Three Candle Reversal Trend Strategy

Author: ChaoZhang, Date: 2024-02-18 09:48:28



The Three Candle Reversal Trend Strategy is a short-term trading strategy that identifies reversals in short-term trends by detecting three consecutive bullish or bearish candles followed by an engulfing candle in the opposite direction, combined with multiple technical indicators to filter entry signals. The strategy trades with a 1:3 risk-reward ratio for gaining excess returns.

Strategy Logic

The core logic of this strategy is to identify the pattern of three consecutive bullish or bearish candles on the chart, which usually implies an impending reversal in the short-term trend. When three bearish candles are detected, wait for the next engulfing bullish candle to go long. Conversely, when three bullish candles are detected, wait for the next engulfing bearish candle to go short. This allows capturing reversing opportunities in short-term trends in a timely manner.

In addition, multiple technical indicators are introduced to filter entry signals. Two SMA lines with different parameter settings are adopted, and entry positions are considered only when the faster SMA crosses over the slower line. Besides, the linear regression indicator is used to judge whether the market is ranging or trending, and trades are taken only in trending conditions. There is also an option to combine the candlestick pattern with SMA golden crosses for additional entry signals. Through the comprehensive judgments of these indicators, most noise can be filtered out and the entry accuracy is improved.

For stop loss and take profit, the strategy requires a minimum 1:3 risk-reward ratio. The ATR indicator based on the price fluctuation of recent N candles is used to determine the stop loss level with an offset percentage. Take profit is then calculated accordingly to target proper excess returns for the risk taken.


The Three Candle Reversal Trend Strategy has the following advantages:

  1. Identify reversals of short-term trends for timely opportunities
  2. Enhanced entry accuracy via multiple indicator filters
  3. Reasonable risk-reward profile with appropriate stop loss and take profit
  4. Simple parameters for ease of understanding and operation


There are also some risks to note for this strategy:

  1. Short-term reversals do not necessarily imply long-term trend reversals. Higher timeframe trends should be monitored. Longer period moving averages can be added as filters.
  2. Single candlestick patterns may produce false signals. Other supplementary judgments can be considered.
  3. Stop loss settings could be too aggressive. Stop loss range can be tightened.
  4. Insufficient backtest data leads to uncertainty in real trading performance.

Enhancement Directions

The strategy can be enhanced in the following aspects:

  1. Adjust parameters for moving averages and linear regression to better identify trends.
  2. Add other indicators like Stoch for supplementary signal confirmation.
  3. Optimize ATR parameters and stop loss percentage to balance risk and return.
  4. Introduce trend breakout tracking mechanisms to improve profitability.
  5. Establish robust capital management schemes to control trading risks.


In conclusion, the Three Candle Reversal Trend Strategy is a simple short-term trading strategy that capitalizes on price patterns and multiple indicators to capture reversing opportunities, built on properly balanced risk-reward profiles. It delivers respectable results with relatively low complexity, and is worth investor attention and testing. There is also ample room for improvement via parameter tuning and rule supplementation for the strategy to grow into a stable high-efficiency algo trading system.

start: 2024-01-01 00:00:00
end: 2024-01-31 23:59:59
period: 3h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]

// This source code is subject to the terms of the Mozilla Public License 2.0 at
// © platsn
// Mainly developed for SPY trading on 1 min chart. But feel free to try on other tickers.

// Basic idea of this strategy is to look for 3 candle reversal pattern within trending market structure. The 3 candle reversal pattern consist of 3 consecutive bullish or bearish candles, 
// followed by an engulfing candle in the opposite direction. This pattern usually signals a reversal of short term trend. This strategy also uses multiple moving averages to filter long or short
// entries. ie. if the 21 smoothed moving average is above the 50, only look for long (bullish) entries, and vise versa. There is option change these moving average periods to suit your needs. 
// I also choose to use Linear Regression to determine whether the market is ranging or trending. It seems the 3 candle pattern is more successful under trending market. Hence I use it as a filter.

// There is also an option to combine this strategy with moving average crossovers. The idea is to look for 3 canddle pattern right after a fast moving average crosses over a slow moving average.
// By default , 21 and 50 smoothed moving averages are used. This gives additional entry opportunites and also provides better results. 

// This strategy aims for 1:3 risk to reward ratio. Stop losses are calculated using the closest low or high values for long or short entries, respectively, with an offset using a percentage of
// the daily ATR value. This allows some price flucuation without being stopped out prematurely. Price target is calculated by multiplying the difference between the entry price and the stop loss
// by a factor of 3. When price target is reach, this strategy will set stop loss at the price target and wait for exit considion to maximize potential profit. 

// This strategy will exit an order if an opposing 3 candle pattern is detected, this could happend before stop loss or price target is reached, and may also happen after price target is reached.

// *Note that this strategy is designed for same day SPY option scalping. I haven't determined an easy way to calculate the # of contracts to represent the equivalent option values. Plus the option
// prices varies greatly depending on which strike and expiry that may suits your trading style. Therefore, please be mindful of the net profit shown. By default, each entry is approxiately equal 
// to buying 10 of same day or 1 day expiry call or puts at strike $1 - $2 OTM. This strategy will close all open trades at 3:45pm EST on Mon, Wed, and Fri. 

// **Note that this strategy also takes into account of extended market data.

// ***Note pyramiding is set to 2 by default, so it allows for multiple entries on the way towards price target. 

// Remember that market conditions are always changing. This strategy was only able to be backtested using 1 month of data. This strategy may not work the next month. Please keep that in mind. 

// *****************************************************************************************************************************************************************************************************

strategy("3 Candle Strike Stretegy", overlay=true, pyramiding=2, initial_capital=5000, commission_type=strategy.commission.cash_per_contract, commission_value = 0.01) 

// ******************** Period **************************************
startY = input(title='Start Year', defval=2011, group = "Trading window")
startM ='Start Month', defval=1, minval=1, maxval=12, group = "Trading window")
startD ='Start Day', defval=1, minval=1, maxval=31, group = "Trading window")
finishY = input(title='Finish Year', defval=2050, group = "Trading window")
finishM ='Finish Month', defval=12, minval=1, maxval=12, group = "Trading window")
finishD ='Finish Day', defval=31, minval=1, maxval=31, group = "Trading window")
timestart = timestamp(startY, startM, startD, 00, 00)
timefinish = timestamp(finishY, finishM, finishD, 23, 59)
t1 = time(timeframe.period, "0930-1545:23456")
window = true

// *****************************************************

isSPY = input.bool(defval=true,title="SPY trading only", group = "Trading Options")
SPY_option =,title="# of SPY options per trade", group = "Trading Options")
reinvest = input.bool(defval=false,title="reinvest profit?", group = "Trading Options")
src = close

// ***************************************************************************************************** Daily ATR *****************************************************
// Inputs
atrlen =, minval=1, title="ATR period", group = "Daily ATR")
iPercent = input.float(5, minval=1, maxval=100, step=0.1, title="% ATR to use for SL / PT", group = "Daily ATR")
// PTPercent =, minval=1, title="% ATR for PT")

// Logic
percentage = iPercent * 0.01
datr =, "1D", ta.rma(, atrlen))
datrp = datr * percentage
// datrPT = datr * PTPercent * 0.01

plot(datr,"Daily ATR")
plot(datrp, "Daily % ATR")

// ***************************************************************************************************************** Moving Averages ************************

len0 =, minval=1, title='Fast EMA', group= "Moving Averages")
ema1 = ta.ema(src, len0)

len1 =, minval=1, title='Fast SMMA', group= "Moving Averages")
smma1 = 0.0
sma_1 = ta.sma(src, len1)
smma1 := na(smma1[1]) ? sma_1 : (smma1[1] * (len1 - 1) + src) / len1

len2 =, minval=1, title='Slow SMMA', group= "Moving Averages")
smma2 = 0.0
sma_2 = ta.sma(src, len2)
smma2 := na(smma2[1]) ? sma_2 : (smma2[1] * (len2 - 1) + src) / len2

len3 =, minval=1, title='Slow SMMA', group= "Moving Averages")
smma3 = 0.0
sma_3 = ta.sma(src, len3)
smma3 := na(smma3[1]) ? sma_3 : (smma3[1] * (len3 - 1) + src) / len3

ma_bull = smma1 > smma2 and smma1 > smma1[1]
ma_bear = smma1 < smma2 and smma1 < smma1[1]

ma_bull_macro = smma1 > smma3 and smma2 > smma3
ma_bear_macro = smma1 < smma3 and smma2 < smma3

// plot(ma_bull? 1 : 0, "MA bull")
// plot(ma_bear? 1 : 0 , "MA bear")

// **************************************************************************************************************** Linear Regression *************************

clen = = 50, minval = 1, title = "Linear Regression Period", group = "Linear Regression")
slen =, minval=1, title="LR Slope Period" , group = "Linear Regression")
glen =, minval=1, title="LR Signal Period", group = "Linear Regression")
LR_thres = input.float(0.03, minval=0, step=0.001, title="LR Threshold for Ranging vs Trending" , group = "Linear Regression")
//Linear Regression Curve
lrc = ta.linreg(src, clen, 0)
//Linear Regression Slope
lrs = (lrc-lrc[1])/1
//Smooth Linear Regression Slope
slrs = ta.ema(lrs, slen)
//Signal Linear Regression Slope
alrs = ta.sma(slrs, glen)

up_accel = lrs > alrs and lrs > 0 
down_accel = lrs < alrs and lrs < 0 

LR_ranging  = math.abs(slrs) <= LR_thres
LR_trending = math.abs(slrs) > LR_thres

plot(slrs, "LR slope")
plot(LR_trending?1:0, "LR Trending")

// *********************************************************************************************************************************** Candle conditions **************************

bull_3s = close[3] <= open[3] and close[2] <= open[2] and close[1] <= open[1] and close > open[1]
bear_3s = close[3] >= open[3] and close[2] >= open[2] and close[1] >= open[1] and close < open[1]

plotshape(bull_3s, style=shape.triangleup,, 0), location=location.belowbar, size=size.small, text='3s-Bull', title='3 Line Strike Up')
plotshape(bear_3s, style=shape.triangledown,, 0), location=location.abovebar, size=size.small, text='3s-Bear', title='3 Line Strike Down')

// ***************************************************************************************************************************************** SL & PT ***********************************
RR = input.float(3.0, minval = 1, step = 0.1, title="Reward to Risk Ratio", group = "Trading Options")

    strategy.opentrades > 0 ? (bar_index - strategy.opentrades.entry_bar_index(strategy.opentrades-1)) : na

last_high = math.max(high, high[1], high[2], high[3])
last_low = math.min(low, low[1], low[2], low[3])

long_SL = last_low - datrp
short_SL = last_high + datrp

long_PT = last_high
short_PT = last_low

last_entry = strategy.opentrades.entry_price(strategy.opentrades-1)
risk = last_entry - long_SL

if strategy.opentrades > 0
    long_SL := math.min(long_SL[barsSinceLastEntry()], last_low)
    short_SL := math.max(short_SL[barsSinceLastEntry()], last_high)
    risk := last_entry - long_SL
    long_PT := last_entry + (last_entry - long_SL) * RR
    short_PT := last_entry - (short_SL - last_entry) * RR
    long_PT := open + (open - long_SL) * RR
    short_PT := open - (short_SL - open) * RR

// plot(short_SL,title = "Short SL",,30))
// plot(long_SL,title = "Long SL",,30))
// plot(long_PT,title = "Long PT",,50))
// plot(short_PT,title = "Short PT",,50))
// plot(last_entry, title = "Last entry")
// plot(risk, title = "Risk")

// **************************************************************************************************************************************** Trade Pauses ****************************************
bool trade_pause = false
bool trade_pause2 = false

if high - low > datr*0.3
    trade_pause := true
    trade_pause := false

no_longat10 = input.bool(true, title="No long entry between 10 - 10:30 (Avoid 10 am dump)", group = "Trading Options")

// ************************************************************************************************************************************ Entry conditions **************************

trade_3s = input.bool(title='Trade 3s candle pattern', defval=true, group = "Trading Options")
L_entry1 = bull_3s and ma_bull and LR_trending 
S_entry1 = bear_3s and ma_bear and LR_trending

trade_ma_reversal = input.bool(title='Trade MA Cross Reversal Signal', defval=true, group = "Trading Options")
L_entry2 = ma_bear_macro and ema1 > smma1 and bull_3s and ta.barssince(ta.cross(ema1,smma1)) < 10
S_entry2 = ma_bull_macro and ema1 < smma1 and bear_3s and ta.barssince(ta.cross(ema1,smma1)) < 10

// ************************************************************************************************************************************** Exit Conditions ********************************

// bsle_thres =, "Bar since entry threshold")

// exit0 = barsSinceLastEntry() >= bsle_thres
exit0 = true

L_exit1 = bear_3s
S_exit1 = bull_3s

// ************************************************************************************************************************************ Entry and Exit orders *****************************
strategy.initial_capital = 50000
trade_amount = math.floor(strategy.initial_capital / close)

if isSPY 
    if strategy.netprofit > 0 and reinvest
        trade_amount := math.floor((strategy.initial_capital + strategy.netprofit) * 0.2 / 600) * 10 * SPY_option
        trade_amount := math.floor(strategy.initial_capital * 0.2 / 600) * 10 * SPY_option

if not(trade_pause) and not(trade_pause2) and time(timeframe.period, "0930-1540:23456")
    if trade_3s
        if not(time(timeframe.period, "1000-1030:23456")) and no_longat10
            strategy.entry("Long", strategy.long, 1, when = L_entry1 and window, comment="Long 3s" + " SL=" + str.tostring(math.round(long_SL,2)) + " PT=" + str.tostring(math.round(long_PT,2)))
        strategy.entry("Short", strategy.short, 1, when = S_entry1 and window, comment = "Short 3s" + " SL=" + str.tostring(math.round(short_SL,2)) + " PT=" + str.tostring(math.round(short_PT,2)))
    if trade_ma_reversal
        strategy.entry("Long", strategy.long, 1, when = L_entry2 and window, comment="Long MA cross" + " SL=" + str.tostring(math.round(long_SL,2)) + " PT=" + str.tostring(math.round(long_PT,2)))
        strategy.entry("Short", strategy.short, 1, when = S_entry2 and window, comment = "Short MA corss" + " SL=" + str.tostring(math.round(short_SL,2)) + " PT=" + str.tostring(math.round(short_PT,2)))

if high > long_PT
    long_SL := low[1]
    strategy.exit("Exit", "Long", when = exit0 and low < long_PT, stop= long_SL, comment = "Exit Long SL/PT hit")
strategy.close("Long", when = L_exit1, comment = "Exit on Bear Signal")

if low < short_PT
    short_SL := high[1]
    strategy.exit("Exit", "Short", when= exit0 and high > short_PT, stop= short_SL, comment = "Exit Short SL/PT hit")
strategy.close("Short", when = S_exit1, comment = "Exit on Bull Signal")

if time(timeframe.period, "1545-1600:246")