The only thing you actually control
You do not control whether a trade wins. Nobody does. What you control completely is how much you lose when it doesn't — and that is where the entire game is decided.
Two traders can take identical trades with identical entries and exits, and one ends the year up while the other is wiped out. The difference isn't analysis. It's position sizing.
This chapter is the least exciting in the manual and by some distance the most important. Skip Chapter 3 and you'll take mediocre entries. Skip this one and you won't have an account to take them with.
Position sizing
The rule is simple: decide what you're prepared to lose on a single trade, then work the stake backwards from your stop.
Not the other way round. "I'll trade £5 a pip because that's what I normally do" is how a wide stop turns into an outsized loss.
The calculation
Stake per pip = amount you'll risk ÷ stop distance in pips
Account £5,000, risking 1% (£50) per trade:
- Stop is 25 pips away → £50 ÷ 25 = £2.00 per pip
- Stop is 50 pips away → £50 ÷ 50 = £1.00 per pip
- Stop is 100 pips away → £50 ÷ 100 = £0.50 per pip
All three risk exactly £50. A wider stop means a smaller stake — never a bigger loss.
What percentage?
You'll see 1–5% quoted widely. Both ends of that range are not equally survivable, and it's worth seeing why in numbers rather than taking anyone's word for it.
| Risk per trade | After 6 losses in a row | Account left |
|---|---|---|
| 1% | −5.9% | £4,706 of £5,000 |
| 2% | −11.4% | £4,428 |
| 3% | −16.7% | £4,165 |
| 5% | −26.5% | £3,675 |
Six losses in a row is not unusual — the next section shows how routine it is. At 1% you barely notice. At 5% you've lost a quarter of your account and, far more dangerously, you're now in the emotional state where people start revenge trading.
Most traders who survive their first few years risk 1% or less. If you're starting out, start at 0.5%. You are buying yourself the room to be wrong while you learn, and being wrong while learning is the entire job.
Losing streaks are normal — here's the maths
Traders abandon perfectly good strategies during ordinary losing runs, because nobody told them what ordinary looks like.
Take a strategy that wins 50% of the time. Over 100 trades, the probability of hitting a run of consecutive losses at some point:
| Losing streak | Chance of seeing it in 100 trades |
|---|---|
| 4 in a row | Around 97% — near certain |
| 5 in a row | Around 81% — expect it |
| 6 in a row | Around 55% — more likely than not |
| 8 in a row | Around 19% — uncommon, not alarming |
| 10 in a row | Around 5% — rare, still not broken |
Read that again. With a perfectly good 50% strategy, six consecutive losses is more likely to happen than not. It says nothing whatsoever about whether your edge still works.
This is why position sizing and psychology are the same topic. At 1% risk, an eight-trade losing streak costs you about 8% and you carry on. At 5% it costs a third of your account, and almost nobody executes their plan calmly from there.
How DEALFx handles this
The backtester runs a Monte Carlo simulation — reshuffling your trade sequence thousands of times to show the range of outcomes the same trades could have produced, including how often they end negative. It's the difference between knowing your strategy made money and knowing how bumpy the ride was. See the evidence features →
Why losses hurt more than wins help
An asymmetry that catches almost everyone: the gain needed to recover a loss is always larger than the loss itself, and it grows viciously.
| You lose | To get back to even you need |
|---|---|
| 10% | +11% |
| 20% | +25% |
| 33% | +50% |
| 50% | +100% |
| 75% | +300% |
Lose half your account and you must double what's left just to return to where you started. This is why professional risk management is obsessive about avoiding large drawdowns rather than chasing large gains. Staying out of the bottom rows is worth more than any entry technique in this manual.
Reward, risk, and why win rate alone is meaningless
"What's your win rate?" is the wrong question. A 90% win rate loses money if the 10% are large enough.
What matters is expectancy — the average amount you make per trade over many trades:
Expectancy
(Win rate × average win) − (Loss rate × average loss)
Strategy A: wins 40% of the time, at 2R per win, 1R per loss.
(0.40 × 2) − (0.60 × 1) = +0.20R per trade
Strategy B: wins 70% of the time, at 0.5R per win, 1R per loss.
(0.70 × 0.5) − (0.30 × 1) = +0.05R per trade
Strategy A loses more often and makes four times as much. Strategy B feels far better to trade, which is precisely the trap.
This is the arithmetic behind the 1.5R minimum in Chapter 2. With a 2R target you can be wrong more often than right and still make money — which is a far more forgiving place to learn than needing to be right most of the time.
| Reward:risk | Win rate needed to break even |
|---|---|
| 1 : 1 | 50% |
| 1.5 : 1 | 40% |
| 2 : 1 | 33% |
| 3 : 1 | 25% |
Stops, and moving them properly
Your stop goes where your analysis is wrong — below the swing low you're buying, above the swing high you're selling, plus a little room for noise. Never at a distance chosen because the loss feels tolerable.
The one legitimate way to move a stop
Never further away. In your favour, though, moving a stop is a real technique:
- To breakeven. Once a trade has moved roughly 1R in your favour, move the stop to your entry. The trade can now no longer lose.
- Trailing. As price makes new highs, move the stop up behind it — under each new higher low. You give back some profit at the end, in exchange for letting a big winner run instead of closing it early.
Be honest about what trailing does
Trailing is not free money. It converts some full winners into partial ones, and in exchange it occasionally catches a very large move. Over a large sample it's often close to expectancy-neutral — what it really changes is the shape of your results, and how easy they are to sit through. That's a legitimate reason to use it. "It makes more money" usually isn't.
How DEALFx handles this
Auto-pilot does all of this on a timer: breakeven at 1R, then a trail, then a profit ratchet that holds a share of the best price seen and pushes the target out while a trend keeps running. Every stop move is logged with the reason. We've also published the honest finding that the ratchet is roughly expectancy-neutral — it's on by default because the result is easier to live with, not because it earns more. See how auto-pilot manages trades →
The four ways people undo a good strategy
Nearly every blown account traces back to one of these.
1. Revenge trading
You take a loss, feel the sting, and immediately take another trade to win it back — smaller setup, bigger size. This is the most reliably destructive behaviour in retail trading. The fix is mechanical, not emotional: a hard rule that after two consecutive losses you stop for the day. Not "if you feel upset" — you won't notice that you are.
2. Moving stops
Price approaches your stop, you widen it "just a bit" to give the trade room. Sometimes it works, which is what makes it lethal — it gets rewarded often enough to become a habit, and then one trade takes a month of gains.
3. Overtrading
Nothing qualifies today, so you lower your standards. Boredom is not a signal. Sitting out a day is a legitimate, professional outcome, and the days you force a trade are disproportionately the days you lose.
4. Cutting winners early
You're up, you're nervous, you take a third of your target. Do it habitually and your winners are all small while your losers stay full size — the one pattern that no win rate can rescue.
Why this is a software problem, not a character problem
Note what all four have in common: they happen while a trade is live, when money is on the line and your judgement is measurably worse than it was an hour earlier. This is exactly why rules written calmly in advance — and, if you want, executed by something that doesn't get nervous — beat good intentions. It isn't a failure of discipline. It's a predictable feature of being human, and the sensible response is to design around it.
The journal
The single highest-return habit available to you, and the one most people skip.
For every trade record: the pair, date, setup, entry, stop, target, size, result in R, and one line on whether you followed your plan.
That last column is the one that matters. Four outcomes:
| Won | Lost | |
|---|---|---|
| Followed plan | Good trade. Repeat. | Also a good trade. Losses are part of the plan. |
| Broke plan | The dangerous one. Rewarded for a bad habit. | Bad trade. At least the lesson is obvious. |
Most people only examine losses. But a rule-break that won is the most expensive box on that table, because it teaches you the wrong lesson at the exact moment you're most inclined to believe it. Mark it as an error regardless of the money.
How DEALFx handles this
The journal reconciles itself against your broker's own record, so exits and P/L match reality rather than what the app assumed. It also saves the market regime at entry, which is how you discover things like "I lose money in choppy conditions" — a pattern almost impossible to spot by memory. See the journal features →
What to take from this chapter
- Stake = risk ÷ stop distance. Always work backwards from the stop.
- Risk 1% or less. Start at 0.5% while you're learning.
- Six losses in a row is normal, even with a good strategy.
- Lose 50% and you need +100% to recover. Avoid deep drawdowns above all.
- Expectancy, not win rate. A 40% win rate at 2R beats 70% at 0.5R.
- Move stops only in your favour, and be honest that trailing changes shape, not edge.
- Stop for the day after two losses — a mechanical rule, not a judgement call.
- Journal everything, and log rule-breaks as errors even when they win.