Part 4: Trading it · Chapter 13 of 14 · 9 min

Analysis routine and risk management

A seven-step routine applied to a real gold day, written with what was known at 08:00 UTC and compared with what happened next. Then lot size, R, costs, losing streaks and recovering from a loss, with real numbers.

The routine in seven steps

  1. Choose and state the timeframes (D1, H4, H1) and the time convention (here, UTC day).
  2. D1: write down the sequence of significant swings (chapter 2). Uptrend, downtrend or range?
  3. H4: write down the latest swings, the latest break (BOS or CHoCH, chapter 3) and the state: trend, range or a picture in doubt.
  4. Levels and zones: support and resistance with their touches (chapter 5), day levels (chapter 6), supply and demand zones (chapter 7). The EMA 50 only if the picture is a clean trend.
  5. Liquidity: equal highs and lows and obvious levels near price (chapter 9).
  6. Scenarios: write two scenarios, A and B, each with its trigger condition and its invalidation level.
  7. Risk: stop, target in R, lot size, the day's loss limit. Then you update the journal.

Step 6 is the one many people skip. Writing down in advance what would make you change your mind is the only way not to change it mid-trade.

A real case: 22 September 2026 at 08:00 UTC

Below, the analysis is written with only the candles closed at 08:00 UTC. Then we look at what happened.

D1. Falling significant highs: $4,700.63 (25 August), 4,514.81 (3 September), 4,405.75 (18 September). Falling lows: 4,318.03 (14 August), 4,288.00 (2 September), 4,243.81 (16 September). Yesterday's close is 4,372.76. D1 is in a downtrend.

H4. On 18 September at 04:00 a bullish CHoCH breaks 4,380.00: price tried to turn. Then it came back down. The latest swings are a high at 4,388.25 (21 September, 00:00) and a low at 4,327.11 (21 September, 12:00). At 08:00 the H4 candle closes at 4,305.70, below that low. A bullish CHoCH inside a bearish D1, with price coming back: the picture is in doubt.

Levels. A 4-touch resistance between 4,380.00 and 4,388.25. Yesterday's low 4,327.11, yesterday's high 4,388.25, Asian low 4,301.82. EMA 50 at 4,358.59: not used, because the picture is not a clean trend.

Scenarios.

  • A: an H4 candle closes above 4,327.11 after making a low below the Asian low. That is a sweep with a return. Invalidation: a close below the sweep's low.
  • B: an H4 candle closes below 4,295.00, the 14 September low. The downtrend continues.

Decision at 08:00. No entry. Neither scenario is active, and doing anything now would be guessing.

Figure 1. The 22 September 2026 analysis at 08:00 UTC, with only the candles already closed.Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 11 Sept 2026 16:00 UTC to 22 Sept 2026 04:00 UTC.
Read the chart in words

H4 gold chart from 11 to 22 September 2026, up to 08:00 UTC. A grey band between 4,380.00 and 4,388.25 dollars is the four-touch resistance. Two dashed lines mark yesterday's low at 4,327.11 and the Asian low at 4,301.82. A blue line is the EMA 50. A pink line marks the 18 September bullish CHoCH. A box sums up: price below yesterday's low, D1 downtrend, H4 in doubt, no entry.

WhenWhatPrice
16 Sept 2026 16:00 UTCResistance · 4 touches4,380.00 – 4,388.25
22 Sept 2026 00:00 UTCPrevious day low4,327.11
22 Sept 2026 00:00 UTCAsian low4,301.82
11 Sept 2026 16:00 UTCEMA 504,400.61 → 4,358.59
16 Sept 2026 16:00 UTCCHoCH4,380.00

Times in UTC.

What happened next

The 08:00 H4 candle drops to $4,295.56, below the Asian low of 4,301.82, and closes at 4,327.02, at yesterday's level. The 12:00 candle closes at 4,334.99, above 4,327.11: scenario A is triggered. Price rises to 4,373.53 (16:00 candle) and then falls. On 23 September the 12:00 candle closes at 4,285.88, below 4,295.00: scenario B is triggered, and scenario A is invalidated, because the close sits below the sweep's low (4,295.56).

A long entry at the trigger of A, with the stop below the sweep's low, would have been stopped out on 23 September. An analysis that says "no entry at 08:00" plus two written conditions was more useful than a forecast. Both conditions triggered, one after the other, in under two days.

Figure 2. The same candles with the 8 that follow the analysis. Scenario A triggers on 22 September at 12:00 and is invalidated by scenario B on 23 September at 12:00.Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 11 Sept 2026 16:00 UTC to 23 Sept 2026 12:00 UTC.
Read the chart in words

H4 gold chart from 11 to 23 September 2026. A grey band on the right marks the eight candles after the analysis. Price drops below the Asian low at 4,301.82, rises back above 4,327.11, reaches 4,373.53 and then falls to close at 4,285.88 on 23 September.

WhenWhatPrice
22 Sept 2026 08:00 UTC → 23 Sept 2026 12:00 UTCafter the analysis
16 Sept 2026 16:00 UTCResistance · 4 touches4,380.00 – 4,388.25
22 Sept 2026 00:00 UTCPrevious day low4,327.11
22 Sept 2026 00:00 UTCAsian low4,301.82
11 Sept 2026 16:00 UTCEMA 504,400.61 → 4,358.59
16 Sept 2026 16:00 UTCCHoCH4,380.00
22 Sept 2026 08:00 UTCbelow the Asian low4,295.56

Times in UTC.

Risk, in numbers

Lot size. On gold 1 lot is 100 ounces: 1 dollar of movement is worth 100 dollars on one lot. The formula: lots = money at risk ÷ (stop distance in dollars × 100). With a 10,000-dollar account and a 1% risk, that is 100 dollars, and a 15.78-dollar stop (the first trade of setup 3, chapter 12): 100 ÷ (15.78 × 100) = 0.063 lots, rounded down to 0.06. The real risk is 0.06 × 100 × 15.78 = 94.68 dollars.

The cost in R. A fixed cost weighs more with short stops. With a 0.30-dollar cost: a 15.78-dollar stop costs 0.02R, a 5-dollar stop costs 0.06R, a 40-dollar stop costs 0.0075R. Over many trades the difference adds up.

Losing streaks. Even with good rules, stops arrive in a row. A simulation of 100 trades, repeated 20,000 times, gives the longest run of consecutive stops:

Targets hitLongest streak (median)Loss at 1% per trade90th percentileLoss
25%1211.4%1816.5%
33%98.6%1413.1%
40%76.8%1110.5%
50%65.9%87.7%

At 33% targets, a streak of 9 stops is normal and one of 14 comes once in ten. If you risk 5% per trade, 9 stops is almost -37% of the account. Decide the risk per trade from the streak you can bear.

Recovering a loss. After a loss of d% you need a gain of d ÷ (100 - d) to get back to even. After -10% you need +11.1%, after -20% +25%, after -50% +100%.

The risk and psychology course develops these themes, with risk of ruin and daily limits.

The day's limit and the journal

Two simple rules. A daily loss limit (for example 2R): once it is hit, the day is over. A journal with date, setup, entry, stop, target, result in R and one line on what did not work. The percentages in chapters 7, 9, 11 and 12 refer to one sample and one instrument. Your journal is the sample that counts for you.

Exercises

Chapter exercises

1. At 08:00 UTC price is below yesterday's low, D1 is bearish, H4 has a bullish CHoCH against D1 and price has come back under. What do you write in the plan?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 11 Sept 2026 16:00 UTC to 22 Sept 2026 04:00 UTC.
Read the chart in words
WhenWhatPrice
16 Sept 2026 16:00 UTCResistance · 4 touches4,380.00 – 4,388.25
22 Sept 2026 00:00 UTCPrevious day low4,327.11
22 Sept 2026 00:00 UTCAsian low4,301.82
11 Sept 2026 16:00 UTCEMA 504,400.61 → 4,358.59
16 Sept 2026 16:00 UTCCHoCH4,380.00

Times in UTC.

2. An account with a 100-dollar risk per trade and a stop 20 dollars away. How many lots of gold (1 lot = 100 ounces) do you open?

3. With 33% of targets hit, over 100 trades the longest streak of stops has a median of 9 and a 90th percentile of 14. What do you plan for?

4. You lose 20% of your account. By what percentage of the remaining account must you gain to get back to the starting level?

Have a question or want to share your exercise?

Post in the community, or join the free signals room on Telegram.

Educational content, not financial advice. Trading involves risk.