# Trading Liquidity Sweeps with Mirage Stops cluster above swing highs and below swing lows. When price runs those stops and snaps back, the move often turns. Mirage Liquidity Sweep Pro finds those moments, grades them and hands you a trade plan; this lesson is about how it decides, how to read what it draws, and how to trade it without fooling yourself. ## Why sweeps happen Look at any obvious swing high. Traders who are short put their stops just above it, breakout traders park buy orders there, and anyone watching the chart can see the level. That pile of resting orders is buy-side liquidity (BSL). Under a swing low sits the opposite: stops of the longs and sell orders of breakdown traders, the sell-side liquidity (SSL). A large player who wants to sell needs buyers. Pushing price up through the high triggers all those buy orders at once, which gives the seller someone to sell to. Once the orders are used up, nothing is left holding price up, and it falls back below the level. Price doesn't turn at random lines on a chart; it tends to turn where resting orders get used up. ## Sweep or break: the close decides On the chart a sweep looks like a wick through the high with a close back underneath. The mirror image, a wick under a swing low that closes back above, is a bullish sweep. Mirage marks the spot with a small **×** at the wick and a dashed line along the swept level, with a thick tick over the part of the wick that went through it. A candle that closes beyond the level, or exactly on it, is different. Buyers or sellers were strong enough to hold the new price, so that is a break, not a sweep. Mirage removes the level from its map and gives no signal. This one rule, judging by the close and not by the wick, filters out most of the noise. ## How Mirage grades a sweep A wick through a level happens all day, and most of them lead nowhere. So every sweep gets a score from 0 to 100 before it can become a signal. Five things go into it: - **Wick depth, 30%.** How long the rejection wick is compared with the ATR. A wick of one ATR or more gets full marks. - **Reclaim, 25%.** How far the close got back past the swept level, again measured in ATR. A close just a hair inside is weak; a close well back inside is strong. - **Close position, 20%.** Where the candle closed within its own range. For a bullish sweep, a close near the high is best; for a bearish one, near the low. - **Volume, 15%.** Volume against its 21-bar average. 1.5 times the average or more counts in full. On markets without volume this part stays neutral. - **Higher-timeframe bias, 10%.** Whether the sweep goes with the trend on the higher timeframe (4h close against its 50 EMA by default). It only nudges the score, so a very clean counter-trend sweep can still pass. A sweep needs at least **Min sweep score**, 50 by default, to count. That single number is the main quality dial. Hover a signal to see its score; with **Score labels** turned on in the Style tab, it is printed under each signal. ## Confirmation: let price prove it A good sweep says liquidity was taken. It does not yet say price has turned. With **Require CHoCH confirmation** on, which is the default, a qualifying sweep becomes a pending setup. The signal only appears when price then closes through the last minor pivot in the new direction, a change of character (CHoCH), within 13 bars. The minor pivots use their own, shorter length (8). Three things can stop a pending setup: - **No CHoCH within 13 bars.** The setup expires. - **Price goes back beyond the sweep wick.** The idea is broken and the planned stop would already be hit, so the setup is cancelled. - **The stop would end up on the wrong side of the entry** after a late confirmation. The setup is rejected rather than patched. A bar that sweeps a high and a low at the same time has no clear direction and is skipped. With the confirmation off, the signal comes on the sweep bar itself: earlier, but you are trusting the wick alone. ## The trade plan on the chart When the signal comes, Mirage draws the whole plan at once: - **Entry** at the close of the signal bar, the **ENTRY** label. - **Stop** beyond the sweep wick: the wick low minus ATR × buffer for a long, the wick high plus ATR × buffer for a short. If that puts the stop closer than half an ATR to the entry, it is widened to half an ATR, so you don't get a stop that any normal candle would hit. - **TP1, TP2 and TP3** as multiples of the risk (R), the distance from entry to stop. The labels show the price and the distance from entry in %. A target that has been touched turns solid teal with ✓. - **Liquidity Target**, the gold dashed line: the nearest untouched pool of liquidity on the other side, the closest BSL above a long or SSL below a short. It is where the move is likely drawn, separate from the fixed R targets. The **Risk preset** sets the buffer and the three targets: - Conservative: buffer 0.5 × ATR, targets 1R / 2R / 4R. - Balanced (the default): 0.25 × ATR, 1R / 2R / 3R. - Aggressive: 0.15 × ATR, 1.5R / 2.5R / 4R. - Scalping: 0.1 × ATR, 0.8R / 1.5R / 2R. - Custom: your own values, with TP1 < TP2 < TP3. Around the plan you see the liquidity map: dotted **BSL** and **SSL** lines at swing highs and lows that haven't been swept yet, six per side by default, and **EQH** / **EQL** connectors where two swings sit almost level. Equal highs and lows are the strongest magnets, because twice as many stops sit there. A level disappears from the map the moment it is taken. ## Step 1: set up the chart Pick one market and one timeframe and stay with it for a while. 15m to 4h is a good range to start; on 1m and 5m the sweeps are smaller and noisier. Add Mirage from **Indicators** › **Price action** and leave the defaults alone at first: Balanced preset, Min sweep score 50, CHoCH confirmation on. With Mirage on, the chart loads older history in the background, about 4,000 bars, so the map and the statistics have more to work with. The dashboard sits folded in the top-right corner as **◆ Mirage · Bullish** or **Bearish**. Tap or click it to open. Check **HTF bias** before anything else. A long while the 4h trend is bearish is not forbidden, but it has to be cleaner to be worth it. ## Step 2: wait for a signal, then check it Most of the time the dashboard shows **Wait**, and you should too. When a **Long ▲** or **Short ▼** appears, run through four questions before you act: 1. **How good is the sweep?** A 52 just passed the filter; a 75 or more is a much cleaner grab. 2. **Where is the Liquidity Target?** If the gold line sits closer than TP1, the move has little room before it reaches the next pool of orders. Skip it or plan to take profit early. 3. **How wide is the stop?** **SL dist %** tells you. A 4% stop on a 15m chart means a huge wick, and your position has to be small to keep the risk the same. 4. **Is anything due?** A sweep right before FOMC or CPI can be swept again by the news. The countdown in the status bar and **Panels** › **News & events** show what is coming. If any answer bothers you, pass. There will be another setup. ## Step 3: size the position from the stop Decide first how much you are willing to lose if the stop is hit, for example 1% of the account. The position size follows from that and from the stop distance, never the other way round. Take the long in the picture above, BTC on 1h with the Balanced preset: - Entry 82,269.1, stop 80,534.8. The stop is 1,734.3 away, 2.11%. - TP1 84,003.4 (1R), TP2 85,737.8 (2R), TP3 87,472.1 (3R). With a $5,000 account and 1% risk, you can lose $50. $50 ÷ 1,734.3 = about 0.029 BTC, a position of roughly $2,370. At 5× leverage that uses about $475 of margin, but the leverage does not change the risk: if the stop is hit you lose about $50 plus fees. Fees count. Every fill pays the exchange fee, and orders from the Market Open chart on Hyperliquid also pay the 0.055% Himbit fee, so a trade closed at TP1 earns a little less than 1R. ## Step 4: manage the trade Mirage moves its stop to entry when TP1 is touched. From the next bar, a wick back to entry closes the trade at break-even; the SL line dims and the entry label reads "→ SL (BE)". Otherwise the trade ends at the stop or at TP3. You don't have to copy that exactly, but write your own rule down before you enter. Three common ones: - **Take part at TP1, move the stop to entry, let the rest run** to TP2 or TP3. This matches what Mirage draws. - **Use the Liquidity Target as the exit** when it sits between TP1 and TP3. Price is drawn to untouched liquidity, and it often stalls there. - **Take everything at TP1.** Less upside, but simple and consistent, and it matches how the dashboard counts a win. Mirage does not place orders. If you trade from the chart, set the stop and targets yourself: drag the TP and SL boxes on your position line to the levels Mirage drew. See [On-chart trading](#trade-on-chart). ## A worked example: TP1, then break-even The picture below is BTC on 1h with the confirmation turned off and score labels on, so every qualifying sweep trades and shows its score. Read it from left to right. Three longs came from swept lows, with scores of 86, 77 and 70. Then price ran up through a swing high and closed back below it: a bearish sweep scored 78, and a **Short ▼** at 82,694.9. The stop went just above the sweep wick at 83,245.2 (+0.67%), so 1R was 550.3 points and TP1 sat at 82,144.6 (−0.67%). Price dropped and touched TP1, the label turned teal with ✓ and the stop moved to entry. A few bars later price bounced back to 82,694.9 and the trade closed at break-even: "ENTRY → SL (BE)". TP2 and TP3 were never reached. The dashboard counts this trade as a win, because TP1 was touched; in money, a trader who took nothing at TP1 ended flat minus fees. Over the 7 closed trades on this chart the dashboard shows 4 wins and 3 losses, 57.1%. Two lessons in one picture. Taking part of the position at TP1 is what turns a "win" like this into real profit. And the win rate tells you how often TP1 is reached, not how much you made. ## Four setups worth watching - **Equal highs and lows.** When an EQH or EQL is swept and reclaimed with a high score, the rejection is often clean, because a double layer of stops was taken at once. - **Draw on liquidity.** Use the gold Liquidity Target as a place to take part of the position; price often travels to the next untouched pool before it pauses. - **Confirmation on or off.** Keep CHoCH on for fewer, more selective entries on 1h and above. On lower timeframes some traders turn it off for faster entries and accept more false starts. - **Trade with the bias.** Take only signals whose direction matches **HTF bias** on the dashboard. You get fewer signals, but each one has the bigger trend behind it. ## Step 5: let alerts do the waiting Staring at a chart for hours leads to forced trades. Right-click the chart, pick **Add alert on Mirage Liquidity Sweep Pro** and choose **Long entry** and **Short entry**. Add **TP1 hit**, **Break-even activated** and **Stop loss hit** if you want to follow an open plan. Alerts fire on bar close, and only while Market Open is open in your browser. ## Tuning without curve-fitting After a few weeks you will want more signals or better ones. Change one setting at a time and give it time: - **Fewer, cleaner signals:** raise Min sweep score to 60 or more, keep CHoCH on and the volume factor on. - **More signals:** lower Min sweep score to 40–45 or shorten Swing length. Turning CHoCH off gives earlier entries but more false starts. - **Choppy market:** lengthen Swing length so only bigger swings count, and lean on the HTF bias. - **Stops too tight or too wide:** switch the preset. Conservative gives the stop more room, Scalping tightens it and brings the targets closer. Don't pick the settings that make the dashboard's win rate highest on the bars you can see. That is fitting the past, and it rarely holds. ## Honest limits - **Statistics are a simulation.** They run on the bars loaded in the chart, about 4,000, with no fees, slippage or funding, and they start over when the chart reloads. It is not a broker-accurate backtest. - **A win means TP1 was touched**, even if the trade then closed at break-even. The win rate is not your profit. - **One trade at a time.** Sweeps that happen while a trade is open are drawn but not traded. - **No repainting, but a delay.** Mirage only uses closed bars, so a signal never disappears afterwards. A swing high or low is only confirmed Swing length bars after it formed, so the map lags behind the chart by design. - **Your fill can be worse.** The entry is the close of the signal bar; in a fast market you may get a worse price. - **Trends sweep and keep going.** A strong trend can sweep the same level twice and carry on. The stop is there for that. ## Try it on Market Open 1. Open BTC on the 1h chart and add Mirage Liquidity Sweep Pro with the default settings. 2. Scroll back through the loaded history. For every Long and Short, hover the marker and the lines, and write down the score, the SL dist % and whether price reached the Liquidity Target before the stop. 3. Turn the confirmation off and score labels on, and compare: how many more signals, and how many of the extra ones reached TP1? 4. Pick a risk per trade and work out the position size for the last three signals, the way Step 3 does. 5. Set alerts for Long entry and Short entry and paper-trade the next five signals in the demo account ("Try demo") before you use real money. _Educational content, not investment advice. Leveraged trading can lose more than your margin._