Part 3: Trading it · Chapter 11 of 14 · 11 min

Order blocks and multiple timeframes: D1 for context, H1 for confirmation

How to use three timeframes around an H4 order block (context on D1, the zone on H4, confirmation on H1), with the measurements: D1 added little, and the H1 confirmation entry gave a positive result that does not depend on the zone.

The top-down method

Multi-timeframe says: look at the context on a high timeframe, mark the zone on a middle one, enter on a low one. For the book:

The test rules are those of chapter 9: risk between 0.5 and 3 ATR, stop before target on the same candle, exit after 24 H4 candles (96 H1), cost 0.30 dollars per ounce. One order per origin (130).

The D1 context

With the chapter 1 rule on D1 candles we find 17 order blocks from December to October. Figure 1 shows one, on H4, from 12 to 25 June 2026: it runs from $4,025.67 to 4,231.25 (205 dollars tall), born on 10 June, and is valid until 25 June at 00:00, when a D1 close goes beyond it. Two H4 demand order blocks overlap it. The first (14 June, from 4,220.50 to 4,235.20) is visited on 17 June at 16:00 and holds. The second (21 June, from 4,139.16 to 4,169.71) is visited on 23 June at 00:00 and four hours later a candle closes below. The same D1 context, two opposite outcomes.

Figure 1. A D1 order block with two H4 demand order blocks inside it: the first holds, the second gives way four hours after the visit.Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 12 Jun 2026 00:00 UTC to 25 Jun 2026 08:00 UTC.
Read the chart in words

H4 gold chart from 12 to 25 June 2026. A very tall grey band between 4,025.67 and 4,231.25 dollars is a D1 order block. Two smaller blue bands within it are H4 demand order blocks: the upper one holds, the lower one is crossed by a candle that closes below.

WhenWhatPrice
12 Jun 2026 00:00 UTCD1 order block4,025.67 – 4,231.25
14 Jun 2026 12:00 UTCH4: holds4,220.50 – 4,235.20
21 Jun 2026 20:00 UTCH4: gives way4,139.16 – 4,169.71

Times in UTC.

The numbers on all zones:

GroupZonesVisitedHoldGive way
H4 inside a live D1 order block14127 (58%)3 (25%)
H4 with no D1 order block13510255 (54%)41 (40%)
Right half of the D1 range775934 (58%)19 (32%)
Wrong half of the D1 range725528 (51%)25 (45%)

With 12 visits in the first group the margin of error is 28 points: nothing can be said about the D1 order block. The D1 range half gives a 7-point gap with margins of 13. In chapter 4 the D1 trend had shown no effect (52% with the trend aligned, 59% against).

In R (limit order at the near edge, 2R, chapter 9 rules, 87 trades in all, −0.20 ± 0.14) there is a result that stands out:

GroupTradesMean R
Inside a D1 order block8+0.11 ± 0.55
With no D1 order block79−0.23 ± 0.15
Right half of the D1 range44+0.08 ± 0.22
Wrong half of the D1 range43−0.48 ± 0.18

The difference between right and wrong half is 0.56 R, about two standard errors. But in this book we tried more than thirty different splits of the same sample: if you try that many, it is normal for one to reach two standard errors by chance. It is treated as a hypothesis to check on new data.

The confirmation on H1

The test has three steps. After the zone's visit on H1, you wait up to 48 H1 candles for the first break of structure in your favour. When it comes, you enter at its close. The stop sits beyond the extreme price touched between the visit and the confirmation. Of the 130 origins, 43 are invalidated at the first touch or never touched, 47 give no confirmation, 19 have a risk above 3 ATR: 21 trades remain.

Figure 2 shows a successful trade, on H1, on 29 and 30 June 2026, on the H4 supply zone from $4,051.01 to 4,078.54. Price touches it on 29 June at 12:00. At 16:00 a candle closes at 4,027.01, below 4,027.50 (the swing low of 09:00): that is the confirmation, and the entry. The stop is at 4,057.89, above the touch's extreme, with a risk of 30.88 dollars (2.35 H4 ATR, the H4 ATR is 13.2 dollars). The 2R target, 3,965.25, is reached on 30 June at 01:00: +1.99R.

Figure 2. An entry with H1 confirmation: price touches the supply zone, a candle closes below the H1 swing low (confirmation and entry) and the 2R target is reached.Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 29 Jun 2026 02:00 UTC to 30 Jun 2026 07:00 UTC.
Read the chart in words

H1 gold chart for 29 and 30 June 2026. A grey band between 4,051.01 and 4,078.54 dollars is an H4 supply order block. A line at 4,027.50 marks the H1 swing low broken at the close by a candle, marked as confirmation and entry. A dashed pink line at 4,057.89 is the stop, a blue one at 3,965.25 the target, reached a few hours later.

WhenWhatPrice
29 Jun 2026 02:00 UTCH4 supply order block4,051.01 – 4,078.54
29 Jun 2026 09:00 UTCH1 structure4,027.50
29 Jun 2026 16:00 UTCStop4,057.89
29 Jun 2026 16:00 UTCTarget 2R3,965.25
29 Jun 2026 12:00 UTCfirst touch4,054.60
29 Jun 2026 16:00 UTCconfirmation, entry4,021.89
30 Jun 2026 01:00 UTCtarget hit3,948.18

Times in UTC.

Figure 3 shows a loss, on H1, on 25 August 2026, on the H4 supply zone from $4,651.57 to 4,687.99. The touch is at 06:00, the confirmation at 13:00 (a candle closes at 4,619.33, below the swing at 4,623.51), the stop is at 4,663.44 and gets hit at 19:00: −1.01R. With the chapter 9 limit order the same zone, sold at 4,651.57 on 25 August at 04:00, had reached the target. No method wins every time.

Figure 3. An entry with confirmation that hits the stop: after the H1 swing breaks, price climbs back and hits the stop above the zone.Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 24 Aug 2026 20:00 UTC to 26 Aug 2026 01:00 UTC.
Read the chart in words

H1 gold chart for 25 August 2026. A grey band between 4,651.57 and 4,687.99 dollars is an H4 supply order block. A candle closes below the H1 swing low at 4,623.51 and marks the confirmation. A few hours later price climbs back to the dashed pink line of the stop at 4,663.44.

WhenWhatPrice
24 Aug 2026 20:00 UTCH4 supply order block4,651.57 – 4,687.99
25 Aug 2026 08:00 UTCH1 structure4,623.51
25 Aug 2026 13:00 UTCStop4,663.44
25 Aug 2026 13:00 UTCTarget 2R4,531.12
25 Aug 2026 06:00 UTCfirst touch4,656.76
25 Aug 2026 13:00 UTCconfirmation, entry4,607.14
25 Aug 2026 19:00 UTCstop hit4,694.53

Times in UTC.

What the data says about the confirmation

The 21 trades, with the stop beyond the touch's extreme:

TargetTradesTargets hitStopsMean R
1R2111 (52%)10+0.04 ± 0.22
2R218 (38%)11+0.35 ± 0.32
3R215 (24%)13+0.26 ± 0.38

They are positive, unlike everything we saw in chapter 9 (−0.20 ± 0.14 with the limit order), but with 21 trades the errors are 0.22-0.38 R. To know whether the zone or the method counts you need the controls.

  1. Fake zones. The same method applied to fake zones (same distance from price and same height, on every H4 candle: 217,360 zones, 48,649 trades) gives +0.14, +0.27 and +0.32 R for 1, 2 and 3R. Real zones (+0.35 with 2R) do no better.
  2. Random entry. With the same stop and target mechanics, but entering after a random number of H1 candles from the touch, the result is −0.02 R: the mechanics alone give nothing.
  3. Periods. The +0.27 repeats in both halves of the history (+0.28 and +0.27).
  4. Side. On demand zones (buys) the method gives +0.04 R, against −0.10 for the random entry. On supply zones (sells) it gives +0.57 R, against +0.06. Almost all the result comes from the sells, in the months when gold fell from the $5,625.34 high (28 January) to lows under 4,000 (3,948.18 on 30 June).

Read together: the H1 confirmation method gave a positive result in the sample, but the zone has nothing to do with it (same result on fake zones), and the result depends on the period, because sells in a falling market did almost all the work. The fake trades overlap in time, so the written standard errors (0.004-0.007) are too small: the true uncertainty is much wider. Ten months of one instrument are not enough to say the method works.

A four-step routine

  1. Mark the D1 context: whether the H4 zone sits inside a live same-side D1 order block and in which half of the D1 range. Note it, without using it as a filter until you have checked it on your own data.
  2. Mark the H4 zone with the chapter 1-5 rules and wait for the first touch on H1.
  3. Write the confirmation rule before looking: the first H1 break of structure in your favour after the touch, within a maximum time, with the stop beyond the touch's extreme.
  4. Measure in R, with a fake-zone comparison and a check on periods and sides. If the result does not depend on the zone, the credit goes to the method.

Exercises

Chapter exercises

1. On the H1 chart four candles are marked with a letter. Which is the confirmation candle, meaning the first to close below the last H1 swing low after the supply zone's touch?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 29 Jun 2026 02:00 UTC to 30 Jun 2026 07:00 UTC.
Read the chart in words
WhenWhatPrice
29 Jun 2026 08:00 UTCPoint A4,068.78
29 Jun 2026 12:00 UTCPoint B4,054.60
29 Jun 2026 16:00 UTCPoint C4,031.55
30 Jun 2026 01:00 UTCPoint D3,990.79

Times in UTC.

2. The H1 confirmation entry gave +0.35 R on 21 trades, and the same method on fake zones +0.27 R. What do you conclude?

3. The null control (random entry after the touch, same stop and target) gave −0.02 R. What is it for?

4. The method gave +0.57 R on sells and +0.04 R on buys, in a period when gold fell from $5,625.34 to under 4,000. What do you conclude?

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.