A trading journal is not a notebook of wins and losses: it is the dataset you use to decide whether your method works, which mistakes cost you money and when to improve. This guide covers what to write down, how to calculate the key metrics with a complete ten-trade example, and how to read the results without fooling yourself.
Why memory is not enough
We remember recent trades, spectacular ones and the ones that hurt. Small losses that followed the plan are forgotten, while that one big win becomes the picture of your system. Without a record, your conclusions depend on that emotional filter.
A journal replaces impressions with data. It will not make you profitable by itself, but it lets you find out whether your method has an edge, whether costs cancel it, or whether the problem is not the strategy but how you execute it.
Which fields to record on every trade
The essentials are: entry and exit date and time, instrument, direction, entry price, stop, target, position size, commissions and spread, result in money and result in R. Add the setup (the pattern or technical reason), the market session, an emotional note (calm, impatient, tired) and a chart screenshot.
You do not need to record everything in equal depth. If a field is so tedious that you stop filling the journal in, drop it. A simple, complete record kept for six months beats a perfect one abandoned after three weeks. Our journal stores data only in your browser and lets you export it to CSV.
Define what 1R is before you start
R is your initial risk on each trade: the distance between entry and stop converted into money. With a 5,000 USD account risking 1%, 1R is 50 USD. A trade that makes 100 USD is +2R; one that loses 50 USD is -1R; one closed by hand with a 25 USD loss is -0.5R.
Measuring in R lets you compare trades of different size, instrument or account, and keeps balance growth from disguising the results. To work out how many lots correspond to your R, you can use our lot size calculator.
The metrics and their formulas
Win rate = winning trades ÷ total trades. Average winner in R = sum of winning R ÷ number of winners; average loser in R is the same with the losses. Expectancy (in R) = win rate × average winner − (1 − win rate) × average loser; it is equal to the sum of all R divided by the number of trades.
Profit factor = gross profit ÷ gross loss. Maximum drawdown = the largest fall from a peak of the equity curve to the lowest point after it. Maximum losing streak = the most consecutive losing trades. A low win rate is not bad if the average winner compensates for it, and a high one is useless if losses are much larger.
A complete example with 10 trades
Assume 1R = 50 USD and a starting account of 5,000 USD with fixed risk per trade. Results in R, in order: +2, -1, -1, +1.5, -1, +3, -1, -1, +2 and -0.5 (the last one closed before reaching the stop).
Winners: 4 (+2, +1.5, +3, +2), totalling 8.5R. Losers: 6, totalling 5.5R (five at -1 and one at -0.5). Win rate = 4 ÷ 10 = 40%. Average winner = 8.5 ÷ 4 = 2.125R. Average loser = 5.5 ÷ 6 ≈ 0.917R. Expectancy = 0.4 × 2.125 − 0.6 × 0.917 = 0.85 − 0.55 = 0.30R per trade; check: 3R ÷ 10 = 0.30R. Profit factor = 8.5 ÷ 5.5 ≈ 1.55.
In money, the net result is +3R = +150 USD, taking the account from 5,000 to 5,150 USD. The equity curve in USD reads 5,100, 5,050, 5,000, 5,075, 5,025, 5,175, 5,125, 5,075, 5,175, 5,150. Maximum drawdown is 2R (100 USD, about 2% from the peak), which happens twice: from 5,100 to 5,000 and from 5,175 to 5,075. The maximum losing streak is 2 trades.
How to read that example with caution
An expectancy of 0.30R sounds positive, but ten trades prove nothing. The dispersion of those results (a standard deviation of about 1.6R) means the standard error of the average is roughly 0.5R: the true value could lie anywhere from a sizeable loss to a larger gain. With 100 trades and similar dispersion, the error falls to about 0.16R, so a measured 0.30R would still be compatible with an expectancy close to zero. This is only an approximation, but it shows why you need a sample, and why it is worth running the numbers in our expectancy calculator.
The same applies to win rate: with a true 40%, the 95% margin of error is about ±30 points with 10 trades, ±17 with 30 and ±10 with 100. And losing streaks are normal: with a 40% win rate, the probability of losing five in a row from any given starting point is 0.6^5 ≈ 7.8%, and eight in a row about 1.7%. If you do not expect them, a sound plan looks broken.