This is the lesson that separates trading from gambling, and it requires nothing beyond arithmetic. If you internalize the three ideas here, R-multiples, expectancy, and streak math, you will understand your results better than the majority of people who have traded for years. Skip it, and every green week will be luck you cannot repeat and every red week a mystery you cannot fix.
R: measure every trade in units of risk
Before entering any trade you define your stop (Module 3 covers where). The distance between entry and stop, in dollars, is your R: one unit of risk.
500 shares → 1R = $100 at risk
Exit $2.60 → made $0.60/share = +3R
Stopped $1.80 → lost $0.20/share = −1R
From today forward, you never say "I made $300." You say "+3R." Why? Because dollars lie across position sizes and account sizes, but R is honest. A +3R trade is excellent whether the account is $500 or $500,000. A trader who makes $900 on 1R=$1,000 of risk did worse than one who made $90 risking $30. R makes your results comparable, auditable, and improvable, and it makes the next concept computable.
Expectancy: the only number that says if you have an edge
Expectancy is what the average trade pays you, in R, over many trades:
Two traders, same month:
→ (0.40 × 2.5) − (0.60 × 1.0) = 1.00 − 0.60 = +0.40R per trade ✓
Trader B: wins 85%, avg win +0.3R, avg loss −2.5R
→ (0.85 × 0.3) − (0.15 × 2.5) = 0.255 − 0.375 = −0.12R per trade ✗
Trader B wins almost every day, feels like a genius, posts green screenshots, and is mathematically guaranteed to lose money over time. Trader A is wrong more often than right and compounds steadily. Win rate without payoff size is not information. This single confusion funds most of the fake-guru economy: 90% win rate is trivially easy to manufacture if you let losers run and cut winners instantly, and it is precisely how accounts die slowly then suddenly.
Payoff asymmetry: how being wrong often still pays
Look again at Trader A: losing 60% of the time, profitable. The entire momentum-trading business model lives inside that arithmetic. Runners offer occasional +3R, +5R, +10R outcomes against defined −1R risk. You do not need to be right often; you need your rights to be LARGE and your wrongs to be uniform, capped at −1R, every single time.
This is why "cut losses fast, let winners run" is not a poster slogan; it is the literal generator of positive expectancy. And it is why the discipline lessons in Module 3 and 4 matter more than any setup: one uncapped loss, one "it will come back" hold that turns −1R into −6R, poisons the entire equation.
Streak math: variance will test you
Even with a real edge, losing streaks are a mathematical certainty, not a sign the edge died:
Probability of a 5-loss streak somewhere: ~97% (near certain)
Probability of a 7-loss streak somewhere: ~50% (coin flip)
Probability of a 9-loss streak somewhere: ~14% (happens to someone every month)
Now connect this to sizing. Risking 1% per trade, a 7-loss streak costs ~7% of the account: annoying, survivable, edge intact. Risking 10% per trade, the same statistically-ordinary streak costs half your account, and the math of holes takes over: a 50% loss requires a 100% gain just to break even. Ruin is not usually caused by bad strategies; it is caused by ordinary variance hitting oversized positions.
What this means before you risk a dollar
- Define R before entry, every trade, no exceptions. No stop defined = no R = no trade.
- Log every trade in R. Wins in R, losses in R, expectancy monthly.
- Judge yourself on process metrics you control (losses capped at −1R, rules followed) rather than green-day counts you do not.
- Size so that a 7-loss streak, which WILL come, is a bruise and not an amputation.
- R is your unit: one planned risk. All results are measured in it.
- Expectancy = (Win% × avg win) − (Loss% × avg loss). Positive or you are gambling with extra steps.
- High win rates can hide negative expectancy: payoff size is half the equation, and the hidden half.
- Momentum trading monetizes asymmetry: uniform −1R losses against occasional multi-R winners.
- Losing streaks are statistically guaranteed: sizing exists so variance cannot kill a real edge.
Drill: audit ten trades
Take your last ten trades (paper trades count). For each, write entry, stop, exit, and the result in R. Compute win rate, average win in R, average loss in R, and expectancy. Two things will likely surprise you: how many trades had no definable R (no stop existed), and what your real expectancy is versus what your memory claims. This 20-minute exercise is the entire foundation of Module 3.