Candle backtests guess which of your stop or target was hit first. Ticks know. Replay XAUUSD on every bid and ask quote since 2016, with stop-loss and take-profit settled on the exact tick, real spreads and 5-second charts. Right in your browser.
We replayed 590,287 random trades on gold from 2016 to 2025, at a 1:2 reward-to-risk. Random entries have no edge, so the honest answer is zero. Here is what each method reported.
per trade. Every candle that touched both levels was booked as a win, so random trades looked like a profitable system.
per trade. Settled on the real tick order, the same trades came out at break-even, which is exactly right for random entries.
| Setup | Trades | Both levels in one candle | Stop hit first | Win rate (candles) | Win rate (ticks) |
|---|---|---|---|---|---|
| M15 · 1:2 | 472,317 | 13,244 (2.8%) | 66.7% | 35.3% | 33.4% |
| H1 · 1:2 | 117,970 | 4,592 (3.9%) | 66.7% | 35.9% | 33.3% |
| M15 · 1:1 | 472,250 | 5,912 (1.3%) | 50.0% | 50.6% | 50.0% |
| H1 · 1:1 | 117,944 | 2,526 (2.1%) | 50.0% | 51.1% | 50.0% |
Method: XAUUSD bid ticks, January 2016 to December 2025. A long and a short at every candle open; stop and target sized from the previous 14 candles' average range (1:2 = stop 0.5 ATR, target 1 ATR). The first candle reaching either level decides the trade; where it reaches both, candle rules book the target and the ticks are scanned for which level traded first. Costs excluded so only fill order is measured. Tighter stops, common on gold scalps, push more trades into these candles.
You replay candles and trade with the mouse the way you always have. The tick layer sits underneath and settles the moments candles cannot.
Every stop and target settles on the first tick that trades the level. Longs exit on the bid, shorts on the ask, just as a broker fills them.
Market entries pay the bid-ask spread actually quoted on the candle's last tick, wider in Asia and around news, instead of one flat number.
S5, S15 and S30 timeframes built straight from the ticks. Zoom into a London open or an NFP spike and see the order the moves came in.
Over 560 million quotes through Brexit, the 2020 crash and the record run of 2024-2026. Test across regimes, not one quiet year.
No tick downloader, no CSV import, no MetaTrader history settings. Each day's ticks stream into the chart the moment the replay needs them.
Win rate, expectancy, profit factor and drawdown from tick-settled trades, plus the full Ultra suite: sessions, MFE, tags and stress tests.
An M1 bar on a news release hides a whole fight. S5 shows it: the first spike, the stop run, the reversal. Scalpers finally get to test entries at the speed they actually take them.
If you already backtest gold on candles, run the same rules again on ticks. An edge that survives is one you can size with confidence.
With Ultra active, the tick layer switches on for gold automatically and the seconds timeframes appear in the switcher.
S5 for scalping the open, M15 or H1 for intraday swings. Everything after your start date stays hidden.
When price reaches your stop or target, the day's ticks decide which traded first, and the report is built on those fills.
Tick backtesting has meant installing a terminal, downloading history and writing an EA. Not here.
| Candle bar replay | MT5 tester (real ticks) | FxBacktest Ultra | |
|---|---|---|---|
| SL and TP in one candle | Target assumed | Real order | Real order |
| Spread | Fixed number | Broker history | Real, per tick |
| Seconds timeframes | No | Custom symbols only | S5 / S15 / S30 |
| Manual bar-by-bar trading | Yes | EA-driven | Yes |
| Gold history | Varies | Depends on broker | Every day since 2016 |
| Setup | Browser | Install, account, EA | Browser |
One plan, the full analytics suite, and every gold tick since 2016.
Not ready? Gold candles are free on every timeframe. How to backtest gold
What traders ask before switching gold to ticks.
No. The ticks stream into the backtesting chart, where they settle your trades and draw the seconds timeframes. They are not offered as a file download. If you need raw CSV files for your own code, use a tick downloader; if you want to replay and trade gold on those ticks, this is the tool.
To January 2016 - every trading day of XAUUSD bid and ask quotes since then, more than 560 million ticks in total.
Stop-loss and take-profit exits are settled on the first tick that trades the level, longs on the bid and shorts on the ask. Market entries take the real spread on the last tick of the candle. Pending limit and stop entries still fill on candle rules, and if a day's ticks are incomplete that candle falls back to candle rules too.
No. XAUUSD is on the free plan with candle data on every timeframe; minute bars cover the most recent months on the free plan. Ultra adds the tick layer: tick-accurate exits, real per-tick spread and the S5, S15 and S30 seconds timeframes.
Gold is the first and currently the only symbol with tick data. It was chosen because its fast candles are where candle-based fills go wrong most often. Meanwhile, every market is on candles - see all 105 markets.
Run your strategy on the ticks that actually printed. If the edge holds, trade it with confidence. If it does not, you just saved your account.