← Back to Blog · 2026-10-08 · 8 min read · Trading Education

You've been there. The 15-minute chart shows a textbook breakout. You enter. Ten minutes later, price reverses hard and stops you out — and when you zoom out to the daily chart, you realize you just bought into a pullback inside a downtrend that's been running for three weeks. The setup wasn't wrong. Your timeframe was.

This is the single most common failure mode among retail traders, and it's especially punishing in Chinese commodity futures, where policy headlines can flip a trend overnight. The fix isn't a new indicator. It's a structure: trade with confluence across multiple timeframes, where the higher timeframe gives you direction, the middle timeframe gives you the setup, and the lower timeframe gives you the trigger. In this article, I'll walk through a concrete version of that method and apply it to two of the most liquid contracts in China — iron ore and rebar.

Why Multi-Timeframe Confluence Works (When Done Honestly)

The core idea is simple: different market participants operate on different horizons. A steel mill's procurement desk thinks in months. A prop desk scalps in minutes. Price at any moment is the collision of all of them. When the monthly, daily, and hourly structures all point the same way, you're trading with the largest pool of order flow behind you. When they conflict, you're gambling on which horizon wins.

Confluence trading doesn't improve your win rate on any single signal. What it does is filter out the signals that fight the dominant flow — the ones that tend to produce your biggest, most avoidable losses. Fewer trades, better average trade. That's the entire pitch.

The discipline part matters more than the theory. Most traders say they check higher timeframes. What they actually do is zoom out after they've already decided they want the trade, and rationalize whatever they see. The method below is designed to remove that discretion: each timeframe has one job, and if any layer fails its check, the trade doesn't happen. No exceptions, no "but this one looks special."

The Three-Layer Structure: Direction, Setup, Trigger

Use three timeframes with roughly a 4x to 6x gap between each layer. A clean, workable combination for Chinese commodity futures:

• Direction layer — Daily chart: defines the only two states you're allowed to trade: long bias or short bias. Everything else is "stand aside."
• Setup layer — 60-minute chart: identifies pullbacks or compression patterns aligned with the daily direction.
• Trigger layer — 15-minute chart: times the actual entry and places the stop.

Daily: Define Direction With Structure, Not Opinion

Pick one objective rule and stick to it. A common one: the daily bias is long when price is above a rising 20-day EMA and the last significant swing high has been broken; short when price is below a falling 20-day EMA and the last significant swing low has been broken. Anything else — sideways, EMAs flat, price chopping through the average — is no-trade. That third state is where most of your account protection lives.

You are not predicting here. You're classifying. The daily chart answers exactly one question: "Am I only allowed to look for longs, only shorts, or nothing at all today?"

60-Minute: Wait for the Pullback, Not the Chase

Once the daily gives you a bias, the 60-minute chart's job is to give you a location. In an uptrend, you want a pullback into a zone of prior resistance-turned-support, a prior consolidation shelf, or the 20-period EMA on the 60-minute chart. In a downtrend, you want a rally into resistance. You mark the zone in advance and you do nothing until price enters it.

This is where confluence gets teeth. A pullback that lands on two or more overlapping references — say, a prior 60-minute range high that also lines up with a daily swing level — is a confluence zone. One reference is a hint. Two stacked references are a setup.

15-Minute: The Trigger and the Stop

Inside the zone, drop to the 15-minute chart and wait for a shift in the short-term structure in your direction — for example, a break above the most recent 15-minute swing high after a higher low forms (for longs), or the mirror image for shorts. Your stop goes just beyond the swing point that invalidates the trigger, plus a small buffer for spread and slippage.

If price enters your zone and just bleeds sideways without ever triggering, you don't chase. The setup expired. That's a win, not a miss — you avoided a trade with no defined risk point.

Know Your Instruments: Iron Ore and Rebar Contract Specs

The method is generic; the execution details are not. If you're going to trade rebar or iron ore — the two contracts that best express China's construction and steel cycle — you need the specs burned into memory, because position sizing math depends on them.

Two practical implications. First, iron ore's tick value is five times rebar's per lot, and its notional per lot is roughly ten times larger — so a "one lot" position in each is a wildly different risk exposure. Size from tick value and stop distance, never from lot count. Second, both contracts trade a day session (9:00–11:30 and 13:30–15:00 Beijing time, with a short break mid-morning) plus a three-hour night session. If you're trading from Europe or the Americas, your evening often maps onto the night session — thin but real liquidity, and the session where global risk sentiment (think a crude oil shock or a US equity selloff) tends to hit first.

Always verify current contract specs, margin requirements, and price limit rules directly with the exchange or your broker before sizing a position — these parameters are adjusted periodically, especially around holidays and volatile periods.

Applying the Method to the Steel Complex

Iron ore and rebar are not independent markets. Rebar is the finished product; iron ore is a key input. They share a common driver — Chinese construction and infrastructure activity — and they're both acutely sensitive to policy. That makes the pair ideal for confluence trading, because you get a second, cross-market layer of confirmation.

Cross-Market Confirmation

Before taking a long rebar trigger, glance at iron ore's daily chart. If iron ore is also in a daily uptrend and holding its own pullback zone, your rebar long has the whole steel complex behind it. If iron ore is rolling over while rebar holds up, one of two things is happening: margins in the steel chain are compressing, or one market is leading the other. Either way, treat the disagreement as a caution flag — reduce size or skip the trade. Divergence between the two isn't automatically a signal, but it's never a green light.

Respect the Policy Layer

China's commodity markets have a documented history of policy-driven regime shifts. The 2021 thermal coal rally — a parabolic run that reversed violently after government intervention on prices — is the canonical example, and iron ore's own surge to record highs in 2021 followed by a sharp correction after supply-side and demand-side policy pressure is another. Nobody's indicator predicted those turning points. What multi-timeframe structure gives you is a response protocol: when a daily trend breaks against your position, the daily layer has flipped to no-trade or opposite bias, and you exit or stand down regardless of what your 15-minute chart says. The hierarchy only protects you if the higher timeframe actually outranks the lower one.

Session Timing for the Trigger Layer

Run your 15-minute triggers during the highest-liquidity windows: the first hour of the day session and the open of the night session. Avoid initiating in the last 15 minutes before session breaks or the close — spreads widen, and your stop placement assumption degrades. Pullback zones formed during the midday break often resolve violently at the 13:30 reopen; that reopen can itself be your trigger candle.

Risk Rules That Make the Method Survive

The structure is worthless without the risk layer. Non-negotiables:

• Fixed fractional risk: risk a fixed small percentage of the account per trade — commonly 0.5% to 1% for retail. Compute lot size as: (account × risk %) ÷ (stop distance in ticks × tick value). If the math gives you less than one lot, the trade is too big. Skip it.
• Stop at the invalidation point: the stop lives where the 15-minute structure fails, not at an arbitrary yuan amount. If that distance implies more than your risk budget in lots, cut the lots, not the stop.
• One bias per day: if the daily flips mid-session, you may exit, but don't immediately reverse. Let the new bias prove itself on the 60-minute chart first.
• Daily circuit breaker: two full stop-outs in a day ends your trading day. Confluence methods produce few setups; revenge trading destroys the edge the filter built.

A Worked Example Walkthrough

Here's the checklist in sequence, using rebar as the vehicle:

• Step 1 (Daily): Rebar is above a rising 20-day EMA and broke its last major swing high two weeks ago. Bias: long only.
• Step 2 (60-min): Price has pulled back for three sessions into a zone where a prior 60-minute consolidation shelf overlaps the daily breakout level. Mark the zone. Set an alert.
• Step 3 (Cross-check): Iron ore's daily is also in an uptrend and sitting on its own 60-minute demand zone. Confluence confirmed.
• Step 4 (15-min): During the day session, price forms a higher low, then breaks the last 15-minute swing high. Enter long on the break or on the retest.
• Step 5 (Stop & size): Stop below the 15-minute higher low, plus buffer. Run the position-size formula. If it yields one lot or fewer at your risk cap, trade it or pass.
• Step 6 (Manage): Trail behind 60-minute swing lows once the trade moves one unit of initial risk in your favor. Exit hard if the daily bias rule breaks — no hoping.

Six steps, every one of them checkable after the fact. That auditability is the real value: at the end of a month, you can review each trade and identify which layer failed — direction, location, trigger, or management — instead of shrugging at your P&L.

Test It Before You Trust It

No method, however disciplined, deserves your capital until it has survived contact with real market data — and Chinese commodity futures have enough structural quirks (session breaks, night-session gaps, policy sensitivity) that backtesting on Western data won't cut it. Paper-trade the checklist above for a few weeks on live iron ore and rebar charts, journal every step, and be honest about which layer you keep bending.

When you're ready to test under realistic conditions, you can run your system against real China futures market data through a structured evaluation at XS Select — evaluations start from $29, so the cost of finding out whether your process holds up is roughly the price of one bad iron ore tick sequence. No promises about outcomes; just a clean environment to prove your rules to yourself before the market does it for you.