This strategy is a comprehensive trading system based on Keltner Channels and dynamic support/resistance levels. It analyzes multiple timeframes, combining moving averages and volatility indicators to form a complete trading decision framework. The core approach is to identify price breakout opportunities while considering market trends and volatility to capture high-probability trading opportunities.
The strategy employs a multi-layer technical indicator system: 1. Uses 21-period Keltner Channels as the primary trend identification tool, with channel width determined by ATR 2. Calculates key support/resistance levels using 21 left-side and 8 right-side bars 3. Incorporates higher timeframe moving averages as trend filters 4. Combines short-term (5-period) and long-term (30-period) moving averages for entry timing 5. Uses ATR for dynamic stop-loss adjustment
This is a well-structured and logically rigorous quantitative trading strategy. Through the coordinated use of multiple technical indicators, it ensures both reliable trading signals and effective risk control. The strategy’s strong extensibility allows for continuous optimization and improvement, potentially maintaining stable performance across different market environments.
/*backtest start: 2024-12-17 00:00:00 end: 2024-12-21 00:00:00 period: 1h basePeriod: 1h exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT","balance":49999}] */ // This Pine Script™ code is subject to the terms of the Mozilla Public License 2.0 at https://mozilla.org/MPL/2.0/ // © sathcm //@version=5 strategy("KMS", overlay=true, initial_capital=100000, default_qty_type=strategy.percent_of_equity, default_qty_value=100, commission_type=strategy.commission.percent, commission_value=0.05, slippage=3) // Inputs for Keltner Channels kcLength = input.int(21, title="Keltner Channel Length", minval=1) // Length for Keltner Channel calculation kcMultiplier = input.float(2.0, title="Keltner Channel Multiplier", minval=0.1) // Multiplier for Keltner Channel width // Calculate Keltner Channels using best practices kcBasis = ta.ema(close, kcLength) // Use EMA for a smoother basis line atrValue = ta.atr(kcLength) // Use ATR for channel width calculation kcUpper = kcBasis + kcMultiplier * atrValue // Upper Keltner Channel kcLower = kcBasis - kcMultiplier * atrValue // Lower Keltner Channel // Inputs for Pivot Point Calculation leftBars = input.int(21, title="Left Bars", minval=1) // Number of bars to the left for pivot calculation rightBars = input.int(8, title="Right Bars", minval=1, tooltip="Number of bars to the right for pivot calculation") // Number of bars to the right for pivot calculation // Calculate Smoothed Pivot Highs and Lows using Weighted Moving Average pivotHigh = ta.pivothigh(high, leftBars, rightBars) // Apply WMA for smoothing pivotLow = ta.pivotlow(low, leftBars, rightBars) // Apply WMA for smoothing // Convert Pivot Highs and Lows to Boolean Conditions isPivotHigh = not na(pivotHigh) // True when a pivot high exists isPivotLow = not na(pivotLow) // True when a pivot low exists // Get Recent Support and Resistance Levels recentResistance = ta.valuewhen(isPivotHigh, high, 0) // Most recent resistance level recentSupport = ta.valuewhen(isPivotLow, low, 0) // Most recent support level // Plot Smoothed Support and Resistance Levels //plot(recentResistance, color=color.red, title="Recent Resistance", linewidth=2, style=plot.style_line) //plot(recentSupport, color=color.green, title="Recent Support", linewidth=2, style=plot.style_line) // Store Entry Price into a Variable var float entryPrice = na // Declare a variable to store the entry price // Input for Higher Timeframe higherTimeframeInput = input.timeframe('W', title="Higher Timeframe for MA Calculation") if (timeframe.period == "240") or (timeframe.period == "120") higherTimeframeInput := "D" if (timeframe.period == "60") or (timeframe.period == "30") or (timeframe.period == "15") higherTimeframeInput := "120" if (timeframe.period == "10") or (timeframe.period == "5") higherTimeframeInput := "30" if (timeframe.period == "1") higherTimeframeInput := "10" prd = input.int(defval=10, title='Pivot Period', minval=4, maxval=30, group='Settings 🔨', tooltip='Used while calculating Pivot Points, checks left&right bars') ppsrc = input.string(defval='High/Low', title='Source', options=['High/Low', 'Close/Open'], group='Settings 🔨', tooltip='Source for Pivot Points') ChannelW = input.int(defval=5, title='Maximum Channel Width %', minval=1, maxval=8, group='Settings 🔨', tooltip='Calculated using Highest/Lowest levels in 300 bars') minstrength = input.int(defval=1, title='Minimum Strength', minval=1, group='Settings 🔨', tooltip='Channel must contain at least 2 Pivot Points') maxnumsr = input.int(defval=4, title='Maximum Number of S/R', minval=1, maxval=10, group='Settings 🔨', tooltip='Maximum number of Support/Resistance Channels to Show') - 1 loopback = input.int(defval=150, title='Loopback Period', minval=100, maxval=400, group='Settings 🔨', tooltip='While calculating S/R levels it checks Pivots in Loopback Period') res_col = input.color(defval=color.new(color.red, 75), title='Resistance Color', group='Colors 🟡🟢🟣') sup_col = input.color(defval=color.new(color.lime, 75), title='Support Color', group='Colors 🟡🟢🟣') inch_col = input.color(defval=color.new(color.gray, 75), title='Color When Price in Channel', group='Colors 🟡🟢🟣') // Get Pivot High/Low src1 = ppsrc == 'High/Low' ? high : math.max(close, open) src2 = ppsrc == 'High/Low' ? low : math.min(close, open) ph = ta.pivothigh(src1, prd, prd) pl = ta.pivotlow(src2, prd, prd) // Calculate maximum S/R channel width prdhighest = ta.highest(300) prdlowest = ta.lowest(300) cwidth = (prdhighest - prdlowest) * ChannelW / 100 // Get/keep Pivot levels var pivotvals = array.new_float(0) var pivotlocs = array.new_float(0) if ph or pl array.unshift(pivotvals, ph ? ph : pl) array.unshift(pivotlocs, bar_index) for x = array.size(pivotvals) - 1 to 0 by 1 if bar_index - array.get(pivotlocs, x) > loopback // remove old pivot points array.pop(pivotvals) array.pop(pivotlocs) continue break // Find/create SR channel of a pivot point get_sr_vals(ind) => float lo = array.get(pivotvals, ind) float hi = lo int numpp = 0 for y = 0 to array.size(pivotvals) - 1 by 1 float cpp = array.get(pivotvals, y) float wdth = cpp <= hi ? hi - cpp : cpp - lo if wdth <= cwidth // fits the max channel width? if cpp <= hi lo := math.min(lo, cpp) else hi := math.max(hi, cpp) numpp += 20 // each pivot point added as 20 [hi, lo, numpp] // Keep old SR channels and calculate/sort new channels if we met new pivot point var suportresistance = array.new_float(20, 0) // min/max levels changeit(x, y) => tmp = array.get(suportresistance, y * 2) array.set(suportresistance, y * 2, array.get(suportresistance, x * 2)) array.set(suportresistance, x * 2, tmp) tmp := array.get(suportresistance, y * 2 + 1) array.set(suportresistance, y * 2 + 1, array.get(suportresistance, x * 2 + 1)) array.set(suportresistance, x * 2 + 1, tmp) if ph or pl supres = array.new_float(0) // number of pivot, strength, min/max levels stren = array.new_float(10, 0) // Get levels and strengths for x = 0 to array.size(pivotvals) - 1 by 1 [hi, lo, strength] = get_sr_vals(x) array.push(supres, strength) array.push(supres, hi) array.push(supres, lo) // Add each HL to strength for x = 0 to array.size(pivotvals) - 1 by 1 h = array.get(supres, x * 3 + 1) l = array.get(supres, x * 3 + 2) s = 0 for y = 0 to loopback by 1 if high[y] <= h and high[y] >= l or low[y] <= h and low[y] >= l s += 1 array.set(supres, x * 3, array.get(supres, x * 3) + s) // Reset SR levels array.fill(suportresistance, 0) // Get strongest SRs src = 0 for x = 0 to array.size(pivotvals) - 1 by 1 stv = -1. // value stl = -1 // location for y = 0 to array.size(pivotvals) - 1 by 1 if array.get(supres, y * 3) > stv and array.get(supres, y * 3) >= minstrength * 20 stv := array.get(supres, y * 3) stl := y if stl >= 0 // Get SR level hh = array.get(supres, stl * 3 + 1) ll = array.get(supres, stl * 3 + 2) array.set(suportresistance, src * 2, hh) array.set(suportresistance, src * 2 + 1, ll) array.set(stren, src, array.get(supres, stl * 3)) // Make included pivot points' strength zero for y = 0 to array.size(pivotvals) - 1 by 1 if array.get(supres, y * 3 + 1) <= hh and array.get(supres, y * 3 + 1) >= ll or array.get(supres, y * 3 + 2) <= hh and array.get(supres, y * 3 + 2) >= ll array.set(supres, y * 3, -1) src += 1 if src >= 10 break for x = 0 to 8 by 1 for y = x + 1 to 9 by 1 if array.get(stren, y) > array.get(stren, x) tmp = array.get(stren, y) array.set(stren, y, array.get(stren, x)) changeit(x, y) get_level(ind) => float ret = na if ind < array.size(suportresistance) if array.get(suportresistance, ind) != 0 ret := array.get(suportresistance, ind) ret get_color(ind) => color ret = na if ind < array.size(suportresistance) if array.get(suportresistance, ind) != 0 ret := array.get(suportresistance, ind) > close and array.get(suportresistance, ind + 1) > close ? res_col : array.get(suportresistance, ind) < close and array.get(suportresistance, ind + 1) < close ? sup_col : inch_col ret // var srchannels = array.new_box(10) // for x = 0 to math.min(9, maxnumsr) by 1 // box.delete(array.get(srchannels, x)) // srcol = get_color(x * 2) // if not na(srcol) // array.set(srchannels, x, box.new(left=bar_index, top=get_level(x * 2), right=bar_index + 1, bottom=get_level(x * 2 + 1), border_color=srcol, border_width=1, extend=extend.both, bgcolor=srcol)) // Improved dynamic support detection float recentSupport1 = na float previousSupport = na float currentsupport = na if na(previousSupport) or currentsupport != previousSupport if array.size(suportresistance) > 1 for i = 0 to math.floor(array.size(suportresistance) / 2) - 1 // Iterate through support levels currentsupport := array.get(suportresistance, i * 2 + 1) // Support is stored at odd indices if currentsupport < close and (na(recentSupport1) or math.abs(close - currentsupport) < math.abs(close - recentSupport1)) previousSupport := currentsupport // Store the newly detected support // Set the most recent support to the new support recentSupport1 := na(recentSupport1) ? ta.lowest(low, 10) : currentsupport // Moving averages for entry and exit maShort = ta.sma(close, 5) maLong = ta.sma(close, 30) + ta.atr(14) // Track entry price entryPrice1 = strategy.position_avg_price // Get the price of the currently open position currentTimeFrame = timeframe.period exitPrice = entryPrice1 * 0.99 if currentTimeFrame == "1H" or currentTimeFrame == "30" or currentTimeFrame == "15" or currentTimeFrame == "5" exitPrice := entryPrice1 * 0.99 // Set the exit price at 99% of the entry price if currentTimeFrame == "120" or currentTimeFrame == "180" or currentTimeFrame == "240" or currentTimeFrame == "D" exitPrice := entryPrice1 * 0.98 // Set the exit price at 95% of the entry price // Calculate Moving Average based on higher timeframe for length of 20 bars higherTimeframeMA = request.security(syminfo.tickerid, higherTimeframeInput, ta.sma(close, 20), barmerge.gaps_off, barmerge.lookahead_on) // Calculate MA with adjusted timeframe // Entry and Exit Conditions for Long entryLong = (close > kcUpper) and (close > recentResistance) and (close > higherTimeframeMA) // Long entry when price breaks above KC upper, recent resistance, and higher timeframe MA exitLong = (close < recentResistance - 1.5*atrValue) // Long exit when price falls below recent resistance with cushion of one ATR // Entry and Exit Conditions for Short entryShort = (close < kcLower) and (close < recentSupport) and (close < higherTimeframeMA+atrValue) // Add RSI filter to reduce false signals by confirming momentum // Short entry when price breaks below KC lower, recent support, and higher timeframe MA exitShort = (close > recentSupport + atrValue) // Short exit when price rises above recent support with cushion of one ATR(close > recentSupport + atrValue) // Short exit when price rises above recent support with cushion of one ATR(close > recentSupport + atrValue) // Short exit when price rises above recent support with cushion of one ATR // Strategy Execution for Long if not na(recentSupport1) and (close <= recentSupport1 +(close*0.01) or close >= recentSupport1 - (close*0.0075)) and (maShort > maLong) and entryLong strategy.entry("Long Entry", strategy.long) //entryPrice := strategy.position_avg_price // Store the entry price when a position is opened if ((maShort < maLong + 3*ta.atr(14)) or close < exitPrice) and exitLong strategy.close("Long Entry") // Strategy Execution for Short if entryShort strategy.entry("Short Entry", strategy.short) entryPrice := strategy.position_avg_price // Store the entry price when a position is opened if exitShort strategy.close("Short Entry") // Plot Keltner Channels plot(kcUpper, color=color.orange, title="Keltner Channel Upper", linewidth=1) plot(kcLower, color=color.orange, title="Keltner Channel Lower", linewidth=1) // Plot Moving Averages plot(higherTimeframeMA, color=color.blue, title="Higher Timeframe MA", linewidth=2) //plot(recentSupport1, color=#04313f, title="Recent Support1") //plot(recentResistance, color=color.purple, title="Recent Resistance") //plot(entryPrice1, color=color.lime, title="Entry Price 1") //plot(exitPrice, color=color.maroon, title="Exit Price") //plot(maShort, color=color.green, title="MA Short") //plot(maLong, color=color.blue, title="MA Long Plus ATR") // Highlight Entry Zones bgcolor(entryLong ? color.new(color.green, 85) : na, title="Long Entry Zone") bgcolor(entryShort ? color.new(color.red, 85) : na, title="Short Entry Zone") // Alerts alertcondition(entryLong, title="Long Entry", message="Price broke above the Keltner Channel and recent resistance for Long Entry") alertcondition(exitLong, title="Long Exit", message="Price fell below recent resistance with cushion of one ATR - Long Exit") alertcondition(entryShort, title="Short Entry", message="Price broke below the Keltner Channel and recent support for Short Entry") alertcondition(exitShort, title="Short Exit", message="Price rose above recent support with cushion of one ATR - Short Exit")