A trading journal records what you planned, what you executed and what happened. Performance measurement uses those records to assess whether your results justify the costs and risks you took. A broker’s profit and loss screen answers only part of that question.
For active trading, a useful journal separates trading results from decision quality. A profitable trade can break every rule in your plan. A losing trade can follow it exactly. Your records need to distinguish the two without giving either a free pass.
What to Record in a Trading Journal
Build the journal around information you can compare later. Record your entry reasoning before placing the order, then add execution details and review notes afterward. Writing the entire explanation after the trade closes gives hindsight too much editing power.
Entry and exit prices, timing, targets and reasons for taking a position form the basic record. Add a short review of what worked and what did not, consistent with CME Group’s trade log framework. Keep observations factual enough to check against your orders and charts.
| Field | What to record | Purpose |
|---|---|---|
| Trade identity | Trade ID, account, instrument, direction and strategy version | Keeps different approaches and accounts distinguishable. |
| Trade plan | Entry condition, initial stop, exit rule and reason for entry | Preserves the decision before the outcome is known. |
| Position risk | Quantity, initial planned dollar risk and percentage of account equity | Shows whether position size followed the plan. |
| Execution | Order and fill times, time zone, actual prices and partial fills | Allows planned and actual execution to be compared. |
| Costs | Commissions, transaction fees and applicable financing or borrowing charges | Supports a result after trading costs. |
| Outcome | Net profit or loss, result in R and holding time | Supports comparisons across trades. |
| Process review | Rule compliance, deviations, relevant market conditions and screenshots | Helps separate execution problems from strategy results. |
Use stable labels. “Breakout v1” and “breakout v2” are more useful than changing a setup’s name whenever its results disappoint. The rules belong in your trading plan; the journal records whether you followed them.
For discretionary trades, save a chart at entry and another at exit. Write “entered before the required close” rather than “bad discipline.” One describes an action you can measure. The other mostly expresses frustration.
Clean the Records Before Calculating Statistics
Reconcile imported transactions against your broker’s records. Check quantities, execution prices, fees and missing transactions. FINRA’s guidance on account statements and trade confirmations identifies these records as checks on transaction accuracy. An attractive dashboard does not repair a faulty import.
Decide what counts as one trade. A practical approach is to group fills belonging to one position from opening through final closure, while retaining the underlying transactions. Three partial exits should not become three independent winning trades simply because the export contains three rows.
Standardize time zones and reporting currency. Otherwise, a session analysis may group trades under the wrong hour, while results in different currencies may be added as though they were interchangeable.
Keep simulated trades separate from live trades, and retain all actual trades, including mistakes. You can filter rule violations for analysis, but deleting them from the main record produces a history you did not trade.
Trading Performance Metrics That Matter
Start with a small set of measurements you can verify. More decimal places do not make a weak sample more persuasive.
Net Profit and Loss
Net profit and loss, or net P&L, measures the result after applicable trading costs. Calculate the price result from actual fills, then deduct separately charged commissions, transaction fees, financing and borrowing costs. Include applicable credits consistently.
Do not subtract the spread or slippage twice. When P&L uses actual entry and exit prices, their execution effects are already reflected in those prices. You can measure slippage separately against a reference price to assess execution, but that comparison is not another charge to deduct.
Track subscriptions, data services and other overhead separately, then deduct them when assessing whether the activity is economically worthwhile. Label whether results are before or after taxes rather than mixing the two.
R Multiples
An R multiple expresses a trade’s net result relative to its initial planned dollar risk:
R multiple = net trade P&L ÷ initial planned dollar risk
Define the denominator consistently. In this hypothetical example, 1R means planned price risk before transaction costs. Buying 100 shares at $50 with an initial stop at $49 creates $100 of planned price risk. Selling at $51.50 produces $150 before costs. With $6 of charges, the net result is $144, or +1.44R.
Keep the original risk figure even if you later move the stop. Rewriting the denominator rewrites the measurement. Actual losses can exceed 1R through gaps, adverse execution or changes to the position. Planned risk is not a guaranteed maximum loss.
R helps compare trade outcomes at different sizes, but it does not replace dollar P&L or account returns.
Win Rate and Expectancy
Win rate measures the percentage of closed trades with positive net P&L. Count zero results consistently and disclose how you classify them. A high win rate alone says nothing about the size of the losses.
Historical expectancy is the average net result per trade:
Historical expectancy = total net P&L ÷ number of closed trades
Consider a separate hypothetical sample of 40 trades. Eighteen win an average of $150 after costs; 22 lose an average of $100. The win rate is 45%, but total net P&L is positive:
(18 × $150) − (22 × $100) = $500
Historical expectancy is therefore $500 ÷ 40, or $12.50 per trade. With those average win and loss sizes, a 40% win rate would break even. Winning more often than losing was not necessary in this example.
The $12.50 figure describes that sample. It is not a promise about the next trade. Also calculate average R when position sizes vary, so larger trades do not obscure changes in trade quality.
Profit Factor
Profit factor divides the sum of winning trade results by the absolute sum of losing trade results. Use results after trading costs on both sides.
In the example, profit factor is $2,700 ÷ $2,200, approximately 1.23. The sample generated about $1.23 in winning results for each $1 of losses.
Inspect how much depends on the largest winner. If there are no losing trades, the denominator is zero; an undefined or “infinite” display is not evidence of a proven strategy.
Maximum Drawdown
Drawdown measures the decline from a previous equity peak. Maximum drawdown is the largest such decline within the period measured.
With no deposits or withdrawals, an account falling from $50,000 to $44,000 has a 12% drawdown. Returning to $50,000 requires a gain of about 13.6% from the trough.
Use equity that includes open positions, not only closed trade balances. Also record the observation frequency: daily snapshots can miss an intraday low. Where external cash flows occur, use a cash flow adjusted performance series rather than treating a withdrawal as a trading loss.
Track recovery time and losing streaks alongside drawdown. Treat the worst recorded decline as historical evidence, not a ceiling on future losses.
Measure Account Returns Separately From Trade Results
Trade statistics assess individual positions. Account returns assess what happened to your capital, including periods when it remained uninvested.
A rising balance is not necessarily a positive return. If an account starts at $20,000, receives a $5,000 deposit and finishes at $24,000, its investment result is a $1,000 loss, assuming no other external flows. Dividing that loss by starting equity does not necessarily give an accurate percentage return when the deposit occurred during the period.
For performance comparisons, a time weighted return removes the effect of external cash flow timing by linking returns between cash flows. A money weighted return reflects the size and timing of those flows. These calculation approaches are detailed in the CFA Institute’s GIPS handbook on performance measurement.
Choose a method that matches the question and label it. Do not present return on margin posted as though it were return on total account equity.
Choose a relevant benchmark before reviewing results, and compare matching dates with income treated consistently. Show drawdown and market exposure beside returns rather than ranking approaches by profit alone.
How Many Trades Make the Results Meaningful?
There is no universal trade count that proves a strategy works. The required evidence depends on outcome variability, the size of the apparent advantage and whether the observations represent comparable conditions.
For illustration, 56 wins from 100 trades gives an observed win rate of 56%. Using the NIST Wilson confidence interval method, the corresponding 95% interval is approximately 46% to 65%, assuming independent outcomes with a stable win probability. That is substantial uncertainty, and it addresses win probability rather than profitability.
Real trading records may not meet those assumptions. Several positions entered during the same market move can share a common driver. Review results across dates and market conditions, not just the total number of rows.
Filtering also creates risk. Testing enough combinations of instrument, weekday, entry time and indicator setting can make chance look like skill. Repeated selection from the same data is the problem examined in Bailey and colleagues’ research on backtest overfitting.
Use journal filters to form questions, not certify answers. If morning trades look better, record the proposed rule before evaluating later observations. Keep the original results and a record of every variation tested. Fresh evidence is more useful than repeatedly polishing the same history.
Turn the Journal Into a Review Routine
After each session: reconcile transactions, finish notes and flag rule violations. Separate what you knew at entry from what became obvious afterward. Brief, timely notes are preferable to elaborate explanations reconstructed days later.
Each week: review net P&L, average R, costs and rule compliance. Examine the largest winner, largest loser and any unusual execution. Ask whether results came from the intended setups or from trades outside the plan.
At a scheduled strategy review: compare strategy versions and market conditions, then choose one question for further investigation. Changing entries, exits and position sizing together makes it harder to identify what caused any improvement.
Score process separately from money. If 24 of 30 trades met every predefined rule, full compliance was 80%. Compare compliant and noncompliant trades, but retain both in the account record. A profitable rule violation still belongs in the violation category.
If trade frequency or position size rises after losses, examine overtrading and loss chasing rather than treating the behavior as a new strategy. The next action may be to stop trading under an existing risk rule, not collect more trades at greater risk.
Spreadsheet or Dedicated Journal Software?
A spreadsheet is a reasonable starting point when manual entry remains manageable. Dedicated software may be worth considering when transaction volume, partial fills or multiple accounts make reconciliation cumbersome.
Evaluate practical requirements: accurate imports, transparent calculations, editable trade grouping, custom tags, screenshots, data exports and suitable account access permissions. Check several trades manually before trusting any summary statistics.
The journal’s value lies in the decisions it supports. Keep enough detail to reproduce results, identify repeated errors and test proposed changes. If a field never helps answer a useful question, remove it. If a number cannot be traced back to the underlying transactions, investigate it before acting on it.