional value of their positions, with this money paid to users holding short positions, who will receive money proportionate to the notional value of their own shorts. Fees are charged from (or paid to) the margin balances of the traders in question. The fact that fees are based on notional value (and not on collateral value) means that, all else equal, higher leverage against a fixed collateral will result in higher fees deducted or received. For example, if a user holds short Doge positions and the majority of Doge futures are long, and this short-position holder increases his leverage by a factor of ten, the fees he receives will also be multiplied by a factor of ten.

4. In every case, the counterparties of cryptocurrency futures are exchanges (i.e., all cryptocurrency futures contracts available on Binance are sold by Binance itself, not between users).

5. Cryptocurrency futures may be “exercised” (closed) at any time.

6. Cryptocurrency futures must be closed at some point. Any trader opening a Futures contract is obligated to close that contract, and will not be permitted to withdraw his collateral until all contracts secured by that collateral have been closed.

7. Given that exchanges are the counterparties of cryptocurrency futures, exchanges must have in place some mechanism to liquidate the losing positions of traders. That is, if a trader’s position moves “into the red” and the unrealized losses on that position approach the same magnitude as the trader’s collateral, the exchange must liquidate that trader immediately, least the trader’s losses exceed his collateral (in which case, the exchange inherits its customer’s losses – never a good thing from an exchange’s perspective).

Based on these principles, we can begin a mathematical treatment of futures contracts. Let us first express the relationship between notional value, leverage, and position size:

These relationships are intuitively reasonable. For example, we see that increases in either margin or leverage will result in the increase in the notional value of a position (which in turn amplifies, in an absolute sense, the financial impact of a particular futures investment, all else being equal). We also see that increasing the notional value of a position or decreasing the margin will increase leverage: the more a man borrows (or the less collateral he puts up), the higher his leverage.

Now, let’s generalize these equations so that we can analyze cases in which a trader has multiple positions against a single margin pool. This will allow us to simulate the positions of traders who use Binance’s “cross-margin” feature, or equivalent features on other exchanges. The simulation of split-margin traders is much simpler, as we can treat each position independently. For cross-margined cases, we write:

We consider n to be the total number of positions held by a trader, and we count positions from i=0, i=1, i=2, etc. until we reach i=n.

Measuring leverage is useful, but this information alone will not tell us when a trader will be liquidated. Even a trader with infinitely-high leverage is in theory no risk to an exchange as long as his open futures are in net profit. Therefore, in the next section of this whitepaper, we will add the opening and current prices of futures positions to our calculations.


IV. LIQUIDATION IN FUTURES MARKETS

As cryptocurrency traders, we are accustomed to thinking about futures markets from the perspective of a trader. It is natural to contemplate a topic from one’s own perspective: you probably don’t think much about economic development from the standpoint of a Tuvaluan, as you don’t live in the tiny island-nation of Tuvalu; you probably don’t think much about cryptocurrency from the perspective of an exchange operator, as you don’t run a cryptocurrency exchange. If, by chance, you live in Tuvalu or operate a cryptocurrency exchange (or both), feel free to congratulate yourself: you are the exception.

As Sun Tzu advised, whenever you are involved in a financial transaction, you ought to walk a kilometer in your counterparty’s shoes. Imagine you run a cryptocurrency exchange and offer leveraged trades: you don’t want to end up rekt when your users take on reckless positions (which some of them inevitably will), so you develop a system to liquidate their positions when prices move against them. Your first step is to calculate a trader’s instantaneous profit or loss (which we will call “return”) at any time t, as a function of the notional value of their position, the direction of the position (long or short), and the delta between the entry price and current price of the traded asset:

If this return-on-position ever falls below margin, then that position is “underwater”: closing it will result in a loss greater than the trader’s collateral, meaning that the exchange cannot confiscate enough money from the trader to make up the loss. An exchange in this position is between a rock and a hard place: it can close the position at a loss and pay off a portion of a user’s bad debt, or it can hold the position and hope it will enter profit, with the risk that the losses will increase over time if past price trends continue. This latter choice is particularly perilous in the case of short futures positions: asset prices cannot fall below zero but can rise without limit, so a short futures position can sink infinitely deep underwater, incurring infinite loss.

To prevent these sorts of situations, exchanges enforce maintenance-margin values. We will refer to them hereafter as “safety margins”, to avoid confusion with other uses of the term “margin”. A safety margin is the degree to which a position’s collateral exceeds its instantaneous profit (or loss). Let’s begin by expressing safety margin as a dollar-value:

We can also express safety margin as a percentage of nominal position value:

Generally, exchanges set larger safety-margin percentage requirements for larger fiat-value positions. This is because, the larger a position, the more difficult it is to liquidate (all else equal). Further, larger positions will result in larger absolute losses if they are not closed before going underwater. Similarly, higher safety-margin requirements are enforced for positions that involve thinly-traded assets, since less liquidity is available. For example, for BTC-USD futures, Binance requires that a $250k position be collateralized at 1% (i.e., $2,500 of minimum collateral), whereas a $250M position must be collateralized at 25% (i.e., $62.5M minimum collateral). BTC-USD is the highest-volume futures pair on Binance and most other exchanges, and so it has the loosest margin requirements of all pairs. Let’s examine safety-margin requirements for certain other pairs, assuming a $100k nominal position in each case:

1. BTC-USD (by far the largest pair on Binance and most centralized exchanges): 0.5%, or $500

2. ETH-USD (a slightly-smaller pair; Ethereum is closer to Bitcoin in trading volume than in market capitalization): 0.65%, or $650

3. LINK-USD (the Chainlink oracle token is not quite a blue-chip cryptocurrency, but still a major player): 2%, or $2,000

4. BAT-USD (the Basic Attention Token is one of the smaller currencies for which Binance offers futures): 5%, or $5,000

Comparing the extremes is helpful: Binance will allow a trader to come within $5k of being underwater on a $100k position in BAT, but that same trader is allowed to come within $500 of being underwater for an identically-sized position in BTC. Binance provides tables listing safety margins for four categories of cryptographic assets (BTC, ETH, select large-caps, and select small-caps); let’s present them here, as we will need them for calculations later:

1. Bitcoin-USD Pair (supports both Tether/USDT and Binance Dollar/BUSD):

2. Ethereum-USD Pair (supports Tether/USDT only; Binance Dollar support under development):

3. Medium-Sized Pairs (supports Tether/USDT only):

4. Smallest-Sized Pairs Not Otherwise Specified (supports Tether/USDT only):

We can see the trend in maintenance margins by plotting the four tiers of futures contracts on a log-log chart. We use longer-wavelength colors to indicate higher-risk trading pairs:

We can put our equations and tables to work by making a sample calculation. For no discernible reason, the ‘Chans have adopted ChainLink as their preferred cryptocurrency, selecting it from among myriad alternatives without any definite cause. We will honor this accident-of-history by analyzing a hypothetical ChainLink long position with nominal value of $1M, entry price of $20, and collateral of $200,000. The leverage factor of this position is:

By the standards of cryptocurrency, this is relatively “moderate” leverage, and well within the limits established by Binance. Looking at Table 3, we see that ChainLink is a “medium-sized pair”, and that a position of $1M requires a 5% safety margin. This theoretically allows leverage as high as (1) / (5%) = 20x, so our ChainLink bull is not over-leveraged (at least, not in the sense of risking a margin call) at the current price.

We can calculate this trader’s liquidation point (i.e., the price that ChainLink must fall to in order to trigger a liquidation) by setting the “safety margin”, S%, to 5%. We can then rearrange to find P(t), the price at time t, which will push the trader’s margin to this level:

This trader will be liquidated if the price of ChainLink (LINK) falls from $20 to $16. This liquidation-point calculation is the first of three key concepts critical to the Liquidator Finance strategy.

V. FUNDING RATES: THE CANARY IN THE COALMINE

The second critical calculation relates to funding rates in futures markets. In order to understand funding rates, we must first understand the difference between “spot” and “mark” prices on exchanges. The “spot” price is what we generally think of as the fiat price of an asset; it is the spot price that one reads on CoinMarketCap (CMC) and CoinGecko (CG), and when prices are quoted in the news, they are invariably spot prices. The “mark” price of an asset, however, is the price that is quoted in futures markets. Cryptocurrency perpetual futures are not excercisable in the sense that options are; therefore the futures market and spot market are isolated and can have different prices. For example, at the time of this writing, the prices of Bitcoin on Binance are as follows:

BTC Spot Price: $41,851.24
BTC Perpetual Futures Price (“Mark Price”): $41,868.19

Although spot and futures markets are isolated from each other, both are subject to the same principles of supply-and-demand (in fact, all assets are subject to these principles). The fact that the BTC mark price is higher than the spot price (by $16.95, or roughly 0.04%) demonstrates that there is more demand for Bitcoin futures contracts than Bitcoin itself. It’s not difficult to understand why: futures offer more leverage than spot positions, and with net bullish sentiment, futures are more attractive than spot positions. As a result, futures are priced at a premium. Note that this divergence is reflective of trader behavior and nothing more; the fact that BTC futures contracts cost more than spot Bitcoin does not imply that futures are “better” in any deep theoretical sense.

In conventional, fixed-term futures markets, the contract price (“mark price”) will naturally converge towards the spot (“true”) asset price, because divergence presents an arbitrage opportunity to those with enough capital to actually exercise options. In this sense, fixed-term futures are self-correcting towards spot prices for the same reason that stock options will likewise converge towards market value. Perpetual futures, however, are not self-correcting. Without any mechanism to force mark prices towards spot prices, exuberant traders could purchase long futures for some asset to such an extent that the mark price would rise far above the spot price, rendering these financial instruments useless as they no longer reflect changes in the price of the base asset. Similarly, a pack of bears opening short futures positions could force mark price far below spot price.

Binance and other exchanges therefore use a system known as “futures funding” to push mark prices towards spot prices. The funding system is, in a sense, a way of financially punishing those who open positions that increase divergence, while rewarding those who open positions that decrease divergence. Funding involves the involuntary transfer of collateral balances between traders, where the amount of transfer is proportional to the notional value of a position. Therefore, two traders with identical collateral balances taking positions in the same direction on the same asset pair at the same time will not necessarily experience equal transfers: if one trader is more leveraged than the other (and hence has a larger notional position against the same collateral), he will experience larger transfers.

Within Binance, funding occurs every 8 hours. A positive funding rate indicates that positions are overall long (and so long-holders will pay short-holders), whereas a negative funding rate implies that positions are overall short (and shorts pay longs). The funding amount a trader will pay (or receive) is simply the product of funding rate and notional position value:

All futures pairs on Binance are synchronized with respect to funding rates; that is, funding on the ETH-USD pair occurs at the same time as funding on the BTC-USD pair, and so-on. For a trader with n positions associated with the same margin account (cross-leverage), the net funding for positions i=0 to i=n will be :

From a trader’s perspective, funding rates are important because of their potential to erode margins and push a position closer to liquidation even if the market itself does not move in a direction adverse to that trader’s position. Simply holding a position for an excessive amount of time can lead to losses if that position is in the same net direction as most other positions. This means that a trader who “goes with the flow” and is generally bullish while others are bullish and bearish while others are bearish will see his collateral decrease the longer a position is held open. This doesn’t apply to all traders, but it applies to most, because most traders are indeed “going with the flow” – by definition, “going with the flow” is whatever most traders are doing. Certain traders will take a contrarian stance, opening shorts while futures markets are net long, and vice versa. These traders can generate modest gains at minimal risk, and their strategy is a reasonable one, despite the mockery they receive on Binance Trollbox. Funding transfers are trader-to-trader, and while exchanges set funding rates, they do not themselves collect or pay out these transfers. Hence, what is to some the cost of doing business (i.e., the cost of holding a position open) is to others a golden opportunity.

Liquidator Finance does engage in opening and closing futures positions, so these considerations are of some interest to us. However, our primary interest is not in how funding rates affect long-term profitability of positions, but rather, in what we can uncover about futures markets by studying funding rates.

In the most general sense, funding rates reveal how ripe the futures market is for liquidations. When funding rates are high, this indicates that most traders are highly-leveraged into long positions, setting the stage for a long squeeze. When funding rates are low (i.e., negative and with high absolute value), this indicates high leverage in short positions and prepares us for a short squeeze. When funding rates are near-zero, this indicates little net leverage and suggests that we at Liquidator would do best to bide our time before entering the markets and triggering liquidations to generate profit.

Funding rates on Binance and other exchanges are the sums of two main components: the static rate and dynamic rate. We will define each in turn.

Static Funding Rate (sometimes called “flat interest rate”): This is a fixed, rarely-changed value which ostensibly reflects the difference in expected interest rates between holding cryptocurrency and holding fiat currency. Static funding rate is currently positive on Binance (at 0.01% per 8 hours), which implies that interest rates are higher for fiat currencies than cryptocurrencies. This is true in a narrow sense – for example, BlockFi offers interest rates of 9% – 10% APY on deposits in USDC or USDT, as opposed to 4%-5% on BTC. In an inflation-adjusted or price-adjusted sense, however, it’s clearly false, or at least misleading: lenders may offer higher interest on fiat than crypto, but a cursory look at bitcoin’s history indicates that buying BTC and holding it has been vastly more profitable than holding fiat currency in interest-bearing accounts. Regardless, Binance’s position is essentially this:

When traders open long futures positions, we as an exchange are forced to purchase crypto on these traders’ behalves, and since interest rates are lower on crypto than on fiat, these traders must pay compensation for forcing us into a lower-interest-rate asset allocation.

As a result, even when long and short futures positions are perfectly balanced (as all things should be), traders with long positions will still pay traders with short positions at 0.03% of notional value per day (three 8-hour funding periods per day at 0.01% each). The static funding rate can therefore be understood as creating a very slight bearish bias to futures markets, i.e. funding rates are naturally “biased” against bullish long-position traders.

Binance reserves the right to adjust the static funding rate as it sees fit, but has rarely made adjustments, and in general, this flat interest rate is minuscule and of little consequence. Importantly, it does not force convergence between mark and spot prices, and it does not reveal anything about leverage or the overall bullishness or bearishness of the futures market.

Dynamic Funding Rate: Here we come to the heart of the matter. The dynamic funding rate is proportional to the divergence between mark price and spot price (i.e., the higher that mark price is above spot price, the higher it will be; if mark price falls below spot, dynamic funding rate will tend to become negative). The dynamic funding rate is also affected by the hypothetical price impact of buying certain quantities of long or short futures. Hence, the dynamic funding rate will become more aggressive (further from zero, i.e., more strongly positive or more strongly negative) in situations in which the futures market in question is thinly-traded and in which individual trades will have substantial impacts on mark price.

Prior to delving into the dynamic funding rate, we need to examine how exchanges model the price-impacts of trades. We know, from supply and demand, that opening a long-futures contract will tend to raise mark prices, while opening a short-futures contract will tend to lower mark prices. The opposite occurs during closure: closing a long position pushes prices down, and closing a short position pushes prices up. These perturbations are caused by the finite liquidity levels seen on all real exchanges. In an “ideal” world, in which all exchanges had infinite liquidity and all trades were infinitesimally small relative to this liquidity pool, trades would have zero price impact, their puny volume being absorbed into an oceanic order-book. The non-ideality of the real world makes Liquidator’s strategy possible.

Binance uses the term “impact margin notional” (IMN) to designate a dollar-value which corresponds to the largest notional futures position that could be opened with $200 in collateral. The IMN therefore depends on the leverage available in a futures market. Currently, there are four IMNs, corresponding to four tiers of cryptocurrency, ranging from the most thickly-traded to the most thinly-traded:

The impact margin notional values for these four tiers are therefore:

1. Bitcoin (125x Maximum Leverage): $25,000
2. Ethereum (100x Maximum Leverage): $20,000
3. Larger Altcoins* (75x Maximum Leverage):
$15,000
4. Smaller Altcoins** (50x Maximum Leverage): $10,000

*Cardano (ADA); Binance Coin (BNB); PolkaDot (DOT); EOS; Ethereum Classic (ETC); ChainLink (LINK); LiteCoin (LTC); Tron (TRX); Stellar Lumens (XLM); Monero (XMR); Ripple (XRP); Tezos (XTZ); BitCoin Cash (BCH).

**All others besides Bitcoin, Ethereum, and designated “Larger Altcoins”.

From these IMN values, Binance performs actual trades on its own platform once per minute in each futures market and monitors the price impact of these trades. Specifically, it monitors the bid-ask spread based on the average price at which its orders fill. If nothing else, this approach should serve as a testament to just how complex liquidity is: so complex that the largest cryptocurrency exchange in history chooses to determine it experimentally rather than to model it, despite having access to order books and various proprietary information of great value to any data scientist.

The Dynamic Funding Rate is expressed in terms of bid-fill and ask-fill prices as below:

There are two special cases in which this equation is trivial and evaluates to zero. The first is the case in which (Abid – Qspot) = (Qspot – Aask), i.e. the upwards price impact of opening a long position is exactly equal to the downwards price impact of opening a short position, assuming equal position values. The second is the infinite liquidity case, in which neither asks nor bids influence the price at all. We will ignore these cases going forward in this whitepaper; at Liquidator, we are interested in cases where dynamic funding rate is non-zero and the market is unbalanced.

We can make the above equation more useful by changing how we think about the difference between bid and spot prices. Specifically, (Abid – Qspot) will always be exactly half the upwards price impact of opening a long futures contract for an IMR value (in the case of BTC, $25k) worth of tokens. Suppose the current price of BTC is $100,000 and a purchase of a $25,000 long-futures contract pushes the price up to $100,001. The very first infinitesimally-small portion of this order fills at $100,000 (the mark price before opening), and the very last infinitesimally-small portion fills at $100,001 (the price after the order completely fills). Therefore the average fill price is halfway between these at $100,000.50, and (Abid – Qspot) = ($100,000.50 – $100,000) = $0.50, or exactly half the price impact. We can say the same thing about (Qspot – Aask), which is exactly half the (downwards) price impact of opening a short futures contract of IMR value. If we treat these variables as changes in price over time (specifically, between the time immediately before an order fills and the time immediately after it fills), we can rewrite them as derivatives:

These derivatives cannot be solved symbolically; they can only be solved approximately through numerical methods based on computations drawing from order book and open interest data. The exact approach we take is a trade secret of the Liquidator DAO, and we will not disclose it here in order to protect our competitive advantage. Disclosing our approach and the API queries involved could also threaten our API access, primarily for exchange APIs but potentially also for third-party data sources, namely Glassnode. Qualitatively, we can say this: while Binance calculates funding rate based upon the price impacts of trades, we have an effective method to reverse the equation. We start with the funding rate provided by Binance and use it in combination with other data sources to determine the price impacts of opening short or long positions. When price sensitivity is high, we rapidly open positions that force other traders into liquidation, allowing us to close our contracts quickly and at enormous profit.


VI. THE LIQUIDATOR STRATEGY

“Don’t trade futures. Don’t trade with leverage”.

You’ve heard it before, including from many of the top thinkers in the cryptocurrency space. They aren’t saying this because they are financially cautious or “conservative”; very few crypto traders could be honestly described as such, at least by the standards of conventional finance. They’re saying this because the overwhelming majority of leverage traders lose money when their positions are inevitably liquidated into insurance funds. Exchanges profit so heavily from busted futures positions that they have been credibly described as being “liquidation engines”. Criticism of leveraged-future trading doesn’t stem exclusively from fears about traders getting rekt; liquidations also affect the spot market, and the volatility caused by over-leveraged traders losing their shirts has contributed to major crashes in the larger crypto market, scaring retail investors and institutions alike away from digital assets. Therefore, there is a widespread belief that futures trading is- for lack of a better word- “bad” for cryptocurrency writ large.

The Liquidator DAO does not take any absolute position on leverage being “good” or “bad”. Only a Sith deals in absolutes, after all, and there’s a massive difference between trading altcoin futures at 125x leverage and using short futures at moderate leverage to hedge against market downturns. What we can say is this: As long as futures exist, traders will from time to time get rekt, and that money has to go somewhere. Why should exchanges have all the fun? We’ll take those profits, thank you.

We generate revenue for token-holders in a three-step process.

Step 1: A small portion of liquidity is temporarily removed from the DEX (decentralized exchange) pool. This removal is “symmetrical”, i.e. Liquidator tokens and

VII. VOTING

IIX. TOKENOMICS

IX. SOLIDITY CONTRACT

X. CONCLUSION