Futures, options and perpetuals: how derivatives work

A market is a place where buyers and sellers meet. A price is not a fact about the world. It is the most recent amount that one buyer and one seller agreed on. If a concert ticket last changed hands at 80, the price is 80, until the next deal happens at 85 or 70. Prices move because new information or new opinions arrive. More people want to buy than sell, and the price rises until sellers show up. More want to sell than buy, and it falls until buyers show up.

The bid is the highest price any buyer is currently willing to pay. The ask (also called the offer) is the lowest price any seller is currently willing to accept. The spread is the gap between them. If the bid is 99.95 and the ask is 100.05, the spread is 0.10.

The spread is a cost of trading. If you buy at the ask (100.05) and immediately sell at the bid (99.95), you lose 0.10 per unit even though nothing about the asset changed.

This list is called the order book: every waiting offer to buy or sell, sorted by price. The best bid and best ask sit in the middle. A trade happens when a buyer and a seller meet at the same price.

A market order says: fill me immediately at the best available prices. A limit order says: fill me only at this price or better. A market order gives certainty of getting in but not of the price. A limit order gives certainty of price but not of getting filled.

Going long means buying, hoping to sell later at a higher price. Going short means selling something you do not own yet: you borrow it, sell it, and later buy it back to return it. You profit if the price falls in between.

Volatility measures how large and fast prices swing. An asset that usually moves 0.5% a day is calm. One that often moves 5% a day is volatile. High volatility means bigger possible gains and bigger possible losses, which matters a great deal once leverage is involved.

Liquidity is how easily you can buy or sell without moving the price. A liquid market has a tight spread and a deep order book, so large orders barely change the price. In a thin market, as the worked example showed, even a modest order walks up the book and costs you extra.

A derivative is an agreement between two parties whose value depends on (is derived from) something else, called the underlying. The underlying can be a stock, an index, oil, wheat, an interest rate, a currency or a cryptocurrency. You do not have to own the underlying to hold the derivative.

Analogy: car insurance is a derivative. Its value depends on what happens to your car, but the policy is not the car. Another: a concert ticket resale contract that says 'I will sell you my ticket for 100 next month' is worth more or less depending on the ticket's market price, without either of you holding the ticket today.

Hedgers want less risk, so they pay to pass it on. Speculators accept that risk, and give hedgers someone to trade with. Arbitrageurs keep related prices lined up. A healthy derivatives market needs all three.

A forward is a private agreement, so it has a weakness: one side might not pay. This is called counterparty risk, the risk that the other party defaults. Two inventions reduce it.

Over-the-counter (OTC) means a contract negotiated privately between two parties, such as a bank and a company. It can be tailored exactly, but you carry the counterparty's credit risk and it may be hard to exit. Exchange-traded means standardized contracts traded on a public venue with visible prices.

On an exchange, a clearing house sits in the middle. Once you and another trader match, the clearing house becomes the buyer to every seller and the seller to every buyer. You no longer depend on the stranger paying; you depend on the clearing house, which demands collateral (margin) from everyone and settles gains and losses every day. The next chapters show how.

The Bank for International Settlements (BIS) reports that the notional value of outstanding over-the-counter derivatives has been on the order of roughly 600 to 700 trillion US dollars in recent years. Notional is the face size of the underlying exposure, not the money that changes hands. The actual market value of those contracts (what they would cost to replace) is far smaller, roughly a few percent of notional. The headline figure is a measure of how much exposure is being managed, not of how much could be lost.

Every contract has a specification (spec). It lists the underlying, the contract size (also called the multiplier), the tick (the smallest price step and what it is worth), the expiry date, and how it settles. Micro contracts are one tenth the size, so a small account can take a smaller position. Size is the main reason to choose one over the other.

Going long means you agreed to buy; going short means you agreed to sell. Profit and loss per contract is the price change in points times the multiplier. Crucially, gains and losses are not saved until the end. Each day the exchange marks your position to the closing settlement price and moves cash between accounts. This is called mark-to-market, or daily settlement.

You do not pay 250,000 for that E-mini. You post margin, a good-faith deposit that covers possible daily losses. The exchange and your broker set the amount, and it changes with volatility. If you were asked for, say, 12,500 (illustrative), that is 5% of the notional. Chapter 4 covers margin and its dangers in depth.

Cash-settled futures (like stock index futures) end with a final payment of the difference; nobody delivers 500 stocks. Physically settled futures (like many commodity contracts) require real delivery of the goods. Most speculators close their position before expiry by trading the opposite side, so they never face a delivery.

If the futures price is above the spot price (the price for immediate delivery), the market is in contango. Intuition: holding the goods until later costs storage, insurance and interest, and the futures price reflects that. If the futures price is below spot, it is called backwardation, which often appears when the good is scarce right now. Either way, the futures price converges to spot as expiry arrives.

Leverage means controlling a position larger than the money you put up. Leverage = position size / your own money in the position. Put 1,000 behind a 10,000 position and you have 10x leverage. A 1% move in the underlying then changes the position by 100, which is 10% of your 1,000. Leverage is a magnifier, and it magnifies in both directions equally.

Initial margin is the deposit required to open a position. Maintenance margin is the minimum account equity you must keep while the position is open. Equity means your deposit plus unrealized gains minus unrealized losses. Maintenance is lower than initial, which creates a buffer. When losses push equity below maintenance, the venue acts.

In traditional futures, the broker issues a margin call: a demand to add money by a deadline. If you do not, they close some or all of your position. A liquidation is that forced closing. Many crypto venues skip the call entirely and liquidate automatically, because markets run all hours and prices can move fast.

The wipe-out move is simply 1 / leverage. In practice liquidation happens sooner, because the maintenance requirement eats part of the buffer. At 100x, an ordinary hourly wiggle in a volatile market can be enough.

The leverage setting on a screen can mislead. Your real risk is how large the position is compared with your whole account, and where your exit is.

• Decide the most you will lose on one trade before entering (for example 1% of the account), then size the position so a stop at your exit costs only that.
• Leave extra cash as a buffer so a normal swing cannot trigger liquidation.
• Remember that gaps and fast markets can liquidate you at a worse price than expected.
• Not trading is also a valid decision, especially while you are learning.

An option is a contract that gives its buyer the right, but not the obligation, to buy or sell something at a fixed price on or before a fixed date. Think of a concert ticket you reserve for a small fee. You can still go and pay the agreed price, or you can walk away and lose only the fee. Nobody forces you to use it.

A call option is the right to buy. A put option is the right to sell. The thing being bought or sold is the underlying, such as 100 shares of a stock called XYZ. Three numbers define every option:

• Strike price: the fixed price at which the option lets you buy (call) or sell (put). Say 105.
• Expiry: the last date the option exists. After it, the contract is worthless if unused.
• Premium: the price you pay today to own the option. It is quoted per share. The seller keeps it whatever happens.

Traders describe an option by comparing the strike with the current stock price. If exercising it right now would give you a gain, it is in the money. If the strike equals the stock price, it is at the money. If exercising would be pointless, it is out of the money. Here the stock XYZ trades at 100.

Every option has two parties. The buyer (long) pays the premium and gets the right. The seller, also called the writer (short), receives the premium and takes on the obligation: if the buyer uses the right, the seller must deliver. The buyer's loss is capped at the premium. The seller's gain is capped at the premium. That asymmetry is the whole game.

An option's price (premium) has two parts. Intrinsic value is what the option would be worth if exercised right now. Time value is everything else: what traders pay for the chance that the option becomes more valuable before expiry.

Time value is larger when there is more time left and when the stock is likely to move a lot. More time means more chances for a big move. A big move can help a buyer a great deal while the buyer's loss stays capped at the premium, so buyers pay more when big moves are likely.

Volatility measures how much a price tends to jump around, usually quoted as an annual percentage. Realised (historical) volatility is what the stock actually did in the past. Implied volatility (IV) is the number you must plug into a pricing model to get the option's current market price. It is the market's forecast of future movement, read backwards from the price.

Black-Scholes is the classic pricing formula, published in 1973 by Fischer Black and Myron Scholes with Robert Merton. It takes five inputs: the stock price, the strike, the time to expiry, the interest rate and the volatility. It gives a fair price by assuming the stock moves randomly and smoothly. Real markets jump, so traders use it mainly as a translator between a price and an IV, not as a prediction. You never need to compute it by hand.

The Greeks are measures of how an option's price reacts when one input changes. Each is named after a Greek letter. Numbers below are per share. Multiply by 100 for a contract.

• Delta: how much the option moves when the stock moves $1. A call with delta 0.30 gains about 0.30 ($30 per contract) if XYZ rises $1. Puts have negative delta. A stock share has delta 1.
• Gamma: how fast delta itself changes. With gamma 0.05, after XYZ rises $1 the call's delta goes from 0.30 to about 0.35. Gamma is highest for at-the-money options near expiry.
• Theta: time decay, the value lost each day if nothing else changes. Theta of -0.04 means the option loses 0.04 a day, $4 per contract per day, and the loss tends to speed up toward expiry.
• Vega: sensitivity to implied volatility. Vega of 0.08 means a one-point rise in IV (say 40% to 41%) adds 0.08, $8 per contract.
• Rho: sensitivity to interest rates. It is usually small for short-dated options. A rate rise helps calls slightly and hurts puts slightly.

A strategy combines options (and sometimes shares) to shape the payoff. Before any detail, sort them by worst case. Defined risk means the maximum loss is known before you enter. Undefined risk means the loss can be very large, even unlimited, or large relative to the gain. Every example below uses XYZ at 100, one contract of 100 shares, 30 days to expiry, and these illustrative premiums per share: 100 call 4.00, 105 call 2.00, 110 call 1.00, 100 put 3.50, 95 put 1.50, 90 put 0.50.

1. Covered call. Setup: you own 100 shares and sell the 105 call for 2.00, receiving $200. If XYZ ends above 105, your shares are called away at 105. Max gain: (105 - 100 + 2) x 100 = $700. Max loss: if XYZ goes to 0, (0 - 100 + 2) x 100 = loss of $9,800. Break-even: 100 - 2 = 98. Used by holders who accept a ceiling in exchange for income. Main risk: the shares can still fall a lot, and the upside is capped.

2. Protective put. Setup: you own 100 shares and buy the 95 put for 1.50, paying $150. This is insurance with a 5-point deductible. Max loss: (100 - 95 + 1.50) x 100 = $650. Max gain: unlimited. Break-even: 100 + 1.50 = 101.50. Used by holders wanting protection through a risky event. Main risk: the premium is a steady drag if nothing bad happens.

3. Cash-secured put. Setup: you hold $9,500 in cash and sell the 95 put for 1.50, receiving $150. If XYZ ends below 95, you must buy 100 shares at 95. Max gain: $150. Break-even: 95 - 1.50 = 93.50. Max loss: if XYZ goes to 0, (1.50 - 95) x 100 = loss of $9,350. Used by people who would be happy to own XYZ at 95 anyway. Main risk: a crash leaves you owning a falling stock bought above the market.

A spread buys one option and sells another to cut the cost and cap the risk. 4. Bull call spread. Buy the 100 call (4.00), sell the 110 call (1.00): net cost 3.00, or $300. Max loss: $300. Max gain: (110 - 100 - 3) x 100 = $700. Break-even: 100 + 3 = 103. Used for a moderate bullish view at lower cost than a plain call. Main risk: the full $300 is lost if XYZ ends at or below 100, and gains stop at 110.

5. Bear put spread. Buy the 100 put (3.50), sell the 90 put (0.50): net cost 3.00, or $300. Max loss: $300. Max gain: (100 - 90 - 3) x 100 = $700. Break-even: 100 - 3 = 97. Used for a moderate bearish view or a hedge. Main risk: the mirror image of the bull call spread.

6. Long straddle. Buy the 100 call (4.00) and the 100 put (3.50) for 7.50, or $750. You profit from a big move either way. Break-evens: 100 + 7.50 = 107.50 and 100 - 7.50 = 92.50. Max loss: $750 if XYZ ends exactly at 100. Max gain: unlimited upward, at most $9,250 downward. Used around events. Main risk: the move must beat the move already priced in, and time decay works against you every day.

7. Long strangle. Buy the 105 call (2.00) and the 95 put (1.50) for 3.50, or $350. Cheaper than a straddle but needs a bigger move. Break-evens: 108.50 and 91.50. Max loss: $350 if XYZ ends between 95 and 105. Main risk: the same as a straddle, with a wider dead zone.

8. Iron condor. Sell the 95 put (1.50), buy the 90 put (0.50), sell the 105 call (2.00), buy the 110 call (1.00). Net credit: 1.50 - 0.50 + 2.00 - 1.00 = 2.00, or $200. Max gain: $200, if XYZ ends between 95 and 105. Max loss: (5 - 2.00) x 100 = $300, if XYZ ends at or beyond 90 or 110. Break-evens: 95 - 2 = 93 and 105 + 2 = 107. Used by those expecting a quiet stock. Main risk: you risk $300 to make $200, and one sharp move gives back many small wins.

Time decay (theta) is the daily loss of time value. It hurts buyers and pays sellers, and it accelerates near expiry. If XYZ stays at 100, an at-the-money option that cost 3.00 with 30 days left is worth roughly 2.12 with 15 days left and 1.22 with 5 days left.

A buyer who is right about direction can still lose if the move is too slow. The stock must move far enough and soon enough to beat the decay.

Assignment is when a seller is told to fulfil the contract. US stock options are American style, meaning the buyer may exercise any time before expiry (early exercise). Short in-the-money calls are most at risk of early assignment just before a dividend date. If you sold the 105 call and XYZ is at 112, you may be forced to sell 100 shares at 105. Index options such as those on the S&P 500 are usually European style (exercised only at expiry) and cash-settled.

Pin risk occurs when the stock closes right at your short strike. You cannot know whether the buyer will exercise. Options that finish even slightly in the money are typically exercised automatically, so you can wake up on Monday owning or owing 100 shares you did not expect, with the stock having moved over the weekend.

Wide spreads: the bid is what buyers will pay, the ask is what sellers want. Illiquid contracts have a big gap. If the bid is 1.00 and the ask is 1.40, you buy at 1.40 and could sell right back at 1.00, losing 0.40 ($40, nearly 29%) immediately. Trade contracts with tight spreads and real volume, and use limit orders.

0DTE options expire the same day they trade. They are cheap, gamma is extreme, and the outcome is settled within hours. They attract a very large share of volume in some index options. Example: XYZ is at 100, a 101 call expiring today costs 0.40 ($40). XYZ ends the day at 100.90. The call expires worthless and you lose all $40, though the stock rose almost a full point. Repeating that many times a day is how small losses add up.

Cheap out-of-the-money options are tempting. A 120 call, 30 days out, costs 0.20 ($20). Fifty contracts cost $1,000. XYZ must pass 120.20 to profit. With IV of 30%, a one standard 30-day move is about 8.6%, so 20% is far outside the usual range. A low delta (around 0.03) hints at a low chance. The price is low because the probability is low. A lottery ticket is not a bargain simply because it is cheap.

• Know your maximum loss in dollars before you click: premium x 100 x contracts for bought options, width minus credit for spreads, and treat naked shorts as unbounded.
• A common rule of thumb is to risk only 1% to 2% of the account on one trade. On a $10,000 account, 2% is $200, so one contract costing $200 fits and fifty $20 lottery tickets ($1,000, or 10%) do not.
• Count overlapping trades. Five option trades on the same stock are one big bet.
• Practise with paper trading first. Most retail option traders lose money over time, and sizing is the main reason a loss stays survivable.

A spread is a position built from two options of the same type (both calls or both puts) on the same stock and the same expiry date, with different strikes. You buy one and sell the other. The option you sell pays for part of the option you buy, and it also puts a lid on your risk. Traders use spreads for three reasons: they cost less than a plain option, the worst case is known in advance, and the best case is capped too. Cheaper and safer is paid for by giving up the big jackpot.

Every example in this chapter uses a stock called XYZ at 100, 30 days to expiry, an implied volatility of about 25%, and contracts of 100 shares. The premiums are illustrative and rounded, quoted per share: calls at 95, 100, 105 and 110 cost 6.20, 3.00, 1.20 and 0.35; puts at 90, 95 and 100 cost 0.20, 0.90 and 2.70. They differ a little from the round numbers in the previous chapter because the volatility is different. Real quotes will vary.

A bull call spread means buying a lower-strike call and selling a higher-strike call, both with the same expiry. You are bullish, but only up to a point. A debit spread is any spread where you pay money to open it (a net debit). The gap between the two strikes is the width, and it sets the most the spread can ever be worth.

Read the last row: at 110 the long call is worth 10 but you owe 5 on the short call, so the spread is worth 5. Nothing beyond 105 is yours. Try your own numbers below.

A bear put spread buys a higher-strike put and sells a lower-strike put. You make money if the stock falls, up to the lower strike.

A credit spread is the opposite way round: you sell the option closer to the stock price and buy a further one as protection. You receive money up front (a net credit). Your best outcome is keeping that credit. The bought option exists to cap the loss, like a seat belt.

• Delta as rough odds. A short option with delta 0.21 (like the 95 put) is often read as roughly a 21% chance of finishing in the money. That is a rough guide, not a promise. Selling closer to the stock price pays more and loses more often.
• Expiry. Many traders use 30 to 60 days. Shorter expiries decay faster but leave little time to be right. Longer ones cost more and move more slowly.
• Width. A wider spread risks and pays more in dollars. Pick the width from the dollars you can afford to lose, not from the premium you hope to get.

The Greeks of a spread are the long leg's numbers minus the short leg's, so they partly cancel. Take the bull call spread above, per share: delta 0.53 - 0.27 = +0.26, theta about -0.053 + 0.043 = -0.010 per day, vega about 0.114 - 0.096 = +0.02 per volatility point. In contract terms, XYZ up $1 adds about $26, time costs about $1 a day and a one-point rise in implied volatility adds about $2. Compare a plain 100 call: delta +0.53, theta -0.053 (about $5.30 a day), vega +0.114. The spread is far less sensitive to everything. A credit spread is the same idea in reverse: you start with positive theta, and a rise in implied volatility usually hurts.

Assignment is being told to fulfil the contract you sold. US stock options can be exercised by the buyer any day. A short call that is deep in the money, especially just before a dividend, may be assigned early. If the short 105 call is assigned, you must deliver 100 shares at 105. Your long 100 call still exists and can be exercised or closed, so the loss is capped, but the cash and share movements can be surprising. Pin risk is when the stock closes almost exactly at the short strike on expiry day: you cannot be sure whether the short leg will be exercised, and you may start the next week with an unwanted share position that moved over the weekend. Many traders close spreads before the last day for this reason.

Managing means deciding in advance when to act. Common rules are to close a profitable credit spread after collecting a large part of the credit, and to close a losing one at a set loss rather than hoping. Rolling means closing the spread and opening a similar one with a later expiry, sometimes at different strikes. A roll can give a trade more time, but it adds new costs, and it only postpones a loss if the view was wrong.

• Ignoring the width: the loss on a credit spread can be many times the credit.
• Choosing the cheapest spread. Far out-of-the-money debit spreads are cheap because they rarely pay.
• Trading illiquid options and paying wide bid-ask gaps on two legs.
• Holding through expiry and meeting pin risk or assignment.
• Oversizing: ten contracts of a $430 risk is $4,300 at stake.
• Forgetting that most retail option traders lose money over time. Paper trade first.

The previous chapter used two options of one type. This chapter combines calls and puts, or options and shares, to bet on how far a stock moves rather than only which way. Numbers are the same as before: XYZ at 100, 30 days, implied volatility about 25%, 100 shares per contract. Per share premiums: calls at 95, 100, 105, 110 cost 6.20, 3.00, 1.20, 0.35; puts at 90, 95, 100 cost 0.20, 0.90, 2.70. The market's typical 30-day move at 25% is about 25% x sqrt(30/365) = 7.2%, so roughly 7 points either way. Keep that 7 in mind: it is the bar a volatility trade has to beat.

Who uses it: traders expecting news such as earnings or a court ruling. The main trap: the 5.70 you pay already assumes about a 5.70-point move. If the stock moves 4, you lose even though you were right about a move. After the event, implied volatility usually falls (IV crush), which cuts the value of both options at once. Time decay (theta) also costs you every day.

Selling a straddle or strangle is the opposite bet: you collect the premium ($570 for the straddle, $210 for the strangle) and win if the stock stays calm. The maximum gain is the premium. Break-evens are the same numbers as the long versions. The maximum loss, however, is not capped.

An iron butterfly is the same shape built from different pieces: sell the 100 call and 100 put (credit 5.70), buy the 105 call and 95 put (cost 2.10). Net credit 3.60, or $360, which is the maximum gain at exactly 100. Maximum loss: (5 - 3.60) x 100 = $140. Break-evens 96.40 and 103.60. These are the same numbers as the call butterfly, which is no accident: the two positions are equivalent in payoff.

A calendar spread (time spread) sells a near-dated option and buys a later-dated one at the same strike. Example, conceptually: sell the 30-day 100 call at 3.00 and buy the 60-day 100 call at about 4.35, for a debit of about 1.35, or $135. The near option decays faster than the far one. If XYZ sits near 100 when the near call expires, the short call is worthless, while the long call still has 30 days of value (about 3.00 in this model), so the spread is worth around 3.00, a gain near $165 before costs. If XYZ runs far away from 100, both calls move in line and the gap narrows, and the loss approaches the debit. Calendars benefit from rising implied volatility and a quiet stock. The trap: you can hardly see the payoff at expiry, because the long option still has value, so results depend on the pricing model and on implied volatility staying put.

A futures contract is a deal to trade something at a set price on a set future date. When that date arrives, the contract ends (it expires). A perpetual future, or perp, drops the date. You can hold it for an hour or a year. It only ends when you close it, or when the exchange closes it for you because your money ran out. Why does that matter? Dated futures split trading across many contracts: one for March, one for June, one for September. Each has its own price and its own thin crowd. A perp puts everyone in one contract, so there is one deep pool of buyers and sellers. Traders also never have to roll, which means closing the old contract and opening the next one before expiry. A common claim is that BitMEX invented the perpetual in 2016. The fuller story is messier. The economist Robert Shiller proposed a perpetual futures contract in a 1993 paper in the Journal of Finance, as a way to cash-settle contracts on assets whose prices are hard to observe directly. BitMEX launched its XBTUSD perpetual swap on 13 May 2016 (BitMEX announcement). BitMEX is widely credited with making perps popular among crypto traders from 2016. The safe summary: the idea is older than crypto, and BitMEX spread it.

A perp screen shows more than one price. They answer different questions.

• Last price: the price of the most recent trade on this exchange's own order book. It can jump on a single thin trade.
• Index price: an average of the spot price (the price to buy the real asset today) across several outside venues. It is the anchor to reality.
• Mark price: the exchange's fair-value estimate, usually built from the index plus some information from the order book. Your profit and loss on screen, and your liquidation, use the mark price, not the last price.

With no expiry, nothing forces the perp price to meet the spot price. Funding fixes that. Funding is a small payment passed between traders holding long positions (bets the price rises) and short positions (bets the price falls). The exchange does not keep it. When the perp trades above spot, longs pay shorts, which makes longing less attractive and pulls the perp down. When the perp trades below spot, shorts pay longs, pulling it up. A typical formula, described on the Hyperliquid documentation as of Nov 2026, adds a premium term (how far perp is from index) to a small fixed interest term of 0.01% per 8 hours. That fixed term is why quiet-market funding is usually a small positive number: longs pay shorts a little.

Funding can turn negative. If the market is panicking and the perp trades below spot, the sign flips. With funding at -0.03% per 8 hours, the same $10,000 position pays nothing and instead the shorts pay: 10,000 x 0.0003 = $3.00 per 8 hours to longs. Positive funding means a crowd leaning long. Negative means a crowd leaning short. It is a mood gauge, not a prediction. Cash-and-carry (also called a basis trade) means owning the real asset and shorting a matching derivative at the same time, so price moves cancel out. The aim is to collect the gap. With perps, the gap is funding.

It sounds like free money. It is not. You need cash for the coin and extra margin for the short. If the price doubles, the short loses on paper before the coin gains are usable, which can force liquidation. Funding can flip negative and make you pay. And you carry exchange risk on both legs. Leverage means borrowing exposure: with $10 of your own money (margin) you control $100 of the asset. Liquidation is the exchange closing your position by force once your equity falls below a minimum called maintenance margin.

If the price jumps so fast that the exchange closes you at a worse price than your bankruptcy level, a hole appears. The insurance fund is a pot of money, built mostly from leftover margin of positions closed profitably, that fills such holes. If the fund is empty, auto-deleveraging (ADL) kicks in. In plain words: the exchange forcibly closes profitable traders on the other side, at the bankruptcy price, to cover the loss. Winners can be closed against their will. Exact ranking rules differ by venue, so read them.

A blockchain is a shared record book that thousands of computers keep identical, so no single company can quietly rewrite it. Hyperliquid runs its exchange on such a book. The order book (the live list of who wants to buy and sell at which price) lives on the chain itself, rather than inside one firm's private servers. This chapter describes how it works as of Nov 2026. It is education, not a suggestion to use it. According to the project's documentation, the system has two parts. HyperCore holds the trading engine and account balances: margin, positions, the order book. HyperEVM is a second environment where general programs (smart contracts, meaning code that runs on the chain) can live alongside it. The chain uses its own consensus system, meaning the way the computers agree on the order of events, called HyperBFT. The docs describe median order latency near 0.2 seconds for co-located users.

Self-custody means you hold the keys to your funds yourself. That removes the risk that a company freezes or loses your money. It also removes the safety net: lose the key or sign the wrong approval, and nobody can reverse it. Per the Hyperliquid docs (as of Nov 2026), perp fees start at 0.045% for takers and 0.015% for makers at the base tier, and fall with 14-day trading volume. A taker removes liquidity by trading against an order already on the book. A maker adds an order to the book and waits.

Funding is paid hourly, at one eighth of the computed 8-hour rate, and has a fixed interest part of 0.01% per 8 hours (about 11% a year). One aggregator, Coinperps, reported a crypto baseline near 10.95% annualised on 20 Oct 2026.

• Cross margin: all your free balance backs all your positions. Fewer surprise liquidations, but one bad position can drain everything.
• Isolated margin: each position gets its own slice of money. A loss is capped at that slice.
• Portfolio margin: a mode in which related positions share margin and offsets are counted. The docs say little beyond that, so read the current rules before using it.

Maximum leverage varies by market. The docs list BTC at up to 40x and ETH at up to 25x, with many smaller assets at 10x or less, and lower limits as position size grows. Maintenance margin is half the initial margin at the maximum leverage. For a 40x market that is 1.25% of position value. Position at 40x with $250 margin on $10,000: a move of only about 1.25% against you removes the buffer between the 2.5% initial margin and the 1.25% maintenance level. Sub-2% moves happen in minutes. When a position falls below maintenance margin, the system first tries to close it on the order book. For large positions it sends only part at a time. If that fails and equity falls far below the minimum, a liquidator vault, part of the community-funded HLP vault (depositors' pooled money that also provides market-making), takes the position. That means ordinary depositors can end up holding the risk of someone else's blown-up trade.

1. Mar 2025JELLY. Reportedly (OAK Research, Cointelegraph), a trader opened a large short in a thin token, pushed its price up and left the position with the HLP vault. Validators voted to delist the token and settle it at a pre-manipulation price. Lesson: thin markets can be pushed around, and validators can intervene.
2. Nov 2025POPCAT. Reportedly (CoinDesk, The Defiant), attacker wallets caused bad debt in the low millions of dollars. Lesson: pooled vaults can take losses from other people's manipulation.
3. 10-11 Oct 2025A market-wide cascade. Press reports describe a record of roughly $19-20B in liquidations across venues in about 24 hours, with Hyperliquid processing a large share. Auto-deleveraging closed some profitable positions. Lesson: in a crash, safety rules that look rare become real.

Figures are from aggregators and may be stale. Reportedly (Datawallet, citing DefiLlama, updated 27 Sep 2026), Hyperliquid did about $216.9B of volume over 30 days and held roughly a third of perp-exchange-on-chain volume. Open interest (the total value of positions still open) was reported at a record near $18B on one counting method, but methods differ by up to 2x. Treat these as order-of-magnitude figures. The exchange has its own token, HYPE. Reportedly (Datawallet, Sep 2026), 99% of perp trading fees go to an Assistance Fund that buys HYPE on the market. Whether that supports the price is not guaranteed and is not a reason to hold anything.

Whether you may legally use an exchange depends on where you live. Reports suggest users in some countries, including the US, are blocked in the terms of the app. This course cannot verify current rules, so read the terms and ask a qualified adviser. Do not use a VPN to get around restrictions: that can void any protection and may be unlawful.

1. Confirm you are legally allowed to use it where you live.
2. Copy the web address from a trusted source and bookmark it.
3. Start with a wallet that holds only trading money.
4. Send a tiny test amount first and check it arrives.
5. Read the fee table, margin rules and liquidation page.
6. Use isolated margin and low leverage, if at all.
7. Decide your maximum loss in dollars before the first trade.
8. Know how to withdraw, and test that too.

HIP-3 is a Hyperliquid rule set that lets an approved outside team, called a deployer or builder, launch its own perpetual markets on the exchange. Per the Hyperliquid docs (as of Nov 2026), a deployer must stake a large amount of HYPE (the docs say 500,000, held for at least 183 days after launch). The deployer sets the oracle, the leverage limits and how a market is settled. An oracle is the feed that tells the system the outside price of the thing being tracked. This is how a perp on an oil price or a stock index can sit on the same trading engine as Bitcoin. Reportedly (Coinperps, 20 Oct 2026), HIP-3 went live on mainnet in Oct 2025, and a large majority of its volume came from one builder. trade.xyz (written trade[XYZ] in its docs) describes itself as a non-custodial interface plus documentation. It does not hold your assets or run the exchange engine. According to its docs it lists perps on stocks, equity indices, commodities and currencies, plus pre-IPO contracts (agreements tied to the expected price of a company before it is publicly listed). It reportedly accounts for over 90% of HIP-3 open interest (Galaxy Research, Jul 2026, and Coinperps, Oct 2026).

Owning a share makes you a part-owner. A contract for difference (CFD) is a private bet with a provider on the price move. A stock perp is neither: it is a derivative whose price tries to follow the stock, held in a crypto wallet. If the tracking breaks, you have no company to appeal to.

The oracle is set by the builder. For a Korean stock market, Galaxy Research described that the mark price was a median of three inputs: the oracle price, the oracle adjusted by a short average of order-book drift, and the middle of best bid, best ask and last trade. When the outside market is closed, the on-chain book keeps trading, so the price can drift away from the last official close. Funding follows the HIP-3 rule: premium is half the sum of the impact bid and impact ask, divided by oracle, minus 1. Per Coinperps, the median funding rate on HIP-3 markets in Oct 2026 was about 0.000625% per hour (roughly 5.5% a year). Fees in standard HIP-3 markets are, per the docs, double the usual exchange fees, while a 'growth mode' can cut them by at least 90%. Leverage reportedly runs from 10x on most markets up to 20-50x on very large, liquid ones such as major indices and currencies.

• Oracle risk: the builder can set oracle prices, so one party's choices drive your liquidation.
• Market-hours gaps: stocks close at night and on weekends. News arrives while the real market is shut, and the on-chain price moves anyway, then the real market opens at a very different price.
• Weekend moves: no one can hedge with the real stock when it is closed, so prices can be jumpy.
• A single builder's rules: markets can be halted or settled by the deployer at its discretion.
• Thin liquidity in new markets: few orders mean large price swings on small trades.
• No shareholder rights or dividends.

Funding step: 0.000625% is 0.00000625 as a decimal. 2,000 x 0.00000625 = $0.0125 per hour, $0.30 per day, $9.00 over 30 days (positive funding paid by longs). That is small beside the price risk. Liquidation step: equity = 200 + 10 x (P - 200). Maintenance = 5% x 10 x P = 0.5P. Set equal: 200 + 10P - 2,000 = 0.5P, so 9.5P = 1,800 and P is about $189.47. A fall of 5.3% from $200 liquidates you, and your $200 is gone, less fees. Gap step: suppose bad news arrives on a weekend and the price opens at $170. Loss is 10 x (200 - 170) = $300. Your margin was $200, so you are $100 short. The exchange covers that from its backstop, and if that fails, from ADL on winners. You did not choose to pass your loss to others, but the system might.

It may suit people who already understand perps, can afford to lose all of the money, want round-the-clock access to a market, and accept builder and oracle risk. It should not be used by anyone who needs the money, who does not understand liquidation, who is not allowed to use it where they live, who hopes to learn by trading, or who would be satisfied simply owning the shares. Plenty of people choose not to trade this at all.

A venue is the place where your trade happens. Each type below suits different needs. None is risk-free. This table is general; real rules differ by country and firm, so check the current terms of anything you consider.

The regulated US futures route uses standardised contracts. A micro contract is a smaller version of a bigger futures contract, with a smaller value per point, so a mistake costs less. The options broker route allows defined risk: if you buy an option, the most you lose is the price you paid (the premium). Selling options without cover can lose far more.

A wallet is an app or device that holds the secret keys proving you own crypto. It does not store coins like a purse. The coins sit on the blockchain, and the keys let you move them. The seed phrase (also called a recovery phrase) is a list of 12 or 24 words that can rebuild all your keys on any device.

• Never share your seed phrase. No real support person, exchange or website will ever ask for it.
• Write it on paper, store it offline, and never photograph it or save it in email or cloud notes.
• A hardware wallet is a small device that signs transactions offline, so malware on your computer cannot read the keys. It is a strong upgrade once you hold meaningful money.
• Phishing means a fake message or site that tricks you into revealing a key or signing something. Fake sites often copy real ones with one letter changed. Type or bookmark the address yourself.
• Approvals: when a site asks you to approve spending, you may be giving it permission to move your tokens. Read what you sign, and revoke old approvals.
• Bridging means moving money from one blockchain to another. Choosing the wrong network or address can lose funds permanently.
• Test with a tiny amount first, every new route, every new site.

This is not tax advice. In many countries, closing a trade at a profit, swapping one crypto asset for another, or earning rewards counts as a taxable event, meaning something you must report and may owe tax on. Losses may sometimes be used against gains, but rules differ by country and by type of product. Funding payments and fees can also matter.

• Keep a record of every deposit, withdrawal, trade, fee and funding payment, with dates and amounts.
• Export statements regularly rather than waiting for the end of the year.
• Ask a qualified tax professional in your country before you trade, not after.

If, after reading all of this, you still want to try something, here is one cautious order of operations. It is not a recommendation to trade.

1. Day 1: confirm you are legally allowed to use the venue and understand the tax rules where you live.
2. Day 1: decide a total amount you can lose without any effect on rent, food, or debts. Treat it as already spent.
3. Day 2: set up a wallet or account, with a unique password and two-factor authentication where available. Store the seed phrase offline.
4. Day 2: bookmark the real website from an official source and avoid clicking links in messages.
5. Day 3: send a tiny test amount, confirm it arrives, then test a withdrawal.
6. Days 3-5: watch the screen without trading. Read prices, mark price, funding and the liquidation price.
7. Day 6: place one very small position at the lowest leverage, with the loss set in advance.
8. Day 7: review: write down fees paid, what you felt, and whether you followed your plan. Stop if not.

Professionals do not start by asking how much they can make. They start by asking how much they can lose and still come back tomorrow. Risk of ruin is the chance that a run of losses shrinks your account so far that you cannot continue, or cannot recover. Every trader, however skilled, hits losing streaks. The only question is whether the account survives them.

A loss and the gain needed to undo it are not the same size. If you lose 50% of 1,000 you have 500, and you need to double it, a 100% gain, just to get back to 1,000. The deeper the hole, the steeper the climb.

Bet size is the lever. Suppose you lose ten trades in a row, which happens even to good systems. Risking 10% of the account on each trade leaves you with 0.9 multiplied by itself ten times, about 0.35, so you have lost roughly 65%. Risking 1% per trade leaves 0.99 multiplied by itself ten times, about 0.90, a loss of roughly 10%. Same bad luck, completely different outcome.

A stop-loss is an order that closes your trade if the price reaches a level you chose in advance. The stop distance is the gap between your entry price and that level. Professionals fix the amount they are willing to lose first, then work out the size that makes the stop distance equal to that amount. The formula is: position size = money at risk divided by the stop distance per unit.

Notice that a wider stop means a smaller position, and a tighter stop allows a larger one. The risk per trade stays the same either way. Many professionals risk somewhere around 0.5% to 2% of the account per trade; beginners often start at the low end. That is a common rule of thumb, not a guarantee of safety.

Expectancy is the average result per trade over many trades. Expectancy = (win rate x average win) minus (loss rate x average loss). A positive number means the approach makes money on average, before costs. A negative number means it loses on average, however often it wins.

Real results are noisier than this table. Averages are estimates from past trades, costs and slippage reduce them, and conditions change. Still, win rate alone tells you almost nothing. Always look at it together with the size of wins and losses.

• Gaps: if a market closes at 100 and opens the next day at 90, a stop at 95 cannot trigger at 95. It triggers at the next available price, around 90.
• Slippage: in a fast market the price you actually get can be worse than your stop level, because there may be few orders at that price.
• Stop hunting by noise: a stop placed too close can be hit by ordinary wiggles, then the price moves your way without you.
• Venue rules: on a leveraged venue, a liquidation can happen before a far-away stop is reached. Keep the stop well inside the liquidation price.

Leverage is using borrowed or margin-backed exposure so a position is larger than your cash. Rule of thumb: use only as much as keeps your stop comfortably away from the liquidation price. At 10x, a move of roughly 10% against you can wipe out the margin (a little less once maintenance margin and fees are counted). Many beginners use 1x to 3x or less, and the leverage figure on the screen is a ceiling, not a target.

Diversification means spreading risk across things that do not always move together. Correlation measures how closely two prices move together. If you hold five long positions in five crypto coins, each risking 1%, they may all fall at once in a market scare, so the real risk is closer to 5% than 1%. Count correlated positions as one bigger bet.

• Keep a journal: record the reason for the trade, entry, stop, target, result and how you felt. Review it weekly to find repeated mistakes.
• Paper trade first: use a simulated account to test your rules and platform skills with no money at stake. It does not reproduce the stress of real losses, so treat it as practice, not proof.
• Set drawdown rules: a drawdown is the fall from an account's peak. A common structure is to stop for the day after losing a set amount (say 2% to 3%), cut size after a larger weekly loss, and pause completely and review after a bigger drawdown (say 10%). Decide the numbers when calm.

Before looking at ways to trade, look at how traders actually do. Regulators in several countries have measured retail results, and the numbers are consistent: most people who trade leveraged derivatives lose money. The sections below name the source of each figure so you can look it up. Definitions and years differ between studies, so compare the direction, not the decimals.

India: the market regulator SEBI studied individual traders in equity futures and options for the financial year 2024-25 (FY25). About 91% lost money, and their net losses were over Rs 1 lakh crore (Rs 1,05,603 crore in the first report, later revised higher), up about 41% on the year before [S, reported by Business Standard and Moneylife]. For FY26, a government statement reported by Moneylife put the losing share at 87.7% [S]. Falling is not winning: most traders still lost.

Brazil: economists Chague, De-Losso and Giovannetti followed 19,646 people who began day trading mini-Ibovespa index futures between 2013 and 2015. Among those who kept going for more than 300 days, 97% lost money, and only 0.4% earned more than a bank teller's wage. Results are after exchange and broker fees [S, FGV/SSRN abstract].

Europe and the UK: ESMA, the EU markets regulator, introduced limits on retail contracts for difference (CFDs, leveraged contracts that track a price without owning it) in 2018. Broker-published figures typically showed 74% to 89% of retail accounts losing money [S]. The UK regulator, the FCA, was reported to have found 82% of spread-betting and CFD clients losing in an older review [S, IBTimes], and it published a consultation paper (DP25/3) on CFD rules on 8 December 2025 [S].

• Costs: every trade pays a fee and a spread, and they add up. On one on-chain exchange, the published base taker fee is 0.045% of position value [P, Hyperliquid docs]. With 10,000 of position value, that is 4.50 to open and 4.50 to close, 9 per round trip. If your margin is 1,000 at 10x leverage, 100 round trips cost 900, or 90% of your margin, before you have made a single price mistake.
• Leverage: it makes small errors fatal. A 10% adverse move is a nuisance at 1x and a wipe-out at 10x.
• Behaviour: cutting winners early because it feels good, and holding losers because closing makes the loss real.
• House money: treating recent profits as free money and taking bigger risks with them. It is still your money.
• Revenge trading: trading bigger right after a loss to win it back, which usually produces a bigger loss.

Some participants do consistently well, but their edge is structural, not magical.

• Hedgers: a farmer or airline uses derivatives to reduce a risk they already have. They are not trying to profit from the derivative; they win by keeping the business steady.
• Market makers: firms that quote both a buy and a sell price and earn the spread between them, thousands of times a day. They have speed, scale and low fees, and aim to hold little net risk.
• Systematic traders: those who follow written rules, size small, and apply them consistently over years. Their edge is discipline and costs control, and even they have long losing stretches.
• Patient traders: people who trade rarely, wait for setups that fit their rules, and avoid the constant costs of overtrading.

Other everyday uses follow the same logic. A household with a variable-rate mortgage can ask a bank about fixing the rate, which is a rate derivative in disguise. A delivery company can lock in fuel: it uses futures to fix 3.00 per gallon on 100,000 gallons, a bill of 300,000. If the market goes to 3.50, it pays 350,000 for fuel but the futures gain 50,000, net 300,000. If the market falls to 2.60, it pays 260,000 and loses 40,000 on futures, net 300,000. The price is fixed either way (ignoring small real-world differences). Farmers do the same with crop prices.

1. Learn: understand margin, liquidation, funding and option pricing before risking money.
2. Paper trade: practise your rules on a simulated account.
3. Start small: use an amount you can afford to lose completely.
4. Write rules: entry, exit, size, maximum loss, and when to stop for the day.
5. Review: keep a journal, check costs, and be willing to conclude that trading is not for you.

Making money here honestly means earning more than the costs and the risks you took, over many months, after accounting for time and stress. The statistics above show that most people do not manage it. No result is promised by this course, and deciding not to trade is a perfectly good outcome. This is education, not advice, and it does not take your own situation into account.

• I can explain margin, liquidation price and funding in my own words.
• I can calculate position size from a stop distance without a calculator app.
• I could lose the whole amount without it affecting rent, debts or sleep.
• I have no emergency-fund or debt problem that this money should go to first.
• I have written rules, including a maximum loss per trade and per day.
• I have paper traded for long enough to see a losing streak.
• I know the total cost per trade, including fees and spread.
• I will not trade to recover a loss or while angry, bored or tired.
• I accept that I might decide not to trade at all.