Why Round-Number Reversals Fail Without a Sweep and Close Confirmation

The problem with trading round numbers too early

Fading a psychological level the moment price touches it is a common way to lose money. Price taps the 4100 handle on gold, you short it, and the next candle blows straight through and stops you out before reversing an hour later. The level was real, but you entered on the touch instead of the confirmation.

Big Figure Trap Signals is a round-number reversal indicator for MetaTrader 4 that exists for one sequence: price runs past a psychological level, takes the orders sitting behind it, and then closes back on the side it came from. The sweep through is the bait; only the return is marked.

What it does

Every closed candle is measured against a set of gates:

• Location - the distance from the nearest level on an auto-built grid of big, half and quarter figures
• Sweep - whether the candle actually traded beyond that level by the sweep distance (default 50 points), not merely touched it
• Reclaim - whether the close came back past the level and past the candle's own midpoint
• Rejection - the wick left on the swept side, as a share of the whole candle range (30 percent for regular tier, 45 percent for premium)
• Context - the candle range against ATR(14), the spacing of the surrounding levels, and optional volume, spread, session and Friday-close filters

What appears on your MetaTrader 4 chart is the level grid itself, signal arrows sized by tier, a measured trade box with entry, stop and target labels, and a separate logic window carrying the distance to the nearest level, the candle range and the reclaim wick ratio as their own lines.

The grid is not typed in by hand. It is derived from the symbol's own digits so the same settings work across instruments with very different prices. The panel names the level in play, its class and how far price sits from it. If the levels sit closer together than 1.2 times the sweep distance, price would be permanently trapped between two of them, and signals are blocked.

How it is used in a real session

You open a MetaTrader 4 chart on gold M15, attach Big Figure Trap Signals, and watch the panel. Price is sitting 58 points below the 4102.50 quarter level. The panel prints the live wick ratios for both directions and the candle range against ATR.

A candle sweeps down below the level by 50 points, closes back above the level and above its own midpoint, and leaves a lower wick that clears 30 percent of the range. The candle range sits between 0.6 and 2.5 times ATR(14), no signal has fired in the same direction in the last 6 bars, and the level spacing is adequate. A BUY arrow prints below the candle, offset by 0.30 ATR.

A trade box appears with entry at the signal candle close, stop beyond the swept extreme plus 0.25 ATR of buffer, and targets at 1R and 2R. The panel updates to show the last signal, its tier, the level it came from, and the running Profit Factor and Win Rate of the signals on that chart. Result markers will appear later where a target or stop is reached.

The whole sequence resolves on one closed candle. There is no arming state to track and nothing carries over to the next bar.

Who it fits

It suits you if:

• You already trade liquidity sweeps and stop hunts around round numbers, and want the criteria written down instead of judged by eye
• You prefer a graded signal - premium versus regular - to a single yes or no
• You want the stop anchored to the swept extreme rather than a fixed distance
• You work intraday on instruments where round numbers are watched: gold, FX majors and indices

Working ranges: M5 to H1 suits the default grid on most instruments, with M15 the reference timeframe used in testing. On symbols with unusual digits or synthetic pricing, check the grid line spacing on the chart before trading it, and set the big-figure step by hand if the auto value does not match the levels you actually watch.

It suits you less if you trade breakouts. A genuine break that keeps running is exactly the case this indicator is built to stay silent through, and during a strong trend it will sit quiet for long stretches.

What it does not do

Big Figure Trap Signals does not mark the level touch itself. Until price has gone through the level by the sweep distance, there is nothing to reclaim, and no arrow can appear. If you enter on the initial contact with the round number, you are on your own.

It does not adapt the grid to intraday supply and demand zones, only to round numbers. If the real liquidity sits at 4103 instead of 4100, the indicator will not see it.

It does not send pop-up alerts. The panel carries the state, and the gates are meant to be read before the arrow, not after it. If you need mobile notifications, you will need to add them separately.

It does not guarantee that the reversal will hold. A candle that meets every gate can still be run over by news or a stronger trend. The stop is part of the logic for a reason.

How it pairs with other tools

Big Figure Trap Signals focuses on one pattern at round numbers. For traders who also work with trendline breaks and fakeouts, Swing Trendline Break & Fakeout MT4 marks a different structure - the diagonal instead of the horizontal. The two do not overlap in what they watch, and they can run on the same chart without conflict.

For traders who want a trend filter before taking a reversal signal, Adaptive ATR Trend MT4 can provide a directional overlay. It will not improve the reversal win rate by itself, but it can help you skip counter-trend signals during a strong move.

Installing it

Copy the .ex4 file into the MQL4/Indicators folder inside your terminal data folder, refresh the Navigator panel in MetaTrader 4, then drag Big Figure Trap Signals onto a chart. If you were issued an Access Key, paste it into the Access Key input. Every threshold - sweep distance, wick ratio, ATR multipliers, cooldown length - is editable in the Inputs tab.

---

Risk disclosure: This indicator is provided for educational purposes only and is not financial advice. Trading involves risk; past performance and backtested behaviour do not guarantee future results.