Oil trading risk management comes down to four controls: size every position from a fixed risk budget (1–2% of capital per trade), place stops using volatility rather than round numbers, respect margin as a leverage warning rather than a suggestion, and cap total exposure across correlated energy positions. Master those four and most of the ways oil traders blow up become unavailable to you.
That matters because oil is a market where the tail events are not hypothetical. Crude has printed a negative price, doubled inside a year, and dropped 30% in a month — all within the last decade. The traders still standing after each of those episodes were not the ones with the best forecasts. They were the ones whose sizing meant a wrong forecast was survivable. Our complete guide to crude oil trading covers how this market moves; this article covers how to make sure it can’t take you off the board.
Fair warning: nothing here is exotic. Risk management is arithmetic plus obedience. The arithmetic takes an afternoon to learn. The obedience is a career.
Start with the Only Number That Matters: Dollars per Point
Every risk decision in energy futures flows from contract math, so put it somewhere you can see it:
| Contract | Size | Tick | $1.00 move is worth | Routine daily range |
|---|---|---|---|---|
| WTI Crude (CL) | 1,000 bbl | $0.01 = $10 | $1,000 | $1.50–$3.00 |
| Micro WTI (MCL) | 100 bbl | $0.01 = $1 | $100 | same, /10 in dollars |
| Henry Hub Gas (NG) | 10,000 MMBtu | $0.001 = $10 | $10,000 | $0.10–$0.30 |
| Micro Gas (MNG) | 1,000 MMBtu | $0.001 = $1 | $1,000 | same, /10 in dollars |
Read the NG row twice. A natural gas contract is worth $10,000 per full dollar of price movement, and gas can move ten or twenty cents in a session without anyone considering it news. A single NG contract in a $25,000 account is not a position; it is a coin flip on the weather. More gas traders die of contract size than of bad analysis — they never did the multiplication.
Everything else in this article is built on this table. If you take one habit away: before any trade, write down what a $1 adverse move costs you (or $0.25 in gas), and check the number against your stomach and your account. If either flinches, the position is too big.
Position Sizing: The 1–2% Rule Done Properly
The rule itself is standard: risk no more than 1–2% of account equity on any single idea. What separates professionals from tourists is the order of operations. Amateurs pick a position size, then find a stop they can afford. Professionals find the technically correct stop first, then let it dictate size:
- Risk budget: account equity × risk percentage. A $50,000 account at 1.5% = $750 per trade.
- Stop distance: set by the market — beyond the invalidation level, scaled by current volatility (more below). Say $1.80 on a CL swing trade.
- Dollar risk per contract: $1.80 × $1,000 = $1,800 for full-size CL; $180 for a micro.
- Size: $750 ÷ $1,800 = 0.4 contracts. You cannot trade 0.4 CL contracts, and rounding up to one doubles your intended risk. The correct trade is four MCL micros ($720 of risk) — or no trade.
Notice what just happened: the arithmetic vetoed the full-size contract. That veto is the entire value of the process, and it is exactly the step most retail traders skip. Micros exist precisely so the math can be obeyed; the walkthrough on how to start trading oil covers choosing contract sizes and brokers for smaller accounts.
Why so small? Because losing streaks are a statistical certainty, not a sign of failure. A strategy that wins 50% of the time will produce a streak of seven straight losses roughly once every hundred trades. At 2% risk per trade that streak costs about 13% of the account — annoying. At 10% per trade it costs half the account, and now you need a 100% return just to get back to even. The table below is the most persuasive argument in trading:
| Drawdown | Gain needed to recover |
|---|---|
| 10% | 11% |
| 25% | 33% |
| 50% | 100% |
| 75% | 300% |
Risk management is not about avoiding losses. It is about keeping every loss in the top row.
Stops That Survive Oil’s Volatility
A stop-loss in crude has to thread a needle: close enough to cap the damage, far enough that ordinary noise cannot reach it. Oil is unusually good at punishing stops placed by feel.
Volatility-scaled placement
The standard tool is Average True Range. Take the 14-day ATR — call it $2.20 on CL — and place the stop 1× to 1.5× ATR beyond the level that invalidates your idea, not beyond your entry. If support sits at $76.40 and ATR is $2.20, a stop at $76.20 is inside the market’s daily breathing room and will be collected by noise; $74.20–$75.00 is a stop with a chance of only triggering when you are actually wrong. Yes, the wider stop forces a smaller position. That is the system working, not a flaw in it. (For how ATR, levels and chart structure fit together, see our piece on crude oil technical analysis.)
Where crude hunts
Round numbers and obvious swing lows are where resting stop orders cluster, and crude has a documented habit of probing through those levels, filling the stops, and reversing. You cannot prevent this; you can refuse to be easy prey. Place stops beyond the cluster (below $74.80, not at $75.00), or use time-based confirmation — exit on a 30-minute or session close beyond your level rather than on a touch. The second approach costs more when you are genuinely wrong but avoids the classic wick-out.
Stop-market versus stop-limit, and the gap problem
A stop-market order guarantees an exit but not a price; a stop-limit guarantees a price but not an exit. In crude, take the fill. A stop-limit that goes unfilled while the market trades through it converts a planned $700 loss into an unmanaged position in a moving market — the worst of all outcomes. The related, harder truth: no stop protects you from a gap. Oil reprices discontinuously on weekend geopolitics and surprise OPEC+ decisions; a Sunday-night open $4 through your stop fills you $4 through your stop. Which is why the last line of defense is never the stop — it is the position size that made a $4 gap survivable, and, for positions held over known event risk, options structures whose maximum loss is fixed in advance (covered in our guide to crude oil options trading strategies).
Margin Is a Leverage Meter, Not a Budget
Futures margin confuses newcomers into catastrophic sizing, so let’s be precise. Initial margin on one CL contract runs in the mid-single-digit thousands of dollars (it floats with volatility; CME’s margin schedule is public and worth checking, since it changes with market stress). The contract itself controls roughly $70,000–$90,000 of crude at recent price levels. That is leverage in the vicinity of 12–15:1 if you put on one contract per margin-minimum of capital — which is exactly what “how many contracts can I afford?” thinking produces.
The professional reframe: margin tells you the exchange’s estimate of a bad day, not your allowance. Clearinghouses size initial margin to cover something like a rough single-day move at high confidence; when margins jump — as they did across energy in spring 2020 and again in 2022 — the exchange is telling you, in dollars, that the distribution has gotten fatter. Treat rising margin requirements as a volatility signal and cut size accordingly, not as an inconvenience to finance with a bigger wire transfer.
And the margin call rule that should be tattooed somewhere visible: a margin call is not a request for more money, it is the market informing you that your position was too big yesterday. Answering a margin call by depositing funds to sustain a losing oil position is how modest losses become account obituaries. Cut the position. The mechanics of maintenance margin and daily settlement are covered in our primer on how to trade crude oil futures.
The Correlation Trap: One Trade Wearing Five Tickets
Position-level discipline fails silently at the portfolio level. Long WTI, long Brent, long RBOB gasoline, short a refiner hedge, long an oil-heavy equity index — that is not five diversified positions; on a demand-shock day it is one large short-the-recession ticket, and it all loses together. Crude benchmarks routinely correlate above 0.9 with each other and strongly with the products; even Henry Hub, which trades its own weather-driven fundamentals most of the year, joins the complex in a macro liquidation.
Practical portfolio rules that survive contact with real books:
- Bucket by driver, not by ticker. “Global demand,” “U.S. supply,” “gas weather” — positions in the same bucket share a risk budget. Two positions with the same driver at 1% each is one position at 2%; label it honestly.
- Cap total energy-complex risk at some multiple of your per-trade budget — many desks use 3–5×. Fully deployed is fully exposed.
- Watch the spread positions. Calendar spreads and crack spreads margin cheaply because they are usually tame — until the relationship itself breaks. Spread traders carry more notional per margin dollar than anyone, which is precisely how the biggest energy blowups happened.
Event Risk: The Calendar Is Half Your Risk Model
Oil’s volatility is not evenly distributed; it clusters around a public schedule, which means a serious risk process treats the calendar as an input, not trivia.
- Wednesdays 10:30 a.m. ET: the EIA petroleum status report. Thursdays 10:30: the EIA natural gas storage report. Both can print an hour of normal range in two minutes.
- OPEC+ meetings: scheduled, but with unscheduled emotional range — the April 2023 surprise cut gapped crude more than $5 at the Sunday open.
- Contract expiry and the roll: liquidity thins and mechanical flows dominate. April 20, 2020 was, at its core, an expiry-mechanics event.
- Weekends: 60+ hours of headline exposure with no exit. Middle East escalations have repeatedly repriced crude between Friday’s settle and Sunday’s open.
Concrete rules that fall out of this: size positions held through data releases and weekends as if the stop does not exist (because functionally it may not); never carry a thin-liquidity front-month position into its final sessions without a reason; and if you would not hold the position through a $5 gap, either cut it Friday afternoon or convert the risk to defined-loss options before the close. None of this is timidity — plenty of professionals deliberately trade the events. They just do it at sizes chosen for the event’s distribution, not the average Tuesday’s.
Case Studies: What the Blowups Actually Teach
Amaranth, September 2006 — liquidity and concentration
Amaranth Advisors lost roughly $6 billion — about two-thirds of the fund — in weeks, on natural gas calendar spreads (long winter, short non-winter months). The trade had worked spectacularly the prior year. The lesson is not “spreads are dangerous”; it is that Amaranth’s positions had grown to a size that was the market in some months, so when the spread moved against them there was no one to sell to at any reasonable price. Position size must be set against market liquidity, not just account equity — if your exit would move the price, you are not carrying a position, you are cornered by one.
April 20, 2020 — the impossible price
May WTI settling at −$37.63 vaporized accounts that were “risk-managed” by any conventional definition — small longs, stops in place. Stops mean nothing when the market goes no-bid, and several retail platforms could not even represent negative prices. The people who lost least had simply followed two boring rules: don’t hold expiring front-month contracts you can’t deliver against, and don’t assume any price level — including zero — is a floor. Tail events in commodities are wilder than equity intuition allows.
Winter Storm Uri, February 2021 — the other tail
When Texas froze, spot gas at some hubs traded from around $3 to hundreds of dollars per MMBtu, and firms on the wrong side of fixed-price obligations failed outright. Futures traders short NG “because it always comes back down” met a market that went up 100× at the physical level. Energy shorts carry unbounded tail risk in a way that longs do not; selling energy into a supply crisis on mean-reversion faith is the single most reliably fatal trade in the complex.
Three different mechanisms — liquidity, settlement mechanics, physical scarcity. One common denominator: in every case, position size relative to the tail, not analytical error, determined who survived.
Daily Loss Limits and the Human Failure Mode
Every professional desk enforces a daily stop on the trader, not just the trades, because the most dangerous instrument in any account is a human being who is down money and wants it back. The pattern is depressingly uniform: a loss, then a bigger position to recover it faster, then a bigger loss — revenge trading is compounding run in reverse. A workable amateur version of the desk rule:
- Daily loss limit: 2–3× your per-trade risk. Hit it and the platform closes for the day. Not after one more trade — closes.
- Weekly circuit breaker: down 5–6% on the week, flatten everything and take two days off. The market will still be there; the pattern in your head won’t be.
- Size-up discipline: increase risk per trade only after a defined sample of trades (say 40–50) at the current size with positive results — never mid-streak in either direction. Winning streaks inflate size before losing streaks arrive; that sequencing has ended more trading careers than any single bad trade.
Write the limits down when you are calm. The whole point of a rule is that it outranks you at the moment you least want to follow it.
Options as Pre-Paid Stops: When to Convert the Risk
A stop order and a long put do the same job — cap the downside — with opposite failure modes. The stop is free but breaks exactly when you need it most: gaps, no-bid tape, exchange halts. The put costs premium but cannot be jumped over; its maximum loss is written on the ticket. A rational risk process uses both, chosen by conditions.
The conversion logic, with numbers: you are long two CL contracts from $75 with stops at $72.50 — $5,000 of intended risk. An OPEC+ meeting lands this weekend. Your stop is worthless against a Sunday gap; a $6 adverse open costs $12,000, not $5,000. Alternatives: flatten Friday (cost: forgone upside if the meeting is bullish), or buy two $73 weekly puts at, say, $0.55 ($1,100 total). Now the weekend worst case is roughly $5,100 all-in — the gap risk is gone, paid for with about a fifth of the intended risk budget. Whether that price is worth it depends on the event: routine EIA weeks, mostly no; genuinely binary OPEC+ meetings or live geopolitical escalation, frequently yes. The point is that it is a calculation, not a mood.
Producers make the same decision at industrial scale — collars and floors on future production are stop-losses on revenue, structured so that no overnight headline can jump them. If you find yourself repeatedly paying to convert stop risk into option risk, that is also information: your underlying position size is probably too big for the tape you are trading in.
Risk of Ruin: The Math That Sets the Percentages
The 1–2% convention is not folklore; it falls out of unforgiving arithmetic. Your long-run survival depends on three numbers: win rate, average win relative to average loss (payoff ratio), and fraction of capital risked per trade. The first two define your expectancy — say a 45% win rate with winners 1.8× losers gives you a positive edge of roughly 0.26R per trade, meaning each trade earns about a quarter of its risked amount on average. Positive expectancy makes profit possible. The third number decides whether you live long enough to collect it.
Risk 2% per trade with that profile and a catastrophic drawdown is a remote possibility; risk 10% and the losing streaks that any 45% win rate will generate — seven or eight straight losers, several times per thousand trades — take enough of the account that the recovery math from the drawdown table starts working against you, and most humans abandon the strategy at precisely the wrong moment anyway. Bet-sizing formulas like Kelly make the same point from the other direction: even when the formula suggests double-digit position sizes on paper, practitioners bet a quarter or half of it, because your edge estimate is noisy and in oil the loss distribution has tails your backtest never met. April 2020 was not in anyone’s sample until it was.
Two practical corollaries. First, expectancy is unknowable without data, which is why new strategies earn micro-sized capital until they have a track record. Second, a small positive edge risked small compounds astonishingly well — 0.25R per trade at two trades a week is roughly 25R a year, which at 1.5% risk is a return most funds would take — while the same edge risked big has a meaningful probability of not being around for year two.
Measure Yourself: The Trade Log Is a Risk Instrument
You cannot manage a risk you do not measure, and the risk most traders never measure is their own behavior. The fix costs one spreadsheet. For every trade, log entry, stop, size, R risked, exit, R realized, and two diagnostic numbers: maximum adverse excursion (how far the trade went against you before resolving) and whether any rule got bent. Review monthly. The patterns that emerge are rarely flattering and always actionable:
- MAE tells you about your stops. If winners routinely draw down 0.8R before working, your stops are one wobble from correct entries — either entries are early or stops are tight. If losers hit full -1R almost immediately, entries are chasing.
- Realized-R distribution tells you about discipline. Losses clustering at -1.3R when the plan says -1R means slippage or hesitation is quietly taxing every mistake. Winners clustering at +0.5R against a +2R plan means profits are being snatched out of fear.
- Rule-bend frequency is the leading indicator. Blowups are trailed by weeks of small violations that all “worked out.” The log catches the drift while it is still cheap.
Desks impose this measurement from outside with risk officers and end-of-day reports. Independent traders must be their own risk department, and the spreadsheet is the department’s only employee. It takes four minutes a day, and it is worth more than any indicator in the previous article.
A Worked Example: One Trade Through the Full Risk Process
Pulling it together with real numbers on a hypothetical swing trade.
Account: $40,000. Per-trade risk at 1.5%: $600. Daily loss limit: $1,500. Total energy-complex cap: $2,400 of open risk.
The idea: WTI has pulled back to a well-tested shelf at $74.50 in a market whose curve is in backwardation; thesis is trend resumption. Entry $75.10. Invalidation: a settle below the shelf. Current 14-day ATR: $2.00, so the stop goes 1× ATR under the shelf at $72.50 — $2.60 of risk from entry.
Sizing: full-size CL risks $2,600 — more than four times the budget; instantly vetoed. Micros risk $260 each, so two MCL ($520) fit inside $600. Notional exposure: about $15,000, or 0.4× account equity — conservative leverage, which is what surviving looks like.
Pre-flight checks: Is EIA within the holding period? Yes — Wednesday sits mid-trade; position must be gap-tolerable, and at $520 of stop-risk with a worst-case $5 gap costing $1,000, it is (barely — note it). Weekend hold? Only if the trade is working. Correlated book? A short natural gas position is a different driver bucket; fine.
Management: stop to breakeven after the first settle above $77.10 (1× initial risk in profit); half off at $78.50; trail the rest under swing lows. If the Wednesday number craters the market to the stop: down $520, which is 1.3% of the account — a Tuesday, not a tragedy. Log it, run the next trade at unchanged size.
The dull genius of the process is that the exciting question — will oil go up? — had almost nothing to do with any decision except direction. Everything else was arithmetic that would have been identical on a losing week or a winning one.
Common Oil Trading Risk Management Mistakes
A checklist of the classics. None of these are rare, none are subtle in hindsight, and every one of them appears in the post-mortem of somebody’s account every single quarter. If two or more feel uncomfortably familiar, fix the process before the market runs the audit for you.
- Sizing by margin instead of by stop distance. “I can afford three contracts” is a leverage statement, not a risk plan.
- One NG contract in a small account. $10,000 per dollar of movement. Do the multiplication before the weather model does it for you.
- Stops at round numbers and obvious swing points. You are placing orders exactly where the market goes fishing.
- Averaging down in a trending commodity. Oil trends longer than intuition expects; 2014–15 fell for eighteen months. Adding to losers in crude is fighting the market’s strongest habit.
- Treating correlated positions as diversification. WTI plus Brent plus gasoline is one bet at triple size.
- Financing margin calls. The call is the risk report. Cut, don’t fund.
- Removing the stop “just this once” into a data release. The one time it goes wrong pays for every time it didn’t.
- Risking real size on an unproven strategy. Fifty trades on micros costs a few hundred dollars of spread and teaches you your actual edge. The same lesson at full size costs a third of an account.
The Risk Framework on One Page
| Control | Rule of thumb | What it protects against |
|---|---|---|
| Per-trade risk | 1–2% of equity, stop-derived sizing | Losing streaks, single bad ideas |
| Stop placement | 1–1.5× ATR beyond invalidation; stop-market, not stop-limit | Noise exits, unfilled exits |
| Leverage | Notional ≤ 1–2× equity for position trades | Gap risk, margin spirals |
| Portfolio cap | 3–5× per-trade budget across the complex, bucketed by driver | Correlation, hidden concentration |
| Event rules | Gap-tolerable size through EIA/OPEC/weekends | Discontinuous repricing |
| Trader limits | Daily 2–3× trade risk; weekly 5–6% circuit breaker | Revenge trading, tilt |
| Tail respect | No naked energy shorts you can’t wear at 3× the move; no expiring front months | Uri, April 2020, the next one |
Survival Is the Strategy
Ask veteran energy traders what their edge is and a surprising number give a version of the same answer: still being here. Oil rewards longevity — the cycles keep handing opportunities to whoever kept their capital through the last one. Sizing, stops, margin discipline and event rules are not the boring part of trading crude; they are the mechanism by which you get to keep playing a game with this much tail. Build the framework before you build the view, pressure-test it against the case studies above, and then go learn the market itself with our crude oil trading guide — with the quiet confidence of someone the market cannot remove in a single bad week.
