Part 4: Checking and deciding · Chapter 14 of 14 · 17 min

A ten-month account, mistakes, reference, glossary and final test

The plan of chapter 13 applied for ten months to LA and LD with one position at a time and 1% risk: without rules the 10,000 $ account closes at 8,640 $ (drawdown 27.1%), with the plan at 9,235 $ (20.3%), because the system has no edge (−0.02 ± 0.08 R) and the plan limits the loss without avoiding it. Then the most frequent mistakes in chapter order, the quick reference with formulas and measurements, the glossary and a final test of 15 questions.

A 10,000 $ account over ten months

Let us put everything together. A trader follows LA and LD (the two systems on levels) with one position at a time (chapter 11), 1% risk on current capital (chapter 5) and the model plan of chapter 13: daily limit −2 R, risk halved at −10% from the high and stop at −20%. It is 606 trades from 15 December 2025 to 5 October 2026. The series has mean R −0.02 ± 0.08, deviation 1.00 R, 50% of trades winning, a drawdown of 29.7 R and a losing streak of 6.

MonthTrades (plan)R of the month (no rules)Capital without rulesCapital with the planPlan status
December 2025 (from the 15th)29 (27)−1.698.3%98.3%
January58 (56)+9.1107.3%107.4%
February52 (45)+3.0110.3%106.6%7 skipped by the daily limit
March60 (59)+5.0115.7%107.0%risk halved
April69 (64)−6.2108.4%102.1%halved
May66 (58)+3.0111.3%101.6%halved
June58 (55)−10.999.5%96.7%halved
July72 (66)−0.398.8%95.5%halved
August65 (63)−0.697.9%94.2%halved
September68 (12)−5.092.9%92.3%stopped from 8 September (−20.3% from the high)
October (to the 5th)9 (0)−7.286.4%92.3%stopped

(Capital as a percentage of the initial one, at the end of the month.) Without rules the account ends at 8,640 $ (−13.6%) with a maximum drawdown of 27.1%. With the plan it ends at 9,235 $ (−7.7%) with a drawdown of 20.3%.

What this account says.

  1. The system has no edge, and the account shows it. Mean R −0.02 ± 0.08: the error contains zero. The estimated Kelly is −1.9% (from the lower limit −9.9%): no fraction to risk. To demonstrate an edge of 0.1 R you would need about 778 trades (chapter 12), and the trader has 606.
  2. The plan did its job, which is not to earn. It limited the loss (−7.7% instead of −13.6%) and stopped the account on 8 September, when the drawdown touched −20%: without rules capital goes from 97.9% at the end of August to 86.4% on 5 October, and the plan did not go through that.
  3. The plan had a cost. In February and March the capital without rules stood 4 and 9 points higher (110.3% and 115.7% against 106.6% and 107.0%): the rules took a little away and the daily limit skipped winning trades along with losing ones.
  4. The month-by-month result swings a lot: from +9.1 R to −10.9 R with the same system, as in chapter 2.
Figure 1. R of each month of LA and LD with one position at a time, marked at the end of the month (December 2025 - October 2026): −1.6, +9.1, +3.0, +5.0, −6.2, +3.0, −10.9, −0.3, −0.6, −5.0, −7.2 (October only up to the 5th). The sum is −11.6 R.Data: gold, XAUUSDT perpetual future (real candles) · daily (D1) · from 15 Dec 2025 to 5 Oct 2026.
Read the chart in words

Daily gold chart with, below, eleven bars of the monthly R of LA and LD together, with a line at 0.

A book on risk that closes with a losing account is consistent with what it measured: the systems we worked on have not demonstrated an edge, and a risk plan does not create one. It serves to stay in the game long enough to find out whether there is one.

The most common mistakes

MistakeChapter
Calling "risk" the possibility of losing and not the loss accepted before entering; comparing trades in dollars instead of in R; judging by the winning share1
Judging a system from 100 trades; stopping when the account is in profit; changing the rules after a bad block2
Thinking that after five losses "a winner is due"; quitting after six losses when they are normal; doubling after losses3
Trusting the maximum drawdown of the first months; believing −30% is recovered with +30%4
Doubling after a loss (martingale); keeping the same ounces when the account halves5
Computing ruin only with the hoped-for edge; believing 5% is small because "after a loss I risk less"6
Computing Kelly on 30 trades; using full Kelly7
Always buying the same ounces with different stops; choosing the stop to make the size work; believing the stop guarantees the price8
Choosing the maximum leverage "to have room to manoeuvre"; adding to a losing position9
Evaluating a narrow-stop system without costs; using the ideal cost for all trades10
Counting three 1% positions as a 1% risk; following two systems on the same instrument to be safer11
Concluding after 30 trades; checking every day and stopping as soon as it is positive12
Writing rules and ignoring them; changing them with the account in the red; expecting the plan to earn more13

Quick reference: the formulas

ConceptFormulaCh.
R1 R = distance between entry and stop; result = gain ÷ risk1
Expected valuep × G − (1 − p) × L (G, L in R)1
Error on the mean±1.96 × σ ÷ √n2
Longest losing streakk such that n × p × (1 − p)^k ≈ 1 (p = winning share)3
Recoverygain needed = 1 ÷ (1 − loss) − 14
Drawdown in % from R1 − e^(−risk × R)4
Ouncescapital × risk % ÷ stop in $ per ounce (round down)5, 8
Growth with fixed fraction≈ f × μ − f² × σ² / 2 per trade5
Ruin at ±1 R((1 − p) / p)^N, N = threshold ÷ risk6
Kellyf* ≈ μ / σ²; p − (1 − p)/b; use 1/4 to 1/27
Leverage usedrisk % ÷ stop as % of price9
Forced closeadverse move = 1/L − threshold (with a 5% margin and close at 50%: 1/L − 2.5%)9
Cost in Rcost in $ ÷ stop in $10
Risk of N positions√(N × (1 + (N − 1) × ρ)) × risk of one11
Trades for an edge μ(1.96 σ / μ)² for error = mean; 7.85 σ² / μ² for power 80%12

Quick reference: the measurements

Ch.Result
1LA −0.03 ± 0.08 R (552 trades), LD +0.03 ± 0.10 (367), SD +0.09 ± 0.15 (55), deviation about 1 R (0.56 for SD); over 100 trades the interval is about ±19 R
2Error ±0.19 at 100 trades, ±0.08 at 552; 100 LA trades from −24 to +11 R; bootstrap at 100 trades: positive 37% (LA)
3Longest losing streak 7 (LA), 6 (LD), 4 (SD); at 50% winning: ≥ 5 in a row in 81% of 100-trade paths; after 3 losses the winning share does not change
4Maximum drawdown 35.3 R (LA), 18.9 R (LD), 1.7 R (SD), 9.0 R (SX); from 12 R at 100 trades to 33 R at 500 for LA; 20 R at 2% = −33%
5At 1% risk fixed size and fixed fraction coincide; martingale: 72% of LA paths wiped out or below 5%
6At p = 50% and 2% risk: losing 50% in 26.4% (500 trades) and 42.9% (1,000); with an edge it stays put
7Kelly: LA 0%, LD 2.9% (−7% / +13%), SX 21%; at full Kelly median drawdown 94%, at half 67.5%, at a quarter 39.8%
8Median stop 21.7 $ (0.48%): 4.6 ounces with 10,000 $ at 1%; widest hourly candle 391.5 $; 2 stopped trades out of 276 with the candle opening beyond the stop (−2.58 R)
9Median leverage used 2.1× at 1%, 10.4× at 5%; at 20× 2.5% against is enough (11.5% of days)
10Mean cost 0.015 R (8.5 R over 552); LD breaks even at 0.87 $, SD at 8.1 $; with a 5 $ stop and 1 $ cost you need 60% winning
1162% of LA trades overlap with LD; same direction: correlation +0.41; LA + LD: drawdown 42.7 R; worst day −5.1 R
12785 trades for μ = 0.1 R (power 80%); 14 months at the pace of LA; check every 25 trades: false positive 16%
13Drawdown steps: LA at 2% from 62% to 99%; daily limit 80% against 81%; SX at 2%: 724% → 572%

All the measurements come from ten months of a single instrument (December 2025 - October 2026, gold perpetual future on Binance, prices only), from series of results already built in books 9 and 10 and from a declared synthetic example (SX). None demonstrates an operational edge.

Glossary

The terms of the book are in the site glossary, where each has an example with numbers.

TermDefinition
Risk of ruinThe probability that the account touches at least once a loss you cannot afford (for example −30% or −50%) within a certain number of trades.
Kelly criterionThe fraction of capital to risk to make the account grow as fast as possible in the long run: about mean divided by variance of the results in R (μ/σ²), or p − (1 − p)/b.
Fixed fractional sizingRisking on every trade the same percentage of current capital: after a loss you risk less, after a gain more.
MartingaleA sizing rule that doubles the stake after every loss to recover with the first win.
Losing streakA stretch of consecutive trades, all losing.
Drawdown recoveryThe percentage gain needed to return to the high after a percentage loss: 1/(1 − loss) − 1.
Bootstrap of resultsDrawing at random, with replacement, the results of a series to obtain many possible paths and see how much the series depends on the order.
Statistical powerThe probability of noticing an edge that really exists, that is that the lower limit of the interval of the mean exceeds zero.
Cost in RThe cost of a trade (spread, commission, slippage) divided by the stop distance, that is how many R it costs to enter and exit.
Correlation between positionsHow much the results of two positions move together, from −1 (opposite) to +1 (identical).
Total open riskThe sum of the risks (stop distance × ounces) of all open positions, as a percentage of capital: the loss if all stops trigger together.
Risk planA set of rules written before trading: risk per trade, cap on positions, daily limit, drawdown steps, reviews and a trial at a fixed number.
Leverage usedThe position value divided by capital: how many times your capital you are moving.
Volatility dragThe loss of capital growth caused by the swings of the results: with fixed fraction the growth per trade is about f × μ − f² × σ² / 2.

What to do now

  1. Write your risk plan (chapter 13) before trading: risk per trade, cap, daily limit, drawdown steps, review, number of trades of the trial.
  2. Record every trade in R, with the cost, and compute mean and error every 25 trades.
  3. Start small (0.5-1%) until the lower limit of the mean exceeds zero; with the data of this book, treat every edge as zero until proof.
  4. Compute the ounces from the stop, check leverage, margin and total open risk, and count the cost in R.
  5. Do the trial at a fixed number, deciding beforehand how many trades and what you will do at the end.
  6. For the basics: the book "Market structure", the book "Order blocks from A to Z", the book "FVG and liquidity", the book "SMC: the whole system", the book "Indicators from A to Z", the book "Elliott and Fibonacci", the book "Chart patterns", the book "Candlesticks", the book "Hours, sessions and news" and the book "Support, resistance and levels". For risk and discipline: the risk and psychology course.

Final test

Fifteen questions on the book. You pass the test with at least 12 correct answers.

Final test of the book

1. On the chart the lines are the entry (5,150.93 $), the stop (5,179.60 $) and the target (5,122.26 $) of a short trade of 11-12 March 2026 (UTC hours). Which letter is the candle in which the stop is hit?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 11 Mar 2026 12:00 UTC to 12 Mar 2026 23:00 UTC.
Read the chart in words
WhenWhatPrice
12 Mar 2026 00:00 UTCentry5,150.93
12 Mar 2026 00:00 UTCstop: −1 R (28.7 $)5,179.60
12 Mar 2026 00:00 UTCtarget: +1 R5,122.26
11 Mar 2026 20:00 UTCPoint A5,182.30
12 Mar 2026 00:00 UTCPoint B5,159.43
12 Mar 2026 06:00 UTCPoint C5,167.39
12 Mar 2026 07:00 UTCPoint D5,182.00

Times in UTC.

2. On the chart the lines are the entry (4,675.66 $), the stop (4,653.47 $) and the target (4,697.85 $) of a long trade of 24 April 2026 (UTC hours). Which letter is the entry candle?

Chart exercise. Data: gold, XAUUSDT perpetual future (real candles) · 1 hour (H1) · from 23 Apr 2026 15:00 UTC to 25 Apr 2026 02:00 UTC.
Read the chart in words
WhenWhatPrice
24 Apr 2026 03:00 UTCentry4,675.66
24 Apr 2026 03:00 UTCstop: −1 R (22.2 $)4,653.47
24 Apr 2026 03:00 UTCtarget: +1 R4,697.85
23 Apr 2026 23:00 UTCPoint A4,702.79
24 Apr 2026 03:00 UTCPoint B4,680.88
24 Apr 2026 07:00 UTCPoint C4,696.61
24 Apr 2026 08:00 UTCPoint D4,698.81

Times in UTC.

3. A 10,000 $ account, 1% risk, stop at 25 $ per ounce. How many ounces do you buy?

4. How does the error on the mean change if you quadruple the trades?

5. A system wins 50% of trades. How likely is it, in 100 trades, to see at least 5 losses in a row?

6. The account has fallen 50% from its high. What gain is needed to return to the high?

7. What happens at a small risk (1%) when comparing fixed size and fixed fraction?

8. In a system with no edge, how does the probability of touching −50% change with the number of trades?

9. What happens using twice the Kelly fraction with a true edge?

10. Risk 1%, stop equal to 0.5% of the price. What leverage do you use, roughly?

11. Cost 0.6 $ per ounce, stop at 12 $. How much is the cost in R?

12. Three 1% positions in the same direction on gold, average correlation 0.5. What total effective risk do you have, roughly?

13. A true edge of 0.1 R, with 100 trades: how often do you recognise it (lower limit above zero)?

14. Did the daily limit at −2 R improve final capital on the real series?

15. With the martingale at a 1% base, how much would you risk after seven losses in a row?

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.