The step-by-step market analysis process

Core idea

Open the chart and scan one timeframe at a time, from the largest down to the smallest, each timeframe only answering two questions: is it rising / falling / sideways and is the force strong or weak. Once you've finished scanning the large-timeframe cluster (1D→1M), stop and settle on an overall read + set your tactical mindset, and only then scan the sub-1D cluster to find which timeframe currently has a pattern. The whole process stops at "assess only as far as the market has actually shown itself" — no step involves guessing further ahead.

Details

Step 0 — The groundwork before you click anything

• Click through timeframes one at a time, don't rush. This is beginner mistake #1: skimming through the whole chain of timeframes in 30 seconds and drawing a conclusion. The order the teacher requires: 1M → 3W → 2W → 1W → 6D → 5D → 3D → 2D → 1D, then 16h → 12h → H8 → H6 → H4 → H2 → H1 → 45m → 30m → 15m → 5m.
• Write it on paper, by hand, while analyzing — don't type it (see tang bang chim trading).
• Your knowledge has to be memorized for you to be able to pull it out while analyzing. The chain the teacher uses: know → memorize → understand → apply fluently → create. Forgetting knowledge in the middle of an analysis is normal for someone who hasn't memorized it yet, not a sign of "being less smart."

Step 1 — Scan the LARGE-timeframe cluster (1D → 1M), settle on a read

For each timeframe, answer briefly: state (rising / falling / sideways) + force (strong or weak).

Once you've scanned the whole cluster, you must reach one of two conclusions:

• The timeframes agree → there's a large trend to follow, and you're allowed to expect a distant target.
• The timeframes are chaotically out of phase (1M against 1W, 1W against 1D/2D/3D...) → the market is broadly sideways, tactical mindset = CAUTIOUS, drop any expectation of a big trend entirely, and just go look for waves on the smaller timeframes.

Step 2 — Cross-check with a "reverse hypothesis"

Before dropping to a smaller timeframe, ask a checking question: "If price were to drop straight from the current zone down to zone X below, which timeframe's wave would that be?"

How to answer: scan for which timeframe currently has a downward pattern clean enough to produce a wave of that size. If only the 1W has a down pattern, while the 2W is still rising and the 1M is accumulating without a clean pattern yet → conclusion: that scenario has low probability, but if it does happen, you already know exactly which timeframe to focus on for evidence.

The value of this step: it turns a vague expectation into a conditional statement, while also pre-defining which timeframe to watch for each scenario. Related: Building a wave scenario.

Step 3 — Scan the SMALL-timeframe cluster, find the timeframe with a pattern

• Whichever timeframe has no trend, leave it alone — don't use it to make a decision. Two metaphors the teacher uses: if your neighbor doesn't buy land, that's their business; you buy with your own money and the price still rises anyway; in a vote, whoever has no opinion has to follow the majority who do.
• Whichever timeframe has a clear trend (RSI has a slope, the two moving averages diverge, and price confirms it) → that's the timeframe to follow.

You must read RSI TOGETHER WITH price. If RSI trends up while price goes sideways, that is not rising buying pressure — it's just selling pressure taking a break. That is not a "credible" trend to base a decision on. Buying pressure/selling pressure can't be seen directly; it's something inferred from two visible pieces of data: RSI and the price line.

Step 4 — For a timeframe with a pattern: ask "is there still force left to continue?"

This is the step done over and over the most. Say H4 currently has a downward pattern:

Open the timeframes smaller than H4 to see whether the downward force is still continuing. There are only two cases:

1. The smaller timeframe still has downward momentum → the H4 wave keeps falling quickly.
2. The smaller timeframe has fully reset its downward force (selling force is spent, but buying force isn't necessarily in yet) → rule out case 1; price will loiter around and the continued fall will take much longer.

Case 2 splits again: price either idles and then goes back to falling, or breaks upward. Don't predict ahead of time — wait for the evidence.

Step 5 — If you plan to follow that timeframe: set your psychology using the adjacent larger timeframe

To ride H4's uptrend wave, you have to check H12 (the adjacent larger timeframe) to see its current state:

• H12 has just turned down → your psychology should be CAUTIOUS, don't expect a distant target.
• Always establish the nearest achievable target = the first resistance zone above.

Step 6 — The STOPPING point of the process

The process stops at the nearest achievable target, no further. At that point:

• If it can keep climbing (the smaller timeframe still supports it) → hold, and roll your expectation up to a larger timeframe.
• If it can't climb any further → switch state: close, and go look for an entry point in the opposite direction.

The motto: "focus only as far as the evidence goes, assess only as far as the market has actually shown itself" — piecing it together one piece at a time, without forcing yourself to see the whole road ahead.

Going along with the market is NOT the same as being late

A common objection: "if I wait for a clear signal before entering, it's already too late." The teacher rejects this directly: go review past waves, and the moment when the timeframes separate out and give an agreeing signal is precisely the right moment. Entering at that moment gives you higher confidence and lower risk — that's what's actually valuable, not a few dozen points of an earlier entry.

When to STOP COMPLETELY and not analyze anymore

Distinguish two forms of sideways market (details in Identifying the range zone):

• The contested form — the timeframes are out of phase with each other but each still shows its own trend → there's still evidence, you can still ride waves to profit from. The cost: short, close, long, close, in and out constantly.
• The forceless form — RSI loitering around 40–60, price also trendless, across most timeframes → sit still. A metaphor the teacher uses: when it's neither raining nor sunny, don't force it. The difficulty here isn't technical, it's discipline.

Also: during major news events (e.g. the Fed announcing rates), price whips up and down very hard — anyone who hasn't mastered the smaller timeframes should stay out, since getting stopped out is nearly certain, sometimes even a very painful fake-out.

Putting it into practice

A handwritten template for every time you open the chart:

[Large cluster 1D→1M]
1M ___ (strong/weak) | 1W ___ | 3D ___ | 2D ___ | 1D ___
→ Overall read: agree ___ / chaotically out of phase

→ Tactical mindset: EXPECT / CAUTIOUS

[Reverse hypothesis] If price returns to zone ___, that's a wave of timeframe ___ → high/low probability because ___

[Small cluster <1D]
12h ___ | H8 ___ | H6 ___ | H4 ___ | H2 ___ | H1 ___ | 45m ___ | 30m ___ | 15m ___ | 5m ___
→ Timeframe currently with a pattern: ___
→ The timeframe below it: still has force / already reset?
→ The adjacent larger timeframe: allows it / out of phase?

[Conclusion] Nearest achievable target ___ | At that point: hold if ___ / reverse if ___

What the teacher stresses / common traps

• The "which timeframe looks nicer" trap. The teacher deliberately baits students by asking: "The 12h's down pattern is nicer than the 1D's up pattern, right?" — students nod along. Wrong. When two timeframes are contested, you must not lean toward either side; whichever one actually happens is the one to focus on. The bigger lesson: one offhand remark, one other person's trade in the group, is enough to bend your own conviction in a direction you're not even aware of. That is exactly the source of "unnecessary" stop-outs.
• Formlessness defeats form. The less trend the market has, the more you should avoid forcing your mind to predict ahead of it. Put yourself inside the market to observe and understand what story it's telling, instead of sitting there arguing whether the 12h or the 1D wins.
• Sideways doesn't mean there are no pullback waves. If H4 is sideways, there's still a 15m/30m structure inside it forming waves — the pullback wave belongs to the smaller timeframe. Saying "sideways means no pullback wave" is a way of thinking that's missing multi-timeframe awareness.
• The BEGINNING of a new trend usually has NO pullback wave. From a sideways zone, price just gradually drips down, the small timeframe has no downward momentum, but it forms progressively lower highs, progressively lower lows, that don't get broken — that is the footprint of gradually rising selling pressure from the small timeframe. At that point you enter following the footprint, without waiting for a clean stop-loss point (there won't be one). Only from the next leg onward is there a pullback wave to enter on cleanly.
• Candles and candlestick patterns are only a reference. A candle that looks "strong" on a small timeframe means nothing when viewed on a larger one. Likewise, a resistance level created by the 30m/45m timeframe only holds as long as the 30m/45m doesn't break — if H4 has upward pulling force, that level is meaningless. Before borrowing outside knowledge, use the system you're already learning well; ultimately everything comes back to buying pressure – selling pressure.
• Starting from session 39 the teacher changed the format: no more model-answer lectures, students do the analysis themselves for the teacher to dissect. The reason: only when told "you analyzed this part well, but your thinking on this part was wrong" does a student see their own gap. "I'd rather encourage you to get it wrong on the training ground than bleed out in the actual market."
• Two people at the same time, one long, one short, can both be right — as long as each one stays consistent within their own timeframe frame of reference.

Don't draw the up/down table when there's no pattern yet

An error corrected live in the session (student Huy's case): laying out a whole row of timeframes — 1M, 1W, 1D, H4, H1... — into two up/down columns, while not a single timeframe currently has a pattern about to trigger. That table has no value for entering a trade: once the analysis is done, you still just sit still, because it doesn't lead to any decision.

The rule: only draw the up/down table when the timeframe you want to enter on currently has a pattern. Only then does the table have something to do — checking which side the surrounding evidence leans toward, to decide whether or not to follow that pattern.

Three steps in the correct order (compiled by student Thao, confirmed by the teacher as the standard version):

1. An overview from large timeframe → small timeframe, identifying where the wave currently sits — which wave of the next-larger timeframe each large timeframe is currently sitting in.
2. Find the timeframe with a pattern / a trend / a point worth noting → focus on exactly that timeframe, and check for agreement/disagreement in the surrounding timeframes. This step should be done by redrawing the chart by hand to get a feel for the price.
3. Only now draw the up/down table, and only for the timeframe you're preparing to enter on.

The structural mistake in doing it backward: the state table is a decision-making tool, not a market-description tool. Using it to "get a general sense of things" wastes time while also creating the illusion of having analyzed carefully.

Closely related: RSI is also a final check step, not a first analysis step — you have to read the story of the price first, and only then use RSI to confirm it. Agreement/disagreement is the layer that does the re-checking. (See The wave's fuel tank (RSI = fuel tank): RSI rising while price goes sideways = selling pressure taking a break, not rising buying pressure.)

Related lessons

• Tactical mindset — the process to determine before entering a trade — the 5-step process for setting your mindset; this process is the "scanning the chart" step that precedes it
• Building a wave scenario — the next step: turning a read into a conditional scenario
• Identifying the range zone — distinguishing the two forms of sideways to know when to sit still
• Multi-timeframe (multi-timeframe logic) — the theoretical foundation for scanning timeframe clusters
• Wave-nature theory (permission from the higher timeframe, backing from the lower timeframe) — "permission from the large timeframe, support from the small timeframe," used in steps 4–5
• Don't predict the market — "to make money, you don't need to know what happens next" — why the process stops at the nearest target
• Combining knowledge with tactics — why knowing still isn't enough to stop you losing — why knowing the process still doesn't mean you can use it
• Entry point — the final step after the analysis is done