Swing-Point Engulfing Candles: Context, Confirmation and Risk

Video: 🔴 The GREATEST Candlestick Pattern on EARTH
Creator: Trader DNA
Original Video: Watch the original video
Difficulty: Intermediate

The Big Idea

Trader DNA combines a two-candle reversal with a price extreme and surrounding context. The bearish version appears at a swing high; the bullish version appears at a swing low. The source then looks for support or resistance and oscillator conditions before entering on the following candle.

The useful lesson is that a candle is an entry component, not a complete edge. The source itself cautions against treating the pattern as a holy grail. Its title and effectiveness claims are not supported by a complete test dataset in the supplied transcript.

If You Remember One Thing

A low that looks obvious afterward may not have been identifiable as the final swing low at entry. Define what was knowable when the reversal candle closed, then test the actual order and stop—not an idealized turning point.

The Source’s Two Patterns

The transcript varies its abbreviations. We use SHEC for swing-high engulfing candle and SLEC for swing-low engulfing candle to keep the explanation consistent.

“Larger Body” and “Enclosing Body” Are Different Rules

The source defines its engulfing event as a color change plus a larger body, explicitly excluding wicks from the body comparison. That does not mathematically require the second body to contain the first.

Suppose a green candle opens at $100 and closes at $101. Its body is $1. A red candle opening at $104 and closing at $102 has a $2 body: it is larger and the color reversed, but it does not enclose the $100–$101 body.

Full enclosure is a stricter rule. For a bullish reversal, it requires the new open at or below the preceding red close and the new close at or above the preceding red open; the bearish version mirrors those inequalities. Our script defaults to the source’s larger-body rule and provides an optional full-enclosure comparison. Do not change definitions after inspecting winners.

Define the Swing Without Looking Ahead

The source describes the highest or lowest point of a trend or pullback. Such a point is easy to label on a completed chart, but the final turning point is uncertain while the trade is unfolding. Price may make another extreme immediately afterward.

Two research approaches answer different questions. A confirmed pivot waits for later bars to establish a local turn, so an entry must occur after that confirmation. A developing extreme uses only past and present bars, permitting an earlier entry but accepting that the turn may fail.

Our baseline uses the second approach. It compares the two-candle pattern’s lowest low or highest high with the preceding 20 bars, excluding both pattern candles. This creates an observable extreme candidate. It does not prove that the pullback has ended or reconstruct the entire trend structure described in the source.

Structure Should Exist Before the Signal

The source looks left for an established support or resistance area. That is important: defining a “major level” after a successful reversal invites hindsight selection.

Our simpler structure proxy is the older rolling 20-bar low or high. The pattern extreme must be within 0.25 preceding ATR(14) of that reference when the structure filter is enabled. A rolling extreme is not equivalent to a discretionary major zone with repeated reactions.

For example, if the reference low is $49.80 and prior ATR is $1, the permitted tolerance is $0.25. A two-candle low at $49.70 passes both the fresh-extreme and proximity tests. A low at $49.20 makes a new extreme but is too far below that old reference for this baseline.

The Oscillator Adds Context, Not Certainty

The source names RSI, CCI and stochastic, with stochastic illustrated in the examples. It describes overbought or oversold conditions and divergence but supplies no exact oscillator settings or complete divergence algorithm.

Our baseline chooses a 14-bar stochastic with three-bar smoothing of %K. For a bullish event, the preceding candle’s smoothed %K must be below 20; for a bearish event, above 80. These settings and the use of the preceding bar are experimental choices. The reversal may move the oscillator out of the extreme by the signal close.

Oversold does not mean price must rise, and overbought does not mean it must fall. Divergence also requires specified price and oscillator reference points. If those points use future-confirmed pivots, their information is available only after the confirmation delay.

The script omits divergence rather than pretending an unspecified visual comparison has been reproduced. Price location, candle shape and an oscillator are related transformations of price, not necessarily independent evidence.

Worked Example: Signal, Fill and Protection

Imagine an eligible support-area sequence. The first red candle opens at $50.20 and closes at $49.80. The next green candle opens at $49.75 and closes at $50.30, making a $0.55 body versus the prior $0.40 body. Assume its pair low is $49.60 and the structure and oscillator tests pass.

With a one-cent tick, the baseline places protection at $49.59. If the next market fill is $50.35, the fill-to-stop distance is $0.76 per share. At 100 shares, planned price risk is $76 before costs. A fill at $50.80 raises that distance to $1.21 and planned price risk to $121.

The candle pattern has not changed, yet actual entry risk has. A gap through the protective level can create a worse exit than planned. Position exposure, stop distance and realized loss are separate quantities.

The source does not specify a complete target, trailing stop or holding rule. Our script therefore uses a ten-bar timed exit as an explicit research choice, with the actual fill bar counted as bar one. It makes no fixed reward-to-risk claim.

Claims and Supplied Evidence

The transcript presents three favorable examples and calls the approach effective or high probability. It does not provide every eligible signal, losing trades, sample size, costs, instruments, dates or out-of-sample results needed to establish that claim. The source’s illustrated trades are reported examples, not a verified track record.

A pattern may produce a positive initial reaction and still lose under an executable exit. Count completed trade outcomes separately from short-term favorable movement. No candlestick label by itself establishes profitability.

The Pine Research Baseline

//@version=6
strategy("Swing-candidate larger-body reversal: research baseline", overlay=true,
     initial_capital=100000, default_qty_type=strategy.percent_of_equity,
     default_qty_value=10, pyramiding=0, margin_long=100, margin_short=100,
     commission_type=strategy.commission.percent, commission_value=0.05,
     slippage=2, process_orders_on_close=false,
     calc_on_every_tick=false, calc_on_order_fills=false)
int lookback = input.int(20, "Historical extreme lookback", minval=5)
bool strictBody = input.bool(false, "Require full prior-body enclosure")
bool useStructure = input.bool(true, "Require proximity to older rolling extreme")
float toleranceATR = input.float(0.25, "Structure tolerance / prior ATR", minval=0.0, step=0.05)
bool useOscillator = input.bool(true, "Require preceding stochastic extreme")
int kLength = input.int(14, "Stochastic length", minval=2)
int kSmooth = input.int(3, "K smoothing", minval=1)
int holdBars = input.int(10, "Experimental timed-exit bars", minval=1)
string side = input.string("Both", "Direction", options=["Both", "Long only", "Short only"])

float atr = ta.atr(14)
float k = ta.sma(ta.stoch(close, high, low, kLength), kSmooth)
// All historical references exclude BOTH candles of the pattern.
float referenceLow = ta.lowest(low[2], lookback)
float referenceHigh = ta.highest(high[2], lookback)
float pairLow = math.min(low, low[1])
float pairHigh = math.max(high, high[1])
float body = math.abs(close - open)
float priorBody = math.abs(close[1] - open[1])
bool bullish = close[1] < open[1] and close > open and body > priorBody
bool bearish = close[1] > open[1] and close < open and body > priorBody
if strictBody
    bullish := bullish and open <= close[1] and close >= open[1]
    bearish := bearish and open >= close[1] and close <= open[1]
bool ready = bar_index >= lookback + 2 and not na(k[1]) and atr[1] > 0
// These are developing extreme candidates, not future-confirmed pivots.
bool lowCandidate = pairLow <= referenceLow
bool highCandidate = pairHigh >= referenceHigh
bool supportOK = not useStructure or math.abs(pairLow - referenceLow) <= toleranceATR * atr[1]
bool resistanceOK = not useStructure or math.abs(pairHigh - referenceHigh) <= toleranceATR * atr[1]
bool oversold = not useOscillator or k[1] < 20
bool overbought = not useOscillator or k[1] > 80
var float frozenStop = na

if barstate.isconfirmed
    if strategy.position_size != 0
        int filledBar = strategy.opentrades.entry_bar_index(0)
        if bar_index - filledBar + 1 >= holdBars
            strategy.close_all(comment="Experimental timed exit", immediately=true)
    else if ready
        if bullish and lowCandidate and supportOK and oversold and side != "Short only"
            frozenStop := pairLow - syminfo.mintick
            strategy.entry("Long", strategy.long)
            strategy.exit("Long protection", from_entry="Long", stop=frozenStop)
        else if bearish and highCandidate and resistanceOK and overbought and side != "Long only"
            frozenStop := pairHigh + syminfo.mintick
            strategy.entry("Short", strategy.short)
            strategy.exit("Short protection", from_entry="Short", stop=frozenStop)
// Protection is frozen at submission, not recalculated from later patterns.
// No divergence, take-profit, additions, trailing stop or session liquidation.
plot(referenceLow, "Older rolling low", color=color.green)
plot(referenceHigh, "Older rolling high", color=color.red)
plot(strategy.position_size != 0 ? frozenStop : na, "Frozen protective stop", color=color.orange, style=plot.style_linebr)

Use standard candlesticks and choose one market and timeframe before examining results. Source rules are color change, a larger body, extreme location and contextual checks. Our rolling lookback, ATR tolerance, stochastic settings, two-candle protective level, exposure and timed exit are experimental definitions.

What This Code Actually Tests

It tests larger-body reversals at developing historical extremes, optionally near older rolling structure and following a stochastic extreme. Signals use completed candles. Market entries normally fill at the next bar’s open, consistent with the source’s explicit bullish next-candle entry.

Protection is submitted with the entry and remains frozen beyond the two-candle extreme. The strategy holds one position at a time; later signals do not replace its stop or add exposure. The timed exit counts from the actual fill bar. No target is calculated from an assumed signal-price fill.

The default 10% equity sizing controls entry exposure, not a fixed percentage of capital at risk. Actual stop distance and product specifications determine the planned loss. The script permits separate long-only and short-only comparisons.

What This Code Does Not Test

It does not recreate discretionary trend and pullback identification, major repeated support zones, oscillator divergence or the exact unseen chart examples. It does not wait for a future-confirmed swing, detect every valid local pullback or classify trend-following versus countertrend trades.

There are no adds, trailing stops, profit targets, session restrictions or end-of-session liquidation. Positions can remain open overnight; ten bars have different elapsed durations across timeframes. Borrow availability and fees for shorts require separate checks. A profitable result would support only this specified experiment.

Execution and Validation

Submission occurs after the signal close; entry fills can differ from that close. Absolute protective orders are attached immediately at submission, allowing protection on the fill bar. The frozen stop may produce an unfavorable or immediate exit if the next open is beyond it. The baseline does not retroactively reject a future fill it could not know at submission.

The timed exit uses an immediate closing-tick market close, a simulation shortcut that may not match a live closing fill. Defaults are 0.05% commission per fill and two ticks of slippage. Review these assumptions and same-bar entry/stop ordering using TradingView’s strategy documentation (paid link).

No higher-timeframe request or future pivot is used. For why backdated swing labels and developing bars require care, see TradingView’s repainting documentation (paid link). Bar Magnifier may clarify ambiguous intrabar paths where suitable data are available. Use standard candles rather than synthetic prices.

Validation status: The script and order logic have been statically reviewed and the full code interface checked. It has not been compiled or backtested in TradingView. No measured returns or success rates are claimed.

Backtest Lab

1. Pattern definition: Compare larger-body-only and full-enclosure modes with identical context and exits. Record trade counts, actual fill-to-stop risk, expectancy and drawdown after costs.
2. Context contribution: Compare structure and oscillator filters individually and together. Keep the base extreme rule fixed and review excluded winners as well as avoided losses.
3. Definition robustness: Compare lookbacks of 10, 20 and 40 bars or holding exits of five, ten and twenty bars in separate experiments. Use enough prior history and a common evaluation start.
4. Execution stress: Raise commission to 0.10% and slippage to five ticks. Inspect gaps, same-bar stops and overnight exposure across several liquid instruments and distinct regimes.

Reserve an untouched later period before selecting settings. Record the number of comparisons made. A good-looking result chosen from many combinations is weaker evidence than stable behavior across nearby definitions and fresh observations.

Check Your Understanding

1. Does a larger opposite-color body necessarily enclose the preceding body? No. Full enclosure imposes additional price conditions.
2. Can the final swing low be known immediately? Not with certainty. An observable extreme candidate can make a new low later.
3. Does stochastic below 20 guarantee a reversal? No. It is context, not a required future outcome.
4. Do three profitable illustrations establish an edge? No. Complete samples and executable rules are needed.
5. Does a small candle stop guarantee a small monetary loss? No. Fill price, quantity, gaps and costs affect the result.

Practice Exercise

Review 30 consecutive opposite-color pairs. Before revealing future bars, record body sizes, whether full enclosure occurs, the older rolling reference, pair extreme and preceding oscillator reading. Preserve valid losers and attractive patterns rejected by the rules.

Then compare signal-close and actual-fill stop distances. The goal is to see how a visually simple entry changes once future information is removed and realistic execution is added.

Final Takeaway

The source’s useful structure is a reversal event at an extreme, supported by previously identified context and an explicit protective level. Its advertised effectiveness remains a hypothesis. Define swings without hindsight, preserve the source’s body rule accurately, and test the complete trade rather than a few appealing candles.

Risk Disclosure

This guide is educational and is not personalized financial advice. Trading involves substantial risk; leverage and short selling can cause losses beyond initial capital. Stops do not guarantee prices or planned loss limits. Hypothetical examples, reported trades and historical simulations do not guarantee future performance.

CandleChasers is an independent educational publisher and is not affiliated with Trader DNA. The examples, experimental definitions and research exercises here are CandleChasers’ independent analysis of the supplied material.