Learn · Gold futures · Template
Gold futures backtesting journal template
A replay session is only useful if you can review it. This template lists the columns worth logging for GC and MGC practice trades, how to fill each one, and the formulas that turn a log into a review.
By the NQ Replay team · Updated
Short answer
Log every simulated gold trade with its session date, contract month, session tag, setup, entry, stop and exit, then compute risk and result in ticks and the R multiple (result ÷ risk). Add whether you followed the rule. Ticks are $0.10/oz: $10 per GC tick, $1 per MGC tick.
The columns and how to fill them
| Column | How to fill it |
|---|---|
session_date | Trading day the session ends on (sessions start 6:00 p.m. ET the evening before). |
contract | GC or MGC plus month, e.g. GCZ6. Changes at each roll. |
session_tag | normal, holiday, roll, cpi, jobs, fomc — so you can filter later. |
window_et | When the signal happened in ET, e.g. 09:40. |
setup | Name of the written rule you tested. |
direction | long or short. |
entry | Fill price in $/oz ($0.10 steps). |
stop | Initial stop price. |
target | Planned exit price, if your rule has one. |
exit | Actual simulated exit price. |
risk_ticks | |entry − stop| ÷ 0.10. |
result_ticks | (exit − entry) ÷ 0.10, sign flipped for shorts. |
r_multiple | result_ticks ÷ risk_ticks. |
contracts | Number of contracts simulated. |
gross_usd | result_ticks × tick value ($10 GC, $1 MGC) × contracts. |
fees_usd | Your own round-trip fee assumption × contracts. |
rule_followed | yes or no. Be honest — this is the most useful column. |
skipped_signal | yes if the rule fired and you didn't take it. |
screenshot | Link or file name of the chart at entry. |
notes | One sentence on what you saw and felt. |
Copy the template
CSV header (paste into a spreadsheet)
session_date,contract,session_tag,window_et,setup,direction,entry,stop,target,exit,risk_ticks,result_ticks,r_multiple,contracts,gross_usd,fees_usd,rule_followed,skipped_signal,screenshot,notes
Turning the log into a review
After each block of trades (for example every 20), calculate:
- Count of trades and of skipped signals.
- Win rate = trades with r_multiple > 0 ÷ all trades.
- Average R = sum of r_multiple ÷ trades.
- Max drawdown in R = largest peak-to-trough drop of the running R total.
- Rule adherence = rule_followed "yes" ÷ all trades.
- The same figures filtered by session_tag (normal vs cpi/jobs/fomc vs roll).
A low adherence score means the result describes your behaviour more than the rule. For the full testing method see how to backtest gold futures.
Filling one row (illustrative)
| Field | Value | Working |
|---|---|---|
| direction / entry / stop / exit | long / 4,000.0 / 3,995.0 / 4,012.5 | Hypothetical prices, not a real trade |
| risk_ticks | 50 | (4,000.0 − 3,995.0) ÷ 0.10 |
| result_ticks | 125 | (4,012.5 − 4,000.0) ÷ 0.10 |
| r_multiple | 2.5 | 125 ÷ 50 |
| gross_usd (1 MGC) | 125 | 125 × $1 × 1 |
Frequently asked questions
What should a backtesting journal include?
At minimum: date, contract, setup, direction, entry, stop, exit, risk and result in ticks, R multiple, whether you followed the rule, and notes. Tags for event and roll days make the review far more useful.
Why log in ticks and R rather than dollars?
Ticks and R stay the same whether you simulate GC or MGC or change size, so results from different sessions stay comparable. Convert to dollars separately.
Does NQ Replay keep a journal automatically?
Yes — simulated trades placed in replay are logged to the Journal. The template here is for the reasoning the simulator can't capture, such as the setup name, tags and whether you followed your rules.
Sources
- CME Group — Gold (GC) futures contract specifications
- CME Group — Micro Gold (MGC) futures contract specifications
- NQ Replay 1-minute archive (GC, MGC, NQ, MNQ), measured October 23, 2026. Measurements describe past sessions only.