Part 1: Risk and its numbers · Chapter 1 of 14 · 10 min
What risk is
The risk of a trade is the loss you accept before entering, and it is measured in R: 1 R is the distance between entry and stop. Three real series from the previous books give mean R of −0.03, +0.03 and +0.09 per trade, with a deviation of about 1 R (0.56 for the hours system): over 100 trades the result swings by ±19 R, far more than any edge we found. That is why the book is about how much to risk, not what to buy.
Risk does not mean losing
In everyday language "risk" is the possibility of losing. In this book it has a precise, measurable meaning: the risk of a trade is the loss you accept before entering. It is set by the stop. If you buy gold at 4,500 $ with the stop at 4,478 $, your risk is 22 $ per ounce. You do not know whether you will lose, you know how much at most (barring price jumps, chapter 8).
That distance is called 1 R. A trade that reaches the target placed at the same distance as the stop is worth +1 R; one hit by the stop is worth −1 R; one closed halfway is worth a fraction. Measuring everything in R has a decisive advantage: two trades with stops of 12 $ and 40 $ become comparable, and the results of different systems can be added.
The step to dollars depends on a single choice, the percentage of risk per trade. With a 10,000 $ account:
| Risk per trade | 1 R is worth | After −1 R the account is | After −10 R in a row the account is |
|---|---|---|---|
| 1% | 100 $ | 9,900 $ | about 9,044 $ |
| 2% | 200 $ | 9,800 $ | about 8,171 $ |
| 5% | 500 $ | 9,500 $ | about 5,987 $ |
(The last column recomputes the loss on the capital left after each trade.) The position size, in ounces or lots, follows from here: dollar risk divided by the stop distance (chapter 8).
Read the chart in words
H1 gold chart of 19-21 February 2026 with the entry, stop and target of a winning trade, in R.
| When | What | Price |
|---|---|---|
| 20 Feb 2026 03:00 UTC | entry | 4,991.35 |
| 20 Feb 2026 03:00 UTC | stop: −1 R (22.4 $) | 4,968.99 |
| 20 Feb 2026 03:00 UTC | target: +1 R | 5,013.71 |
Times in UTC.
The book's four series
The rest of the book uses real series of trades, those of the previous books, in R and already net of the 0.3 $ cost:
- LA, bounce at time-based levels (book 10, variant A): entry on the bounce at yesterday's, the week's and Asia's levels.
- LD, break of H4 swings (book 10, variant D).
- SD, hours system of Asia with filters (book 9, variant D): breakout of the Asian range, stop at the opposite extreme, target 1 R, time exit.
- SX, a synthetic example: 552 trades worth +1 R with probability 0.60 and −1 R otherwise, at the same instants as LA. It is not a measurement: it serves to see what happens when an edge really exists.
| Series | Trades | Mean R ± error | Standard deviation | Winners | Average win | Average loss | Breakeven at |
|---|---|---|---|---|---|---|---|
| LA | 552 | −0.03 ± 0.08 | 0.99 | 49% | 0.98 R | 1.01 R | 51% |
| LD | 367 | +0.03 ± 0.10 | 1.00 | 52% | 0.98 R | 1.00 R | 50% |
| SD | 55 | +0.09 ± 0.15 | 0.56 | 51% | 0.52 R | 0.35 R | 40% |
| SX (synthetic) | 552 | +0.21 ± 0.08 | 0.98 | 60% | 1.00 R | 1.00 R | 50% |
"Breakeven at" is the share of winning trades needed for a mean R of zero with that ratio of average win to average loss: 1/(1 + average win/average loss). For LA, with wins and losses almost equal, 51% is needed, and 49% is obtained: hence the −0.03. For SD wins are larger than losses (1.48 times), so 40% is enough and the 51% obtained gives +0.09.
Read the chart in words
Daily gold chart with, below, three curves of cumulative R.
The expected value
The expected value of a trade is the mean of what you expect to gain, in R. With winning share p, average win G and average loss L (in R, positive): p × G − (1 − p) × L. For LA: 0.49 × 0.98 − 0.51 × 1.01 = about −0.03. For SX: 0.60 × 1.00 − 0.40 × 1.00 = +0.20.
A positive expected value is not enough: you also need a small error. The error is ± 1.96 times the standard deviation divided by the square root of the number of trades. For LA (552 trades) it is ± 0.08: the true expected value lies between −0.11 and +0.05. None of these three systems demonstrates a positive expected value; the only series that has one, SX, has it because we built it that way.
A ±1 R coin flip
In the LA, LD and SX series R is almost always −1 or +1: for LA 50% of trades sit between −1.05 and −0.9, 48% between +0.9 and +1, only 2% in between (the time exits). The reason is the rule they are built with: stop and target at the same distance, time exit after 24 candles that almost never triggers. The result of a trade is, in practice, a coin flip rigged by a few points.
SD is different: in 76% of cases R is in between, because the time exit (at 21:00 UTC) often triggers; and its deviation is 0.56.
This has a consequence the whole book develops: the series of results is very noisy. A series of 100 LA trades has an expected R of −3.3, but a standard deviation of 10 R. A 95% interval (±1.96 deviations) goes from about −22 to +16 R:
| Series | Mean R per trade | 100 trades: expected R | 95% interval (± 1.96 deviations) |
|---|---|---|---|
| LA | −0.03 | −3.3 | from −22.7 to +16.1 |
| LD | +0.03 | +2.9 | from −16.7 to +22.5 |
| SD | +0.09 | +9.2 | from −1.8 to +20.2 |
| SX | +0.21 | +20.7 | from +1.5 to +39.9 |
With a 1% risk per trade, a hundred LA trades can end between −22% and +16% of the account, with no real edge: chance alone makes this difference. Even SX, with a built-in edge of 0.2 R per trade, starts from +1.5 R in the worst case over a hundred trades.
Read the chart in words
H1 gold chart of 22-23 January 2026 with the entry, stop and target of a losing trade and four candles marked with the letters A, B, C and D.
| When | What | Price |
|---|---|---|
| 23 Jan 2026 00:00 UTC | entry | 4,960.86 |
| 23 Jan 2026 00:00 UTC | stop: −1 R (34.7 $) | 4,926.21 |
| 23 Jan 2026 00:00 UTC | target: +1 R | 4,995.51 |
| 22 Jan 2026 20:00 UTC | Point A | 4,929.64 |
| 23 Jan 2026 00:00 UTC | Point B | 4,961.63 |
| 23 Jan 2026 07:00 UTC | Point C | 4,956.93 |
| 23 Jan 2026 08:00 UTC | Point D | 4,946.34 |
Times in UTC.
What to do with it
- Before entering, write the stop and compute 1 R in dollars. Without a stop there is no risk to measure.
- Choose the risk percentage before looking at charts. Afterwards the temptation is to raise it when "the setup is beautiful".
- Record every trade in R. It is the only way to compare trades with different stops and to add different systems.
- Look at the expected value and its error, not the share of winning trades. 51% winners with wins smaller than losses loses; 40% with wins twice the losses wins (0.4 × 2 − 0.6 = +0.2).
- Do not judge a system from a hundred trades. The interval is ±19 R: a much larger number is needed (chapter 12).
Procedure in four steps
- Set the stop and compute the dollar risk (entry minus stop, for the size you will buy).
- Choose the risk percentage per trade and compute what 1 R is worth on your account.
- Record every result in R, not in dollars.
- Compute expected value, deviation and error of the series, and always look at the error before the value.
Exercises
Chapter exercises
1. On the chart the lines are the entry (4,960.86 $), the stop (4,926.21 $) and the target (4,995.51 $) of a long trade on 23 January 2026 (UTC hours). Which letter is the candle where the stop is hit?
Read the chart in words
| When | What | Price |
|---|---|---|
| 23 Jan 2026 00:00 UTC | entry | 4,960.86 |
| 23 Jan 2026 00:00 UTC | stop: −1 R (34.7 $) | 4,926.21 |
| 23 Jan 2026 00:00 UTC | target: +1 R | 4,995.51 |
| 22 Jan 2026 20:00 UTC | Point A | 4,929.64 |
| 23 Jan 2026 00:00 UTC | Point B | 4,961.63 |
| 23 Jan 2026 07:00 UTC | Point C | 4,956.93 |
| 23 Jan 2026 08:00 UTC | Point D | 4,946.34 |
Times in UTC.
2. With a 10,000 $ account and a 1% risk per trade, how much is 1 R worth?
3. A system wins 60% of trades, with an average win of 0.5 R and an average loss of 1 R. What is the expected value per trade?
4. LA has mean R −0.03 with an error of ± 0.08 over 552 trades. What does it mean?
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.