Trend, Pullback, Trigger: Filtering Indicator Signals in the Right Order

Video: Once You Master This Strategy, Trading Becomes SO SIMPLE …
Creator: TradeGenius
Original Video: Watch the original video
Difficulty: Intermediate

The Big Idea

An arrow says a condition occurred; it does not establish that the trade fits the market. This lesson organizes three checks: identify the permitted direction, wait for a pullback to the trend area, and only then evaluate a matching trigger.

If you remember one thing: matching direction is necessary in this framework, but a matching arrow before the pullback is still rejected.

Why the Arrow Cannot Make the Whole Decision

The source shows the same signal tool producing appealing moves, countertrend entries, and alternating signals in sideways conditions. The arrow’s visual appearance does not distinguish those environments. Selection therefore requires information beyond the trigger itself.

That does not make every rejected arrow objectively bad. A countertrend signal can win, and a filtered signal can lose. The proposed benefit is a change in the distribution of opportunities, not the elimination of uncertainty. A few selected chart examples cannot establish that distribution.

The lesson pairs Zero Lag Trend Signals by AlgoAlpha with Swing Arm HT Trend Indicator. The named tools have separate jobs: one supplies signal events, and the other supplies the directional and pullback reference. Do not confuse either with the creator’s separately promoted membership indicator.

Preserve the Source Settings Without Inventing the Algorithm

Those settings do not disclose every calculation inside either published script. In particular, an ATR-based trend tool with modified trailing behavior is not automatically identical to TradingView’s built-in Supertrend. An independently written proxy must be labeled as such rather than presented as a clone.

The extra timeframe table also does not automatically create five additional entry filters. In the source’s described decision sequence, the hourly trend band has the central directional role. Adding a unanimous vote across every displayed timeframe changes the method and should be investigated separately.

Order Matters More Than Stacking Conditions

Imagine a bearish hourly reference. A sell arrow appears while price is already stretched below it. Direction matches, but the pullback condition is absent. A later rally touches the reference area, and only afterward a fresh sell trigger occurs. The second sequence fits the rule; the first does not.

A simple Boolean expression that asks whether all conditions were true somewhere recently can accidentally accept the wrong sequence. The implementation needs state: when was the pullback observed, how long does it remain valid, and what resets it? Otherwise an old touch may authorize arrows indefinitely.

The indicator uses a finite waiting window and clears it when the confirmed higher-timeframe direction changes. One accepted signal consumes the pullback. Another accepted signal needs a new touch. This is an explicit definition for investigation, not a rule fully specified by the transcript.

Worked Example: Direction Does Not Fix Reward

Consider a hypothetical bearish reference near $102.00. Price rallies into the area and later produces a sell trigger at $101.70. If the chosen invalidation is $102.20, planned price risk is $0.50 per share. A 2R objective would be $100.70 before costs.

If a nearby support area sits at $101.30, there is only $0.40 of clear room, or 0.8R, before that obstacle. A nominal 2R target does not remove the support area. Direction, pullback, and trigger can all agree while the actual trade remains unattractive under a room-to-target rule.

A short fill at $101.55 with the same $102.20 stop increases risk to $0.65. The unchanged $100.70 target offers $0.85, about 1.31R. Always distinguish signal-close planning from actual execution. The indicator places neither an entry nor a bracket, so its marker cannot establish either fill.

Confirmed Hourly Context Has a Timing Cost

On a fifteen-minute chart, the current hourly trend value may still be developing. A line that looks settled on a historical chart can have fluctuated during the live hour. The proxy below instead requests the previous completed hourly state and holds it until a new confirmed value becomes available.

This avoids using the hour’s final result early, while necessarily reacting later to a developing reversal. That delay is part of the definition, not a bug to remove by enabling future data. TradingView’s higher-timeframe documentation (paid link) explains the confirmed-value offset pattern.

Sideways Markets Still Require Respect

The source presents the trend-and-pullback rule as a way to filter countertrend and poorly timed arrows. It does not establish a perfect ranging-market detector. An hourly trend state may persist while the smaller chart chops sideways, and a lagging reference can change after several losses have already occurred.

Observe how often signals cluster, how far price travels after confirmation, and whether the trend reference is repeatedly crossed. If you add a separate range filter later, compare it against the same baseline opportunity set instead of giving it credit for unrelated rule changes.

Pine Script Indicator

//@version=6
// Guide: https://candlechasers.com/trend-pullback-trigger-filter/
// Original educational indicator, not an exact reconstruction or a trading system.
indicator("CC Trend Pullback Trigger Map", overlay=true, max_labels_count=100)
htf = input.timeframe("60", "Confirmed trend timeframe")
if timeframe.in_seconds(htf) <= timeframe.in_seconds()
    runtime.error("Trend timeframe must be higher than the chart timeframe; try 15m chart and 60m trend.")
length = input.int(150, "Zero-lag EMA length (proxy)", minval=2)
bandMult = input.float(1.7, "Chart ATR band width", minval=0.1)
atrLength = input.int(5, "HTF Supertrend ATR length (proxy)", minval=1)
factor = input.float(5.0, "HTF Supertrend factor (proxy)", minval=0.1)
tol = input.float(0.5, "Pullback tolerance in chart ATR", minval=0)
life = input.int(12, "Trigger window after touch", minval=1)
confirmedTrend() =>
    [lineValue, dirValue] = ta.supertrend(factor, atrLength)
    [lineValue[1], dirValue[1]]
[trendLine, direction] = request.security(syminfo.tickerid, htf, confirmedTrend(), lookahead=barmerge.lookahead_on)
lag = int(math.floor((length - 1) / 2))
zeroEMA = ta.ema(close + (close - close[lag]), length)
atr = ta.atr(14)
bandATR = ta.atr(length)
upCross = ta.crossover(close, zeroEMA)
downCross = ta.crossunder(close, zeroEMA)
var int touched = na
var int lastDir = 0
var int accepted = 0
bool buy = false
bool sell = false
if barstate.isconfirmed and not na(direction)
    int currentDir = direction < 0 ? 1 : -1
    if currentDir != lastDir
        touched := na
    lastDir := currentDir
    if not na(touched) and bar_index - touched > life
        touched := na
    // Require the trigger on a later bar, not before or during the first touch bar.
    if not na(touched) and bar_index > touched
        buy := currentDir == 1 and upCross and close > trendLine
        sell := currentDir == -1 and downCross and close < trendLine
        if buy or sell
            accepted += 1
            touched := na
    if na(touched) and not buy and not sell and low <= trendLine + atr * tol and high >= trendLine - atr * tol
        touched := bar_index
plot(trendLine, "Confirmed HTF Supertrend proxy", color=direction < 0 ? color.green : color.red, linewidth=2, style=plot.style_stepline)
plot(zeroEMA, "Zero-lag EMA proxy", color=color.blue)
plot(zeroEMA + bandATR * bandMult, "Context upper band", color=color.new(color.gray, 70))
plot(zeroEMA - bandATR * bandMult, "Context lower band", color=color.new(color.gray, 70))
plotshape(barstate.isconfirmed and (upCross or downCross), title="Unfiltered crosses", style=shape.circle, location=location.abovebar, color=color.gray, size=size.tiny)
plotshape(buy, title="Filtered long setup", style=shape.triangleup, location=location.belowbar, color=color.lime)
plotshape(sell, title="Filtered short setup", style=shape.triangledown, location=location.abovebar, color=color.red)
alertcondition(buy, "Filtered bullish setup", "CC: confirmed trend-pullback-trigger long setup on {{ticker}} {{interval}}.")
alertcondition(sell, "Filtered bearish setup", "CC: confirmed trend-pullback-trigger short setup on {{ticker}} {{interval}}.")
var table status = table.new(position.top_right, 1, 3, bgcolor=color.new(color.black, 15))
if barstate.islast
    table.cell(status, 0, 0, direction < 0 ? "Prior confirmed HTF trend: UP" : "Prior confirmed HTF trend: DOWN / warming up", text_color=color.white)
    table.cell(status, 0, 1, na(touched) ? "Waiting for band touch" : "Pullback known; waiting for cross", text_color=color.white)
    table.cell(status, 0, 2, "Accepted setups: " + str.tostring(accepted), text_color=color.white)

What This Indicator Shows

This is an original proxy: a zero-lag EMA price cross supplies the trigger, while a previous completed higher-timeframe built-in Supertrend supplies direction and a pullback reference. A candle must intersect a configurable tolerance around that reference before a later matching cross can qualify. The default trigger window is twelve chart bars; a confirmed direction change clears it.

The 150-period length, 1.7 displayed chart-band multiplier, and 5/5 Supertrend inputs echo the source’s settings but do not reproduce its named scripts. The chart bands are context visuals; they do not gate the EMA-cross trigger. Gray circles show raw crosses, and colored triangles show accepted sequences.

What This Indicator Does Not Show

The code is not AlgoAlpha’s Zero Lag Trend Signals or the modified Swing Arm calculation. It omits their full arrow, trailing, Fibonacci, and multi-timeframe-table logic. It does not detect every range, place orders, compute 2R brackets, or establish a win rate. Touches can be uncommon when the confirmed trend reference is far from price; visible raw crosses are not missing filtered trades.

Chart Setup, Alerts, and Validation

Paste the complete script into Pine Editor, save it, and add it to a standard-candlestick chart. Follow the timeframe and session instructions above. Configure alerts for the named condition using Once Per Bar Close. The code gates setup markers on completed chart bars; developing reference plots can still move while the current bar forms.

Allow enough history for every lookback and any higher-timeframe calculation. Inspect the visible references and status table before concluding that a quiet chart is broken. A restrictive filter can correctly produce no signals. Changing the chart session, timeframe, inputs, or available data can change the observations; create new alerts after such changes.

Validation status: code, state handling, confirmation timing, and HTML preservation were reviewed statically. The indicator has not been compiled or chart-tested in TradingView here, and alerts and browser copying have not been tested live. No orders, fills, backtest results, or performance claims are produced.

Practice and Research

1. Sequence audit: record raw crosses, matching-direction crosses, and crosses after a qualifying touch as three separate populations.
2. Freshness: compare trigger windows of four, eight, and twelve bars. Count stale permissions and missed continuations without changing the trend calculation.
3. Reference timing: study the confirmed-hour delay in replay. Document when information became available; do not compare it with a historical line that silently uses the unfinished hour’s final state.

Begin on a fifteen-minute chart with a sixty-minute trend reference. Use later sessions to review the same definitions, and record the effect of spread and fill differences separately if you trade the observations.

Check Your Understanding

Does a sell arrow in a bearish trend automatically qualify? No. The pullback must occur first under this framework.

Does a matching indicator name prove matching code? No. Published settings do not reveal every calculation.

Replay assignment: label the direction, touch time, trigger time, and permission expiry before advancing to the next candle.

Final Takeaway

Use direction to restrict the search, a fresh pullback to establish location, and a later trigger to identify timing. An arrow that passes those checks remains a setup observation rather than a promise of profit.

Risk Disclosure

This material is for education and research, not personalized financial advice or a recommendation to trade any security, contract, or account program. Trading can result in substantial losses. Leveraged products can produce losses beyond the amount initially committed, and short selling can involve unlimited loss. Options carry exercise, assignment, liquidity, and expiration risks. Stops and displayed prices do not guarantee execution; gaps, spreads, slippage, fees, funding, halts, and platform interruptions can materially change outcomes. Hypothetical examples, historical charts, indicators, and creator-reported results do not establish future performance. Understand the instrument, account rules, and total exposure, practice before committing capital, and use only risk capital you can afford to lose.

CandleChasers provides independent educational analysis and is not affiliated with, endorsed by, or sponsored by the source creator, platforms, or firms discussed.