Asset Classes¶
In principle, a good idea is asset-class agnostic: momentum is momentum whether the instrument is a stock, a bond future, or a perpetual swap. In practice, the choice of asset class decides most of your operating reality before you write a line of strategy code — how much leverage you get and on what terms, when you can trade, what execution costs, whether shorting is trivial or a daily negotiation, and whether decent data costs nothing or five figures a year. Professionals choose instruments the way engineers choose materials: by mechanical properties, not sentiment. This lesson walks the five classes you might realistically trade and ends with an honest ranking for the independent systematic trader.
The five arenas¶
Equities. A share is fractional ownership of a firm — no expiry, no roll, dividends along the way. US margin follows Regulation T: 50% initial (2:1 leverage), with 4:1 intraday available under pattern-day-trader rules ($25k account minimum). Regular hours run 9:30–16:00 ET bracketed by the open and close auctions, with thin pre/post-market sessions. Costs: commissions near zero at retail, spreads of 1–5 bps in liquid names, plus small regulatory fees — but shorting requires locating borrowable shares, and hard-to-borrow names carry fees from ~0.3% to over 100% annualized. Data: end-of-day is cheap or free; clean consolidated intraday tick data is a genuine expense; corporate actions (splits, dividends, mergers) make historical data maintenance a real chore, and survivorship bias in free datasets is a classic backtest killer.
Futures. Standardized contracts to buy or sell at a future date, traded on centralized exchanges (CME dominant), spanning equity indices, rates, energy, metals, and agriculture. Contracts expire, so a continuous position must be rolled — and every historical price series must be roll-adjusted, your first data hygiene test. Margin is a performance bond set by risk-based models: typically 3–12% of notional, so leverage of 8–30x is structurally available; positions are marked to market daily through the clearinghouse. Trading runs nearly 23 hours a day on Globex, five days a week. Costs: roughly $1–3 round-turn in commissions and fees against contract notionals of $50k–$250k — among the cheapest access in finance — with the caveat that tick sizes make minimum spreads meaningful for small moves. Data: one venue per contract means no fragmentation; exchange-quality historical data is affordable; shorting is symmetric with going long — no borrow, no locate.
Options. Rights, not obligations, to buy (calls) or sell (puts) at a strike by an expiry. The payoff is nonlinear, which changes everything: an option position is exposure to direction, volatility, and time simultaneously, summarized by the greeks. Margin ranges from "fully paid" (long options) to substantial and model-based (short options; portfolio margin for larger accounts). Hours track the underlying's regular session for most contracts. Costs are the trap: outside the top tickers and near-the-money strikes, spreads of 2–10% of premium are routine — per-trade costs an order of magnitude above the other classes. Data is the other trap: the US options tape (OPRA) is enormous — full chains across thousands of underlyings, each with dozens of strikes and expiries — and quality historical chains with greeks are priced for institutions.
FX. Spot currency trading is over-the-counter: no central exchange, no consolidated tape — your "market" is whatever prices your counterparties stream to you. The interbank core (EBS, LSEG Matching) is inaccessible at retail size; retail traders face dealers or aggregators, some of whom apply last look — the right to reject your trade after seeing it. Leverage: capped at 50:1 for majors at US retail brokers, higher offshore. Hours: 24 hours, five days a week, following the sun from Wellington to New York. Costs: spreads on majors are genuinely tight (fractions of a pip), but all-in costs are opaque — embedded in the spread, rejections, and swap points on overnight carry. Data: there is no "true" historical FX price, only some venue's or broker's; backtests inherit that ambiguity.
Crypto. Native digital assets trading 24/7/365 on exchanges that are simultaneously venue, broker, and custodian — which is exactly why counterparty risk here is a first-order concern (recall the missing clearinghouse from the previous lesson). The dominant instrument is the perpetual future: no expiry, tethered to spot by a funding rate exchanged between longs and shorts every few hours — a cost and a signal in its own right. Leverage of 10–100x is offered; almost no one survives using the upper end. Costs: maker/taker fees of roughly 1–10 bps plus funding; spreads tight on majors, wide on everything else. Data: the standout feature — full-depth order books and complete trade history stream free over public websocket and REST APIs. Nowhere else can an individual assemble institutional-grade microstructure data at zero cost.
Side by side¶
| Class | Contract mechanics | Margin & leverage | Trading hours | Typical costs | Data accessibility |
|---|---|---|---|---|---|
| Equities | Perpetual shares; dividends; corporate actions; shorting needs borrow | Reg T 2:1 (4:1 intraday) | 9:30–16:00 ET + auctions, thin extended hours | ~0 commission; 1–5 bp spreads; borrow fees on shorts | EOD cheap; clean intraday costly; survivorship traps |
| Futures | Standardized, expiring; must roll; symmetric short | Performance bond, ~3–12% notional (8–30x) | ~23h/day, 5 days | $1–3 round-turn on $50k–250k notional | Single venue; affordable and clean |
| Options | Nonlinear payoff; strikes and expiries; greeks | Long: premium; short: model-based, punitive | Underlying's regular session (mostly) | Spreads 2–10% of premium off the top names | OPRA is huge; quality chains priced institutionally |
| FX | OTC spot; dealer-streamed prices; last look | Up to 50:1 retail (US majors) | 24h, 5 days | Tight headline spreads; opaque all-in costs | No consolidated tape; broker-dependent history |
| Crypto | Spot + perpetuals with funding; exchange is also custodian | 10–100x offered; funding costs carry | 24/7/365 | 1–10 bp maker/taker + funding | Free full-depth books and trade history via API |
What should an independent systematic trader actually trade?¶
Rank the classes by what matters when you are the whole firm — research, execution, risk, and operations in one chair — and a clear order emerges.
Futures rank first. One venue per contract, so no fragmentation games; roll-adjusted data is a solved problem you can afford; costs are low, explicit, and honest; shorting is structurally identical to going long; and the margin system gives real capital efficiency inside a clearinghouse's protections. Liquid index, rate, and energy futures are where systematic trading is most nearly a pure test of whether your signal is real.
Crypto ranks a close second — for different reasons. The free full-depth data means you can do genuine microstructure research on a laptop; access is direct and programmatic from day one; the market is young enough that inefficiencies documented and arbitraged away decades ago in equities still surface. The offsets are severe and must be managed, not ignored: exchange counterparty risk in place of a clearinghouse, and a 24/7 market that never gives your operations a weekend.
Equities are workable but frictionful at independent scale: shorting frictions, fragmented and costly intraday data, and corporate-action bookkeeping consume time that a solo operation cannot spare. FX at retail means trading against your dealer's prices with no tape to check them — a structurally weak position for a data-driven trader. Options are a special case: real, persistent edges live there, but extracting them requires volatility modeling as a core competency, not a sideline. We treat options and volatility trading properly in the advanced volatility module later in the course; until then, treat option backtests that ignore spread costs as fiction.
Leverage is the common killer
Every class above will extend you more leverage than you should use — 4:1 intraday equities, 20x futures, 50:1 FX, 100x crypto perps. Leverage does not change a strategy's expected return per unit of risk; it scales both until the first bad week is the last one. A 20x levered position is finished by a 5% adverse move — an ordinary day in crypto and an ordinary week in futures. Professionals size positions from estimated risk and survivable drawdown, then check the margin allowed; amateurs size from the margin allowed. Part of this course's job is to make the first habit automatic before real money is involved.
Key takeaways
- Asset class choice fixes your leverage terms, hours, costs, shorting mechanics, and data reality before any strategy code exists.
- Futures offer the cleanest laboratory: single venue, honest low costs, symmetric shorting, affordable data — at the price of roll mechanics you must handle correctly.
- Crypto offers free institutional-grade market data and direct access, offset by exchange counterparty risk and a market that never closes.
- Equities carry shorting and data frictions at independent scale; retail FX means trading a dealer's prices without a consolidated tape.
- Options contain real edges but demand volatility expertise and impose per-trade costs an order of magnitude above other classes.
- Size positions from risk and survivable drawdown, never from the leverage a broker offers.
Where this goes next¶
You now know what markets are, who trades them, how the machinery works, and which instruments suit an independent operation. But markets do not behave the same way all the time — the same strategy that prints money in one environment bleeds in another. Lesson 6 introduces market regimes: trending, mean-reverting, high-volatility, and crisis conditions, and why every strategy is an implicit bet on which one you are in.