Gold strategies
JOURNALING THE GOLD PULLBACK
A pullback entry generates more decision points than most XAUUSD setups — depth, shape, confluence type, order type, trigger quality — and a journal that only records win or loss throws away every one of them. This page is the field list that turns a trade log into something that can actually diagnose why a pullback strategy is or is not working.
A gold pullback journal needs, at minimum: level type, retracement depth in both dollars and percent, the specific confirmation trigger used, and time-to-fill from zone arrival to entry. Tracked over a rolling sample, these fields expose edge decay through R by retracement bucket, fill rate, maximum adverse excursion and hit rate by confirmation type — the four metrics that show whether the setup is degrading before the account does.
WHY WIN/LOSS ALONE HIDES THE PROBLEM
Two pullback trades can share the same outcome — both stopped out — for entirely different reasons: one entered on a corrective retracement that simply failed to continue, the other entered on an impulsive move mislabelled as a dip. A journal that records only the outcome cannot distinguish them, which means thirty losing trades produce zero information about which of the setup's several moving parts is actually broken.
The fields below are chosen because each answers a specific diagnostic question rather than adding volume to the log. If a field cannot change a future decision, it does not belong in the journal regardless of how easy it is to record.
None of this requires abandoning a simple spreadsheet or the Gold Hunts journal's default fields — it requires adding the handful below that are specific to how a pullback entry is built, on top of whatever the platform already captures automatically from the MetaTrader 5 sync.
CORE FIELDS TO RECORD ON EVERY TRADE
These four fields are the minimum. Everything downstream in this page is derived from them, so skipping one removes an entire category of later analysis.
Tags worth adding
Tags are cheap to add and expensive to reconstruct after the fact, so they are worth attaching at entry time even before a use for them is obvious.
- Session tag — London impulse, post-London retest, NY overlap retest, or "outside core windows" — matched to the best-sessions reference table.
- Shape tag — corrective or impulsive — recorded from the candle-body comparison at the moment of entry, not reconstructed afterward.
- Order-type tag — resting limit, stop-entry on trigger break, or split entry across both.
- Re-entry tag — first attempt, or re-entry after a stopped first attempt in the same trend leg.
Screenshots to keep
Two screenshots per trade are enough: one taken the moment the confluence zone is marked, before the pullback begins, and one taken at the trigger candle close. The first proves the zone was not drawn with hindsight; the second is the only chart evidence of what the trigger candle actually looked like once spread and slippage are no longer visible on a later chart.
| Field | What to record | Why it matters |
|---|---|---|
| Level type | Fib band, 15m 20 EMA, session VWAP, broken swing level, or a stated combination | Reveals which confluence reference is actually carrying the edge versus which is decoration |
| Retracement depth ($ and %) | Both the dollar distance from impulse extreme and the equivalent fibonacci percentage | Dollar depth normalises across differing impulse-leg sizes; percent depth lines up with the standard 38.2–61.8% zone |
| Confirmation trigger | 15m rejection close, 5m rejection close, stop-entry break, or limit fill with no confirming candle | Separates confirmed entries from unconfirmed ones so hit rate by trigger type can be measured directly |
| Time-to-fill | Minutes from price first touching the confluence zone to the entry being filled | Long delays flag chasing or hesitation; near-zero delays on limit orders flag unconfirmed fills |
EDGE-DECAY METRIC 1 — R BY RETRACEMENT BUCKET
Group closed trades into three depth buckets — 38.2% to 50%, 50% to 61.8%, and 61.8%+ — and compute average R for each bucket on a rolling basis. The hub page treats these bands as roughly equivalent in principle; in practice one bucket usually carries most of the expectancy on any given account, broker and gold regime, and the other two are close to break-even or negative.
This is the single most valuable pullback-specific metric because it converts a qualitative rule ("38.2% to 61.8% is the continuation zone") into an account-specific, testable one. A trader whose 61.8%+ bucket is consistently negative should stop taking deep retracements even though the hub-page rule technically permits them, because the rule is a starting filter, not a guarantee that every sub-range performs equally.
| Bucket | What it usually reveals |
|---|---|
| 38.2% – 50% | Often the highest hit rate in strong-trend conditions; smaller sample sizes if trends are shallow |
| 50% – 61.8% | Usually the largest sample and the closest to the account's overall average R |
| 61.8%+ | Frequently the weakest bucket once confluence and shape filters are applied honestly; a common source of hidden negative expectancy |
EDGE-DECAY METRIC 2 — FILL RATE
Fill rate is the percentage of marked confluence zones that actually produce a filled entry, tracked separately for limit orders and stop entries. A falling fill rate on limit orders across a rolling twenty-zone sample usually means retracement depth has been trending shallower than the zone assumes — the market is not giving back as much as the strategy expects, which is itself useful information about the current trend's strength.
A near-100% fill rate on a resting limit order combined with a below-average hit rate is a specific red flag: it usually means the limit is placed too shallow, filling on noise before the retracement is actually finished, rather than filling inside a genuine confluence zone.
EDGE-DECAY METRIC 3 — MAXIMUM ADVERSE EXCURSION (MAE)
MAE on a pullback entry measures how far price traded against the position, in dollars per ounce, before either the target or the stop was reached. On a well-placed pullback stop, winning trades should show a small MAE relative to the eventual stop distance — the entry, by definition, was meant to be close to where the retracement finished.
A rising average MAE on winning trades over time, even while win rate holds steady, is an early warning that entries are drifting earlier into the retracement than the confluence zone justifies — a slow-motion version of the front-running pattern discussed on the psychology page, visible in the numbers before it shows up as a losing streak.
QUANTIFY THE STOP DISTANCE PROPERLY
- GOLD ATR STOP CALCULATOR
Size the stop to gold volatility rather than to a round number you like the look of.
- GOLD R MULTIPLE CALCULATOR
Grade every closed trade in R so the record survives changes in account size.
EDGE-DECAY METRIC 4 — HIT RATE BY CONFIRMATION TYPE
Split the record by confirmation trigger — 15-minute rejection close, 5-minute rejection close, stop-entry break, and unconfirmed limit fill — and compute hit rate separately for each. The hub page's stated default is the 15-minute trigger; this metric is what confirms whether that default actually holds for a specific account's execution and typical trading hours, or whether the 5-minute trigger is quietly performing better or worse than assumed.
Unconfirmed limit fills — a filled order with no rejection candle following it — should be tracked as their own category rather than merged into "limit orders" generally. This category typically shows the lowest hit rate of the four, and isolating it is what makes the case, in a trader's own numbers rather than in a page like this one, for requiring confirmation even at the cost of occasionally missing a fill.
| Confirmation type | What a healthy pattern looks like | Investigate when |
|---|---|---|
| 15m rejection close | Hit rate near or above the account baseline | It falls meaningfully below the 5m figure over 20+ trades |
| 5m rejection close | Lower average R per trade, similar or higher hit rate | Hit rate drops without a corresponding tighter average stop |
| Stop-entry break | Fewer trades, generally cleaner outcomes | Frequent missed fills on fast rejection candles worth reviewing separately |
| Unconfirmed limit fill | Should be the smallest and weakest category | It approaches the hit rate of confirmed entries — a sign confirmation is not adding real value on this account |
A GOLDHUNTS JOURNAL WALKTHROUGH
Connect MetaTrader 5 to the Gold Hunts journal and every closed XAUUSD trade is attributed automatically to its UTC session, entry and exit price, and R multiple — the infrastructure this page assumes is already in place rather than something to rebuild manually. The pullback-specific fields above are added as custom tags and notes at the point of entry, not reconstructed from the fill afterward.
A practical routine: mark the confluence zone and screenshot it the moment the impulse leg completes, tag the session window and confluence type before the trigger candle even forms, then complete the remaining fields — trigger type, depth in both dollars and percent, time-to-fill — the moment the order is filled. Reviewing rolling twenty-trade windows on the four decay metrics above, rather than lifetime aggregates, is what catches a decaying setup while it is still a small sample rather than after a drawdown has already happened.
Weekly review is the appropriate cadence for this specific setup, because pullback frequency on XAUUSD is lower than for scalping or breakout setups — a monthly review risks acting on too few trades, while a daily review risks reacting to single-trade noise inside one of the four metrics.
SET UP THE JOURNAL
- FREE GOLD TRADE JOURNAL
Sync MetaTrader 5 and let the journal attribute each gold trade to its UTC session, setup type and R multiple automatically.
- TRADING EXPECTANCY CALCULATOR
Expectancy per trade and per 100 trades, plus the win rate you actually need to break even.
- DRAWDOWN RECOVERY CALCULATOR
The gain a drawdown demands, in cash, in R and in trades at your current edge.
THE MODEL THAT TRADES THIS
Gold Hunts runs models built on this method, so the rules above can be applied on a practice account with your own limits rather than watched by hand. Where a model has closed gold trades on Gold Hunts accounts, its recorded win rate and sample size are printed beside it — what happened, not what will.
- Pullback Continuation — Waits for a healthy pullback inside the trend before entering.
No closed gold trades recorded on this model yet.
- Trend Pullback Failure — Trades the case where a pullback fails to resume the trend.
No closed gold trades recorded on this model yet.
RISK NOTE
Leveraged gold carries a high risk of rapid loss. Every level, stop distance and session window on this page is a working reference measured on XAUUSD, not financial advice, a signal, or a projection of returns. Test any rule set on a demo account and journal it before risking capital.
GOLD TRADING FAQ
- What fields should I record for every gold pullback trade?
- At minimum: level type (fib, EMA, VWAP, swing level), retracement depth in both dollars and percent, the specific confirmation trigger used, and time-to-fill from zone arrival to entry. These four fields drive every downstream diagnostic metric.
- How do I know if my gold pullback setup has stopped working?
- Track average R by retracement-depth bucket, fill rate on resting limit orders, maximum adverse excursion on winning trades, and hit rate split by confirmation trigger type — all on a rolling twenty-trade window rather than a lifetime aggregate.
- Should I journal unconfirmed limit fills separately from confirmed entries?
- Yes. An unconfirmed limit fill — one with no rejection candle following it — typically shows a materially lower hit rate than a confirmed trigger, and merging the two categories hides exactly the evidence needed to justify requiring confirmation.
- Which retracement depth performs best on gold, according to journaling data?
- It varies by account and regime, which is the point of tracking it — bucket trades into 38.2–50%, 50–61.8% and 61.8%+ and compute average R for each. One bucket usually carries most of the expectancy, and it is not always the textbook 50–61.8% zone.
- How often should I review my gold pullback journal?
- Weekly is the right cadence for this setup, since pullback frequency is lower than scalping or breakout strategies. Monthly reviews risk acting on too small a sample, while daily reviews risk reacting to noise within a single metric.
JOURNAL THIS GOLD SETUP FREE
Live XAUUSD signals on a public ledger, a free gold trade journal that syncs with MetaTrader 5, and the macro pulse before the London open.
MORE ON GOLD PULLBACK ENTRIES
- GOLD PULLBACK ENTRIES
The institutional guide to XAUUSD pullback entries: measuring retracement depth and shape, confluence zones, limit versus stop entry, trigger candles, invalidation, target logic and two fully worked gold trades.
- BEST SESSIONS
When XAUUSD pullback entries actually work: hourly behaviour in UTC, the post-London-open retest, the NY overlap retest of the London break, a session-suitability table, day-of-week effects, DST handling and the hours to avoid.
- WHEN IT FAILS
Why XAUUSD pullback entries fail: chasing a missed zone, impulsive retracements mistaken for dips, confluence-free fib levels, entering ahead of the trigger, ignoring higher-timeframe supply and demand, and buying every dip in a range.
- RISK MANAGEMENT
Position sizing XAUUSD pullback entries on the 100-ounce contract: stops beyond the retracement extreme, split-entry sizing, deep-retracement discounting, spread cost at entry, correlation risk and prop-firm constraints.
- PSYCHOLOGY
The four behavioural failures that break XAUUSD pullback entries: waiting forever for a perfect retest, front-running the level, moving the limit order deeper, and entering a pullback that is really a reversal. A hard rule and journal prompt for each.
- WORKED EXAMPLES
Three fully annotated XAUUSD pullback entries: a clean win with textbook confluence, a valid loss that followed every rule and still lost, and a trade that should never have been taken — with level, entry, stop in dollars per ounce, target, lot sizing from a stated balance and risk percentage, R outcome and a post-trade review.
OTHER GOLD STRATEGIES
- XAU SESSION BREAKOUT
Trade the range expansion when London and New York arrive: how the Asian range sets the trap, and how gold actually breaks.
- GOLD LIQUIDITY SWEEP
The stop hunt, traded deliberately: where gold stops sit, how the sweep prints, and how to enter the reversal instead of funding it.
- GOLD MEAN REVERSION
Fading stretched XAU moves back to value: the bands and averages that work on gold, and the trend days that make this setup dangerous.
- GOLD TREND CONTINUATION
Riding the gold trend day: identifying a real XAU trend, entering with it, and holding long enough for the move to pay for the chop.
- GOLD NEWS MOMENTUM
Trading XAU around CPI, NFP and FOMC: what actually happens in the first ten minutes, and the only entries worth taking.
- GOLD RANGE SCALPING
Fade the edges of a defined gold range on thin liquidity: how to confirm the range is real, where to place stops, and why costs decide the outcome.
Hubs: all gold strategies · the complete gold trading guide · gold pairs and instruments.
RELATED ON GOLDHUNTS
- Gold Trading StrategiesCluster hub
- Journaling the Gold Liquidity Sweep — Fields, Tags, MetricsSame cluster
- Journaling Gold Range Scalps: Cost Columns and Edge DecaySame cluster
- Gold Pullback Entries — XAUUSD Retracement Entry ExecutionSame cluster
- Free Gold Trading CalculatorsCalculator
- Gold Hunts vs Tradezella: XAU Setup & Sweep TaggingComparison