How to Build a Futures Trading Journal That Records Decisions, Not Just P&L

If your journal is mostly screenshots of green and red trades with a note like 'good trade', it can't tell you what to change. Short-term results are noisy even for a sound plan, so a journal built on outcomes will reward lucky mistakes and punish good decisions that happened to lose. The fix is to grade the decision separately from the result.

Start with the free journal template for Excel or Google Sheets if you need a place to record your trades. This guide explains how to record the decision behind each trade, so your review has more to work with than a dollar result.

Process and outcome are different measurements

A good trade here means you followed your plan: the setup was valid, the size fit your rules and you managed it as written. Whether it hit the target is a different question. Crossing the two gives four cases:

• Good process, profit: expected, keep doing it.
• Good process, loss: normal variance, no change.
• Bad process, loss: expected, fix the error.
• Bad process, profit: the dangerous one. The mistake got rewarded, so you're more likely to repeat it, and it'll cost you when the same habit meets a losing stretch.

The fields to log

You don't need software. A spreadsheet with these columns works.

Before the trade: date and time, setup type (for example a reference level test, a VWAP reclaim or a trend continuation), planned entry, stop and target, one sentence on why the setup is valid, your emotional state from 1 (calm) to 5 (anxious or excited), and the market condition (trend day, range day, news-driven).

If you trade from KPL reference levels, remember they're fixed prices anchored to the regular-session opening print, so a level-based setup can only be tagged after the open. Don't backfill a setup name you didn't have at entry.

After the trade: actual entry, stop and exit, the exit reason (target, stop, or manual, with the reason for a manual exit), result in dollars and in R, a process grade, and a one-sentence lesson.

Grading process

• A: setup defined before entry, entry at the planned price or better, stop where planned, size within your rules, exit per plan.
• B: one small deviation, such as a tick of extra slippage or a slightly different exit.
• C: a significant deviation: oversized, entered without confirmation, moved the stop with no rule allowing it, or exited on feel.
• F: several large deviations, no plan, a revenge trade, or breaking your sizing rules.

Set your own target for how many trades should be A or B. I like having one, but no data I can cite says what the right number is, so pick a target you'll actually track and adjust it.

What to look for in the data

Setup type. After enough trades, compute average R by setup. Keep in mind how noisy small groups are. If one setup has 12 trades, a few outliers decide its average, so treat a result as a hypothesis to keep logging, not a verdict. Hypothetical example: setup A has 10 trades with a total of +4R, so its average is +0.4R. Setup B has 10 trades with a total of -3R, an average of -0.3R. That's worth watching, not worth deleting B on its own.

Emotional state. Compare your 1 to 5 rating with your process grade. You may find your worst-graded trades cluster at high ratings. Whether that's true for you is something only your data can say, which is why it's worth recording.

Market condition. A strategy built for ranges may do badly on strong trend days. Logging the condition lets you see where yours does and doesn't fit, again with the small-sample caveat.

Exit reasons. Treat every manual exit as a question. Common patterns are cutting winners early, holding losers too long, and flipping direction mid-session without a plan. They're easy to miss in single trades and become visible across many.

Review routine

Weekly, 15 to 20 minutes: count the grades, find the most common process error, and read every F trade to see what triggered the deviation.

Monthly, 30 to 45 minutes: average R by setup and by market condition, plot emotional state against process grade, compare planned risk to reward against what you actually got, and choose one behavior change for next month. One change is enough. It keeps the effect traceable.

Getting trades out of NinjaTrader

In NinjaTrader 8, the Trade Performance report lists your trades with entry and exit prices, profit and loss, and duration, so you don't have to retype the mechanical fields. Export it to CSV, paste it into your journal and add the qualitative columns by hand. Take a chart screenshot at exit with your entry and exit marked and your grade written on it, then look through the week's screenshots during your review. Expect the whole thing to take 5 to 10 minutes a week once the template exists.