Part 1: How a candle is tested · Chapter 1 of 14 · 8 min

What a candlestick is and how to test it

A Japanese candlestick summarises a period in four prices. On H1 gold, in 9.7 months, the 29 formations of the book appear 2,547 times, as many as in a random series (2,526), and give −0.06 R: the name of a formation does not say what price does next.

One candle, four prices

A candle covers a period (here one hour, H1) and has four prices: open, high, low and close. The body is the distance between open and close; the wicks are the stretches above and below the body, up to the high and the low. A candle is bullish if it closes above the open, bearish if it closes below. The name comes from Japan.

Figure 1 shows a real candle: on 21 December 2025 at 05:00 gold opens at 4,337.77 $, touches a low of 4,333.00 $, a high of 4,343.99 $ and closes at 4,343.99 $. The body is 6.22 $, the lower wick 4.77 $ and the upper wick zero: the close coincides with the high.

Figure 1. A real H1 candle: open 4,337.77, low 4,333.00, high 4,343.99, close 4,343.99 (21 December 2025, 05:00).Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 20 Dec 2025 18:00 UTC to 21 Dec 2025 13:00 UTC.
Read the chart in words

H1 gold chart from 20 to 21 December 2025 with the four prices of the 05:00 candle marked on the candle.

WhenWhatPrice
21 Dec 2025 05:00 UTChigh 4343.994,343.99
21 Dec 2025 05:00 UTClow 4333.004,333.00
21 Dec 2025 05:00 UTCopen 4337.774,337.77
21 Dec 2025 05:00 UTCclose 4343.994,343.99

Times in UTC.

Sizes are measured in ATR

Is a body of 6 $ big or small? It depends on the volatility of the moment, which the book measures with the ATR (the average range of the last candles). All the rules are written in ATR: "body of at least 1 ATR", "lower wick of at least 60% of the range". On H1 gold the median body is 0.35 ATR (25th percentile 0.15, 75th percentile 0.68), the median range 0.88 ATR and the median ATR 15.59 $.

The data are 7,091 H1 candles from 15 December 2025 to 6 October 2026, of the gold perpetual future on Binance (XAUUSDT). 50% of the candles are bullish (3,553). The future is open 24 hours a day: 78% of candles open exactly at the close of the previous one. Formations that require a gap (like the "abandoned baby") do not exist here, and the book does not propose them.

How the book recognises a formation

A formation is recognised at the close of its last candle, with rules written in ATR and without looking at the future. The book uses 29: marubozu, doji (standard, dragonfly, gravestone), spinning top, hammer, hanging man, shooting star, inverted hammer, engulfing (bullish and bearish), harami, inside bar, piercing, dark cloud, tweezers, morning and evening star, three inside, three outside, three soldiers, three crows and three methods. Each has its own chapter.

For each formation the book makes three measures.

  1. How often it appears, against 100 random series built with the same candles in shuffled order (in blocks of 6 candles). If the number is the same, the formation describes common candles.
  2. What follows: the first threshold touched between +1.5 and −1.5 ATR within 10 candles, in the direction the theory expects or against. The control is any candle: 37% go up first, 36% go down, 27% touch neither threshold (7,050 candles).
  3. The trade: entry at the close of the last candle, stop beyond the extreme of the formation plus 0.25 ATR (risk between 0.5 and 4 ATR), target equal to the risk (1 R), exit after 24 candles, cost 0.30 $. The mean R is reported with the error (±, 95% interval) and the comparison with the random series.

The trade control is any candle with a random direction, a stop at 1.5 ATR and a target of 1 R: 7,036 trades, mean R −0.03 (the target arrives in 44% of cases, the stop in 45%). The error the book reports assumes independent trades; formations close in time overlap, so the true error is a little wider.

How many formations appear

The rules of the book find 2,547 formations in 9.7 months (263 a month). A random series with the same candles has a median of 2,526 (5th percentile 2,469, 95th percentile 2,580): the same number. In 29% of cases (2,040 candles out of 7,091) the candle is the last one of at least one formation. The valid trades are 2,346 (201 skipped because the risk is outside 0.5-4 ATR): 1,074 targets, 1,136 stops, 136 time exits, mean R −0.06 ± 0.04.

A formation is not rare and is not a signal in itself. The rest of the book measures, for each one, whether anything changes.

Same formation, two outcomes

Figure 2 shows a bullish engulfing of 21 December 2025: the 04:00 candle (open 4,343.58 $, close 4,337.88 $) is bearish and the 05:00 candle (open 4,337.77 $, close 4,343.99 $) engulfs it, after a fall of 1.6 ATR in the previous ten candles. Entry at 4,343.99 $, stop at 4,332.08 $ (11.91 $, 3.3 ATR), target at 4,355.90 $. The target is reached on 22 December at 01:00, twenty candles later: +0.97 R.

Figure 3 shows the same formation on 21 April 2026, at 06:00 and 07:00 (4,787.10 → 4,780.84 $ and 4,780.83 → 4,788.09 $), after a fall of 2.6 ATR. Entry at 4,788.09 $, stop at 4,773.11 $ (14.98 $, 1.3 ATR), target at 4,803.07 $. The stop is hit at 12:00: −1.02 R. One formation, two outcomes: the two cases say nothing on their own.

Figure 2. A bullish engulfing that works: entry at 4,343.99, stop at 4,332.08, target at 4,355.90 reached on 22 December 2025 at 01:00 (+0.97 R).Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 20 Dec 2025 20:00 UTC to 22 Dec 2025 07:00 UTC.
Read the chart in words

H1 gold chart from 20 to 22 December 2025 with the two numbered candles and the lines of entry, stop and target.

WhenWhatPrice
21 Dec 2025 04:00 UTCPoint 14,334.02
21 Dec 2025 05:00 UTCPoint 24,333.00
21 Dec 2025 05:00 UTCEntry4,343.99
21 Dec 2025 05:00 UTCStop4,332.08
21 Dec 2025 05:00 UTCTarget4,355.90
22 Dec 2025 01:00 UTCtarget reached4,383.45

Times in UTC.

Figure 3. The same formation failing: entry at 4,788.09, stop at 4,773.11 hit on 21 April 2026 at 12:00 (−1.02 R).Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 20 Apr 2026 22:00 UTC to 21 Apr 2026 18:00 UTC.
Read the chart in words

H1 gold chart of 21 April 2026 with the two numbered candles and the lines of entry, stop and target.

WhenWhatPrice
21 Apr 2026 06:00 UTCPoint 14,777.18
21 Apr 2026 07:00 UTCPoint 24,776.00
21 Apr 2026 07:00 UTCEntry4,788.09
21 Apr 2026 07:00 UTCStop4,773.11
21 Apr 2026 07:00 UTCTarget4,803.07
21 Apr 2026 12:00 UTCstop hit4,768.47

Times in UTC.

Procedure in four steps

  1. Write the rule of the formation in ATR before looking at the chart.
  2. Count how often it appears and compare with a random series.
  3. Measure what follows, with a control (any candle).
  4. If you trade, write the stop and the target and report the R with the error and the cost.

Exercises

Chapter exercises

1. On the chart the letters A-D are on four candles of 21 December 2025. Which letter is the candle that engulfs the body of the previous one?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 20 Dec 2025 20:00 UTC to 21 Dec 2025 13:00 UTC.
Read the chart in words
WhenWhatPrice
21 Dec 2025 00:00 UTCPoint A4,342.91
21 Dec 2025 04:00 UTCPoint B4,334.02
21 Dec 2025 05:00 UTCPoint C4,333.00
21 Dec 2025 08:00 UTCPoint D4,342.99

Times in UTC.

2. An H1 candle opens at 1,200, has a high of 1,230, a low of 1,190 and closes at 1,220 (hypothetical numbers). What are the body and the upper wick?

3. The rules of the book find 2,547 formations on gold and a random series with the same candles has a median of 2,526. What does it mean?

4. Why does the book not propose formations that require a gap, like the abandoned baby?

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.