Part 1: Understanding the order block · Chapter 1 of 10 · 10 min

What an order block is and what the data says

The rule GoForexHub's engine uses to find an order block, why many traders think it works, and what 149 real gold zones show: how often price comes back, how often the zone holds and how long it lasts.

The idea in one sentence

An order block is the last opposite candle before a strong move. If price rises decisively, the last bearish candle before the rise is the demand order block. If price falls decisively, the last bullish candle before the drop is the supply order block. The zone is its range, from low to high.

Book 1 gave you this rule in short form (chapter 7). Here we take it as the subject of study: how it forms, how you read it, when it works and when it does not.

The engine's rule

The engine keeps at most two zones per side and ranks them with a few criteria: whether the zone sits in the right half of the range (demand in discount, supply in premium), whether it is recent, whether price has already visited it, whether it coincides with an FVG, and how far it is from price. The next chapters take these criteria apart one by one.

Why people think it works

The most common explanation goes like this. Big operators cannot buy or sell everything at once: the order is too large for the market. They execute part of it, move price and leave orders waiting. When price returns to the zone it started from, those orders fill and the move restarts.

It is a hypothesis, and from prices alone it cannot be checked. We cannot see the orders, and other explanations lead to the same chart: a strong candle marks where one side had been aggressive, and price tends to pass back through there for plain retracement reasons. In this book the hypothesis helps you understand the rule. The measurements decide whether the rule works.

What the data says

The engine showed 149 order blocks from January to October 2026: 68 demand and 81 supply, 54 of them with an overlapping FVG. For each one we looked at what happened in the 40 H4 candles after the displacement.

  • 114 (77%) were visited: a candle came back to touch the zone. The visit comes quickly, with a median of 2 candles (8 hours) after the displacement. The other 35 were not touched in 40 candles.
  • After the visit we look at which event comes first in the 10 candles after the visit candle: a reaction of at least 1.5 ATR in your favour, measured from the extreme the visit reached, or a close beyond the far side of the zone. If the candle that touches the zone already closes beyond, that is a failure and it is counted. The visit candle alone does not count as a reaction.
  • 62 (54%) hold, 44 (39%) give way, 8 do neither.
  • As a benchmark we rebuilt, on every H4 candle in the history, a fake zone with the same height and the same depth of penetration as each of the 114 real visits (198,360 fake visits) and applied the same measure: 44% hold, 40% give way.

With 114 cases the margin of error on the 54% is about 9 points. The hold rate is 10 points above the benchmark (54% against 44%), the failures are identical (39% against 40%): an uncertain advantage, and a 10-point gap with a 9-point margin of error does not settle the question. The sample is short (ten months), comes from one instrument, and the engine picks zones with criteria that do not look at the future, but which are its own.

A second figure counts as much as the first. Of the 62 zones that hold, 29 (47%) still close beyond the far side within 30 candles of the visit (five days). "Holds" means "price reacted by 1.5 ATR before turning back". It does not mean "held forever" and it does not mean "you would have made money". An order block is a zone that is good for a few days.

Two examples, one that holds and one that gives way

The first order block in the sample is from 2 January (supply, from $4,378.00 to 4,397.23) and it gives way at the first visit: on 5 January at 00:00 price enters and four hours later a candle closes above. For a case that holds we start from the fifth one.

The zone that holds (figure 1, H4, 10 to 15 January 2026). On 12 January at 08:00 there is a bearish candle with a low of $4,578.32 and a high of 4,596.98. The 12:00 candle is the displacement: it has a body of 37.01, or 2.07 ATR. At 20:00 price comes back into the zone, with a low of 4,581.00, and closes at 4,600.86, above it. In the next ten candles the high reaches 4,638.99. On 16 January at 12:00 a candle closes below the zone: the zone has had its day.

Figure 1. A demand order block that holds: origin, a 2.07 ATR displacement, first visit and restart.Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 10 Jan 2026 16:00 UTC to 15 Jan 2026 04:00 UTC.
Read the chart in words

H4 gold chart from 10 to 15 January 2026. A blue band between 4,578.32 and 4,596.98 dollars starts at the last bearish candle before a strong bullish candle. Labels mark the origin, the displacement and the first visit, after which price climbs to 4,638.99.

WhenWhatPrice
12 Jan 2026 08:00 UTCDemand order block4,578.32 – 4,596.98
12 Jan 2026 08:00 UTCorigin4,578.32
12 Jan 2026 12:00 UTCdisplacement4,624.99
12 Jan 2026 20:00 UTCfirst visit4,581.00

Times in UTC.

The zone that gives way (figure 2, H4, 19 to 23 January 2026). On 21 January at 08:00 a bullish candle with a low of $4,850.46 and a high of 4,877.40 precedes a bearish displacement (the 16:00 candle, a body of 1.59 ATR). On 22 January at 12:00 price climbs into the zone as far as 4,871.65 and closes at 4,868.77, still inside. Four hours later a candle closes at 4,911.16, above the far side: the zone is invalidated.

Figure 2. A supply order block that gives way: price enters the zone and four hours later closes beyond the far side.Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 31 Dec 2025 16:00 UTC to 6 Jan 2026 08:00 UTC.
Read the chart in words

H4 gold chart from 19 to 23 January 2026. A grey band between 4,850.46 and 4,877.40 dollars starts at the last bullish candle before a bearish displacement candle. Price returns into the band, labelled first visit, and a later candle closes above the band, invalidating it.

WhenWhatPrice
2 Jan 2026 08:00 UTCSupply order block4,378.00 – 4,397.23
2 Jan 2026 08:00 UTCorigin4,397.23
2 Jan 2026 12:00 UTCdisplacement4,331.68
5 Jan 2026 00:00 UTCfirst visit4,417.01

Times in UTC.

What this book will do

The next chapters follow the zone from beginning to end. First the ingredients (structure, displacement, liquidity), the anatomy and the validity criteria. Then how to mark and refine it, and the six types: reversal, continuation, breaker, mitigation, rejection and propulsion. Then entry, stop and target, confluence with FVG, OTE and killzones, multiple timeframes and failures. At the end, five gold setups with measured results, a complete example, the mistakes and the glossary. Every percentage is measured on the engine's 149 zones.

Exercises

Chapter exercises

1. On the chart candle C is a strong bearish displacement. Which letter is the origin of the supply order block, the last bullish candle before the displacement?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 4 hours (H4) · from 14 Feb 2026 12:00 UTC to 18 Feb 2026 00:00 UTC.
Read the chart in words
WhenWhatPrice
15 Feb 2026 00:00 UTCPoint A5,039.78
16 Feb 2026 04:00 UTCPoint B5,009.96
16 Feb 2026 12:00 UTCPoint C5,007.89
16 Feb 2026 16:00 UTCPoint D4,999.78

Times in UTC.

2. The ATR is 20 dollars and a candle has a body of 24 dollars. Under the engine's rule (1.3 ATR), is it a displacement?

3. In the sample, 62 visited zones out of 114 (54%) reacted by 1.5 ATR before giving way, against 44% for a no-zone benchmark of the same height and depth, with a margin of error of about 9 points. What do you conclude?

4. What does it mean that a zone "holds" in the book's measurements?

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.