โ Back to Blog ยท 2026-09-16 ยท 7 min read ยท Trading Education
You're long soda ash at the open. Within minutes, the market is moving against you in chunks of 20, 40, 60 yuan per tick sequence, and your stop โ sitting a comfortable 40 points below entry โ gets blown through in a single print. You're out 80 points instead of 40, and by the time you've processed what happened, the market has bounced. Sound familiar?
This is the reality of trading fast-moving Chinese commodity futures. Lithium carbonate and soda ash are two of the most volatile products on China's exchanges, and they punish traders who import stop-loss habits from slower markets without adapting them. This article is about building stops that actually work when the tape is moving faster than your fingers.
Why Fast Markets Break Normal Stop Rules
Most stop-loss advice assumes a market that trades continuously and fills you near your stop price. In fast Chinese commodity markets, three assumptions fail at once:
- Slippage is structural, not exceptional. When a policy headline hits or a margin hike is announced, order books thin out instantly. Your stop becomes a market order into a vacuum.
- Limit moves exist. Chinese exchanges apply daily price limits (roughly ยฑ4% to ยฑ8% on these products, adjusted by the exchange depending on conditions). If a market locks limit-down against you, your stop doesn't execute at all. You're along for the ride until the lock breaks.
- Volatility is regime-based. These products don't drift; they trend violently or chop. The 2021 thermal coal rally โ where prices roughly doubled before government intervention triggered a cascade of limit-down days โ is the canonical example. Anyone holding with a "normal" stop learned the lesson in one week.
Lithium carbonate gave a similar lesson more recently: after spiking to extreme highs in late 2022, it spent much of 2023 in a relentless decline. Traders who kept buying dips with tight stops got death by a thousand cuts. Traders with no stops at all got wiped out on the bounces.
The takeaway isn't "don't trade these markets." It's that your stop logic must be designed for the market you're actually in โ not the market you wish you were in.
Know the Contract Before You Place a Stop
You cannot size a stop sensibly without knowing what a point is worth and how far the market typically moves. Here's what you're dealing with:
| Contract | Exchange | Contract Size | Tick | Value per Tick |
|---|---|---|---|---|
| Lithium Carbonate (LC) | Guangzhou Futures Exchange (GFEX) | 1 ton per lot | Small tick increments on a high-priced product | Small per tick, but percentage swings are huge |
| Soda Ash (SA) | Zhengzhou Commodity Exchange (ZCE) | 20 tons per lot | 1 yuan/ton | 20 yuan per tick |
Two practical notes:
- Lithium carbonate's 1-ton multiplier is deceptive. One lot looks tiny, but when the underlying is trading in the tens of thousands of yuan per ton, a 2% day is a four-figure yuan swing per lot. Percentage risk is what matters, not lot size.
- Soda ash's 20-yuan tick adds up fast. A 100-point adverse move is 2,000 yuan per lot. If your account equity is, say, 50,000 yuan, that's 4% gone on one trade โ before slippage.
Also check the exchange's margin and limit settings before every session. Both exchanges have raised and lowered limits and margins on these products multiple times when volatility spiked. A stop placed based on last month's parameters may be invalid today.
Stop on Volatility, Not on Points
The single biggest upgrade most traders can make: replace fixed-point stops with volatility-based stops. The logic is simple โ your stop should sit just outside normal noise, so you're only stopped when the trade is actually wrong, not when the market is merely breathing.
The ATR method
- Calculate the Average True Range (ATR) on your trading timeframe โ 14 periods is standard.
- Place your stop at a multiple of ATR beyond your entry or beyond the nearest structural level. For fast markets like these, 1.5x to 2.5x ATR is a common working range. Tighter than 1x ATR and you're trading noise; wider than 3x and a single loss becomes an equity event.
- Size your position so that (stop distance ร contract value) equals a fixed small fraction of your account โ 1% is the classic ceiling, and in these markets many experienced traders use less.
Example: soda ash's ATR on the daily chart expands to 80 points during a trending phase. At 2x ATR, your stop is 160 points, or 3,200 yuan per lot. If that's more than 1% of your account, you trade fewer lots โ you don't shrink the stop to fit your position. This order of operations (stop first, size second) is where most retail accounts die.
Why this beats fixed ticks
A fixed 50-point stop means completely different things in a quiet range versus a policy-driven trend. In quiet conditions it's needlessly wide; in fast conditions it's inside the noise and you'll be stopped on the first pullback. ATR adapts automatically. It's not magic โ it's just measuring the market you're actually in.
Structural Stops: Use Levels, But Not the Obvious Ones
Volatility sizing tells you how far; structure tells you where. The best stops sit at prices that, if traded through, genuinely invalidate your idea.
- Behind swing points, with a buffer. If you're long off a support level, don't put your stop at the level โ put it beyond it by a fraction of ATR. In fast Chinese commodity markets, obvious levels get swept precisely because everyone's stops sit there.
- Respect the round-number sweep. Big figures attract resting orders. If your stop is at a psychologically obvious number, assume it's fuel.
- Invalidate, don't hope. The test for any structural stop: if price trades through it, is your original thesis still intact? If yes, the stop is in the wrong place. If no, the stop is doing its job even if it feels painful.
One warning specific to these products: structure matters less during policy-driven moves. When a government agency signals intervention in a commodity โ as happened with coal in 2021 โ technical levels become suggestions, not laws. During those windows, reduce size or step aside. No stop placement survives a limit-locked session.
Event Risk: The Stop You Can't Place
Chinese commodity futures have a distinctive feature global traders need to internalize: policy is a first-order price driver. NDRC statements, exchange margin adjustments, export policy changes, and industry capacity announcements can move lithium carbonate and soda ash several percent in minutes โ often at session open or in the night session.
You cannot stop-loss your way out of a gap. So manage it before it happens:
- Know the calendar. Chinese policy announcements, economic data releases, and exchange notices cluster around specific times. Trading straight into them with a full position is a choice, not an accident.
- Size for the gap, not the tick. Ask yourself: if this market opens 3% against me tomorrow, what does that cost me? If the answer makes you sweat, the position is too big regardless of where your stop sits.
- Respect the night session. These products trade at night, and some of the sharpest moves happen when day-session liquidity is absent. If you won't watch the night session, either flatten or cut size before it opens.
- Have a lock-limit plan. If a market locks at its daily limit against you, decide in advance: do you hold and hope for a reversal day, or exit at whatever price appears when the lock breaks? Deciding mid-crisis is how small losses become account events.
A Worked Rule Set You Can Steal
Here's how this all comes together for a practical soda ash or lithium carbonate setup:
- Define the stop before entry. Nearest structural level plus a 0.5xโ1x ATR buffer, with a floor and ceiling of 1.5xโ2.5x ATR from entry. If the two disagree wildly, skip the trade.
- Size to the stop. Position size = (risk budget, e.g. 1% of equity) รท (stop distance ร per-point contract value). Round down. Always down.
- Never widen a stop. In fast markets the temptation is constant. Widening a stop once rewires your discipline permanently. If you want more room, exit and re-enter smaller.
- Move stops only in your favor. Trail behind swing points or by ATR โ not tighter than 1x ATR of current price, or normal noise will take you out of a winner.
- Time-stop your trades. If a breakout trade hasn't worked within a few bars, the edge has likely decayed. Fast markets don't owe you patience.
- Stand down around known event windows unless your size is small enough that a limit move is survivable.
None of this guarantees wins โ nothing does. What it does is ensure that when you're wrong, you're wrong by a known, survivable amount, and that you live to take the next trade. In markets that can lock at their limit overnight, that's the whole game.
Test It Before You Trust It
Stop-loss rules look great on paper and fall apart the first time real slippage hits your fills. That's why testing against real market data โ real tick movement, real session dynamics, real Chinese commodity futures contracts โ matters more here than in most markets. If you want to pressure-test your stop logic on lithium carbonate, soda ash, or the products you already trade like iron ore and rebar, you can run your system through a China futures evaluation on real historical data at XS Select, with evaluations starting from $29. Whether you pass or not, the feedback on how your stops behave in fast tape is worth more than another hundred hours of backtesting in a spreadsheet.
Fast markets don't forgive, but they do reward preparation. Set the stop before the trade, size to the stop, and respect the gap. Everything else is noise.