Orion Trading Strategy

Pine ScriptMean reversion Momentum Trend following RSI MACD EMA SMA Stochastic Pivots / S&R

Author: ChaoZhang
Complexity: Advanced · 307 lines
Parameters: 17
Backtest bar: 3m · intraday
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

Overview

The Orion trading strategy integrates multiple technical indicators for quantitative trading. It aims to identify market tops and bottoms early so traders can make timely buy and sell decisions. The strategy uses a unique prediction curve mechanism to try generating trading signals before actual price reversals occur.

Strategy Logic

The core of the strategy is the proprietary Orion signal curve. This curve synthesizes multiple indicators including MACD, WPR, Stoch, RSI etc to generate a composite signal. It is then processed by supersmoothing to create the final curve.

Critically, the curve also incorporates a prediction model, which analyzes the slope changes of the curve to try forecasting potential reversals 1-2 bars ahead. When the prediction curve diverges from the actual curve, early trading signals can be generated.

In addition, a momentum wave indicator is used to determine the trend direction on a larger timeframe. When the wave changes direction, it suggests a larger degree reversal may be upcoming.

Finally, the strategy provides buy and sell suggestions when signals are triggered. Traders can decide whether to follow them.

Advantage Analysis

Multiple indicators improve accuracy Combining indicators helps confirm trends and spot reversals, avoiding single indicator pitfalls.
Prediction model provides early reversal alerts The prediction curve may front-run actual signals, granting trading decisions a head start.
Momentum wave judges overall trend direction Incorporating higher timeframe momentum wave avoids trading against major trends.
Customizable parameters suit different products Users can tune indicator parameters to suit characteristics of different trading products.

Risk Analysis

Prediction model may cause over-trading The prediction model can generate false signals. Blindly following it may lead to over-trading.
Difficult optimization with multiple parameters With numerous parameters, finding the optimum combination requires extensive datasets and prolonged testing.
Indicator effectiveness needs prudent assessment The actual incremental benefit of each indicator needs careful evaluation to avoid redundancy.
Real-world trading costs should be considered Frequent trading incurs higher costs. Real-world costs need incorporating into backtests.

Improvement Directions

Evaluate and adjust prediction model
Assess prediction accuracy and optimize parameters to improve reliability.
Simplify model by reducing redundancy Adopt indicator effectiveness evaluation and model simplification to remove unnecessary complexity.
Robustness test across markets Conduct multi-market backtests to verify optimization results and robustness.
Adjust strategy based on real-world costs Introduce real-world costs into backtest to adjust strategy parameters for lower trade frequency.

Summary

The Orion strategy synthesizes multiple indicators and a unique prediction curve to try identifying turns early. It has merits but scalability is also limited. Cautious attitude is needed. Continuous optimizations from aspects like signal efficacy and cost effectiveness are required to achieve steady long-term gains in automated trading.

Parameters

Name	Default	What it controls
v_input_14	0	strategy: dca long \| simple
v_input_15	10	pyramid orders
v_input_16	2	percent exit in profit
v_input_17	2	percent drop before each entry
v_input_1	true	(?User Agreement ─────────────)I understand that Orion Algo cannot be 100% accurate and overall performance will shift with market conditions. While Orion Algo increases my chances of entering better positions, I must use smart trade management.
v_input_2	10	(?Signal ────────────────────)Super Smooth
v_input_3	false	Prediction
v_input_4	0.45	Bias
v_input_5	true	Curve
v_input_6	true	(?Momentum Wave ─────────────)Momentum Wave
v_input_7	3	momentumWaveLength
v_input_8	true	Position Outside
v_input_9	true	(?Visuals ───────────────────)Dark Mode
v_input_10	0	Mode: Pro \| Beginner
v_input_11	true	(?Dashboard ─────────────────)Dashboard
v_input_12	true	(?Pivots ────────────────────)Signal Pivots
v_input_13	false	Prediction Pivots

Source Pine Script

/*backtest
start: 2023-09-17 00:00:00
end: 2023-09-21 22:00:00
period: 3m
basePeriod: 1m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

// © OrionAlgo
//          () /? | () |\|   /\ |_ (_, ()          //
//@version=4

version = '2.0'

strategy("Orion Algo Strategy v"+version, shorttitle="Orion Algo Strategy v"+version, overlay=false, pyramiding=100)


// Getting inputs --------------------------------------------------------------
userAgreement = input(true, title='I understand that Orion Algo cannot be 100% accurate and overall performance will shift with market conditions. While Orion Algo increases my chances of entering better positions, I must use smart trade management. ', type=input.bool,group='User Agreement ─────────────',
  tooltip='In order to use Orion Algo, you must click the checkbox to acknowledge the user agreement')

src = close
//smoothing inputs -------------------------------------------------------------

//superSmooth = input(true, title='Super Smooth', inline='Super Smooth', group='Smoothing ─────────────────')
superSmooth = true
smoothType = 1
superSmoothStrength = input(10, title='Super Smooth',minval = 3, inline='Super Smooth', group='Signal ────────────────────',
  tooltip='Smooths the signal. Lower values move pivots to the left while increasing noise, higher values move pivots to the right and reduce noise. 8 is a good mix of both') // set to timeframe for decent results?
//trendSmoothing = input(30, title='Trend Smooth',minval = 3, group='Smoothing ─────────────────') // set to timeframe for decent results?
trendSmoothing = 30 // set to timeframe for decent results?

showPrediction = input(false, title='Prediction', group='Signal ────────────────────',inline='prediction')
predictionBias = input(0.45, minval = 0.,maxval=1., step=0.05, title='Bias', group='Signal ────────────────────',inline='prediction')
showPredictionCurve = input(true, title='Curve', group='Signal ────────────────────',inline='prediction', tooltip='Prediction model that attempts to predict short range reversals (0-2 bars). Adjust Bias to change the prediction curve.')

//momentum wave inputs ---------------------------------------------------------
showMomentumWave = input(true, 'Momentum Wave', group='Momentum Wave ─────────────', inline='mom')
momentumWaveLength = input(3, '', group='Momentum Wave ─────────────', inline='mom', tooltip='Secondary signal that shows medium to large movements based on the input variable. The wave will change depending on the current timeframe.')
momentumOutside = input(true, 'Position Outside', group='Momentum Wave ─────────────', inline='mom2', tooltip='Positions the wave outside of the main signal area.')

//visuals input-----------------------------------------------------------------

useDarkMode = input(true, 'Dark Mode', group='Visuals ───────────────────',inline='Colors')

// 0:backgroundlines, 1:signal, 2:bullish, 3:bearish, 4:hiddenbull, 5:hiddenbear, 6:deltav, 7:prediction, 8:predictionbull, 9:predictionbear, 10:dash, 11:mom2

visualMode = input('Pro', 'Mode',options=['Beginner', 'Pro'] ,group='Visuals ───────────────────')

dashOn = input(true, "Dashboard", group='Dashboard ─────────────────', inline='dash', tooltip='A dashboard with some usefual stats')
  
dashColor = color.new(#171a27, 100)

showPivots = input(true, title='Signal Pivots', group='Pivots ────────────────────',inline='pivots')
showPredictionPivots = input(false, title='Prediction Pivots', group='Pivots ────────────────────',inline='pivots')


// Functions -------------------------------------------------------------------

f_secureSecurity(_symbol, _res, _src) => security(_symbol, _res, _src,barmerge.gaps_on, lookahead = barmerge.lookahead_on) 

f_slope(x) =>
    slopePeriod = 1
    (x - x[slopePeriod]) / slopePeriod

f_superSmooth(inputVal,smoothType) =>
    smoothType==1? (hma(inputVal,superSmoothStrength)) :
      smoothType==2? (ema((ema((ema(inputVal,3)),3)),superSmoothStrength)):
      smoothType==3? linreg(inputVal,superSmoothStrength,0) : 
      smoothType==4? (hma(inputVal,superSmoothStrength * momentumWaveLength)) : na

f_bias(bias, min, max) =>
    (bias * (max - min) ) + min

f_resInMinutes() =>
    _resInMinutes = timeframe.multiplier * (
      timeframe.isseconds ? 1. / 60. :
      timeframe.isminutes ? 1.       :
      timeframe.isdaily   ? 1440.    :
      timeframe.isweekly  ? 10080.   :
      timeframe.ismonthly ? 43800.   : na)

f_resFromMinutes(_minutes) =>
    _minutes     <= 0.0167       ? "1S"  :
      _minutes   <= 0.0834       ? "5S"  :
      _minutes   <= 0.2500       ? "15S" :
      _minutes   <= 0.5000       ? "30S" :
      _minutes   <= 1            ? "1":
      _minutes   <= 1440         ? tostring(round(_minutes)) :
      _minutes   <= 43800        ? tostring(round(min(_minutes / 1440, 365))) + "D" :
      tostring(round(min(_minutes / 43800, 12))) + "M"
      
f_output_signal()=>    
    a = ((ema(close, 12) - ema(close, 26)) - ema((ema(close, 12) - ema(close, 26)), 8))/10
    b = wpr(8)
    c = (100 * ( close + 2*stdev( close, 21) - sma( close, 21 ) ) / ( 4 * stdev( close, 21 ) ))
    d = (rsi(close - sma(close, 21)[11],8)*2)-100
    e = (rsi(fixnan(100 * rma(change(high) > change(low) and change(high) > 0 ? change(high) : 0, 1) / rma(tr, 1)) - fixnan(100 * rma(change(low) > change(high) and change(low) > 0 ? change(low) : 0, 1) / rma(tr, 1)),8)*2)-100 //causes slow down
    f = rsi((((close-( (sum(volume, 20) - volume)/sum(volume, 20)) + (volume*close/sum(volume, 20)))/((close+( (sum(volume, 20) - volume)/sum(volume, 20)) + (volume*close/sum(volume, 20)))/2)) * 100),8)-100
    g = (rsi(sma(highest(high,14)-lowest(low,14)==0.0?0.0:(close-lowest(low,14))/highest(high,14)-lowest(low,14)-0.5,max(1,int(2))),8)*2)-100 //causes slow down
    avg(a,b,c,d,e,f,g)*2
 
output_signal = f_output_signal()
output_signal := f_superSmooth(output_signal,1)

// output_signal2 = plot(f_superSmoothSlow(f_output_signal()), color=color.blue, linewidth=2)

//Orion Signal Higher Timeframe / Momentum Wave --------------------------------
f_momentumWave(wavelength,smooth) =>
    currentMinutes = f_resInMinutes()
    m = currentMinutes * wavelength //multiply current resolution by momentumWaveLength to get higher resolution
    momentumWaveRes = f_resFromMinutes(m)
    f_secureSecurity(syminfo.tickerid, momentumWaveRes,f_superSmooth(f_output_signal(),1))



// Plot ------------------------------------------------------------------------
f_color(x) =>
    if userAgreement
        white      = useDarkMode ? #e5e4f4 : #505050ff
        lightgray  = useDarkMode ? #808080 : #909090ff
        gray       = useDarkMode ? #808080 : #505050ff
        //blue       = useDarkMode ? #007EA7 : #007EA7ff
        blue       = useDarkMode ? #2862FFFF : #2862FFFF
        
        // 0:backgroundlines, 1:signal, 2:bullish, 3:bearish, 4:hiddenbull, 5:hiddenbear, 6:deltav, 7:prediction, 8:predictionbull, 9:predictionbear, 10:trendbull, 11:trendbear, 12:dash, 13:mom1, 14:mom2
        x==0? lightgray : x==1? gray : x==2? white : x==3? blue : x==4? white : x==5? blue : x==6? blue : x==7? blue : x==8? white : x==9? blue : x==10? blue : x==11? blue : na

// Lines -----------------------------------------------------------------------

h1 = plot(0, "Mid Band", color=f_color(0),editable=0, transp=80)

// Signal ----------------------------------------------------------------------

orionSignal = plot(output_signal, title="Orion Signal Curve", style=plot.style_line,linewidth=1, transp=0, color= f_color(1), offset=0,editable=0)


// Momentum Wave ---------------------------------------------------------------
momWave = f_momentumWave(momentumWaveLength,1)


p_momWave = plot(showMomentumWave? momentumOutside? (momWave/2) -150 : momWave : na, color=f_color(11), linewidth=showMomentumWave and momentumOutside ? 1 : 2, editable =0, transp=50, style=momentumOutside? plot.style_area : plot.style_line, histbase=-200) //two tone color doesnt want to work with this for some reason.

// Divergence ------------------------------------------------------------------

osc = output_signal

plFound = osc > osc [1] and osc[1] < osc[2]
phFound = osc < osc [1] and osc[1] > osc[2]

// bullish

plot(
     plFound and visualMode=='Pro'?  osc[1] - 10 : na,
     offset=0,
     title="Regular Bullish",
     linewidth=3,
     color=showPivots ? f_color(2) :na,
     transp=0,
     style=plot.style_circles,
     editable=0
     )
plotshape(
     plFound and visualMode=='Beginner'?  osc[1] - 10 : na,
     offset=0,
     title="Regular Bullish",
     size=size.tiny,
     color=showPivots ? f_color(2) :na,
     transp=0,
     style=shape.labelup,
     text = 'Buy',
     textcolor= color.black,
     location=location.absolute,
     editable=0
     )


// bearish
plot(
     phFound and visualMode=='Pro'? osc[1] + 10: na,
     offset=0,
     title="Regular Bearish",
     linewidth=3,
     color=showPivots ? f_color(3):na,
     transp=0,
     style=plot.style_circles,
     editable=0
     )
plotshape(
     phFound and visualMode=='Beginner'? osc[1] + 10: na,
     offset=0,
     title="Regular Bearish",
     size=size.tiny,
     color=showPivots ? f_color(3):na,
     transp=0,
     style=shape.labeldown,
     text = 'Sell',
     textcolor= color.white,
     location=location.absolute,
     editable=0
     )



// Delta v ---------------------------------------------------------------------

slope    = f_slope(output_signal)*1.5

// Prediction from Delta v -----------------------------------------------------
output_prediction = f_bias(predictionBias, slope, output_signal)

prediction_bullish = output_prediction>output_prediction[1] and output_prediction[1]<output_prediction[2] ?true:false
prediction_bearish = output_prediction<output_prediction[1] and output_prediction[1]>output_prediction[2] ?true:false

plot(showPrediction and showPredictionCurve?output_prediction:na,title='Prediction Curve', color=f_color(7), editable=0)
//prediction bull
plot(showPrediction?showPredictionPivots?output_prediction>output_prediction[1] and output_prediction[1]<output_prediction[2]?showPredictionCurve?output_prediction:output_signal:na:na:na,
  title='Prediction Bullish',color=f_color(8), style=plot.style_circles, linewidth=2, editable=0)
//prediction bear
plot(showPrediction?showPredictionPivots?output_prediction<output_prediction[1] and output_prediction[1]>output_prediction[2]?showPredictionCurve?output_prediction:output_signal:na:na:na,
  title='Prediction Bearish', color=f_color(9), style=plot.style_circles, linewidth=2, editable=0)

// User Aggreement -------------------------------------------------------------

plotshape(userAgreement==false?0:na,title='Welcome', text='Welcome to Orion Algo! Please double click me to enable signals',textcolor=color.black,color=color.white,offset=0,size=size.huge,style=shape.labeldown,location=location.absolute, transp=0, show_last=1, editable=0)
plotshape(userAgreement==false?0:na,title='Welcome', text='Welcome to Orion Algo! Please double click me to enable signals',textcolor=color.black,color=color.white,offset=-100,size=size.huge,style=shape.labeldown,location=location.absolute, transp=0, show_last=1, editable=0)

// Alerts ----------------------------------------------------------------------

alertcondition(plFound,title='1. Bullish (Big Dot)', message='Bullish Signal (Big Dot)')
alertcondition(phFound,title='2. Bearish (Big Dot)', message='Bearish Signal (Big Dot)')
alertcondition(prediction_bullish,title='3. Prediction Bullish (Small Dot)', message='Prediction Bullish Signal (Small Dot)')
alertcondition(prediction_bearish,title='4. Prediction Bearish (Small Dot)', message='Prediction Bearish Signal (Small Dot)')





// Strategy --------------------------------------------------------------------
i_strategy = input(defval='dca long', title='strategy', options=['simple','dca long'])
i_pyramid = input(10, 'pyramid orders')

// Simple Strat
if (i_strategy == 'simple')
    longCondition = crossover(output_signal, output_signal[1])
    if (longCondition)
        strategy.entry("My Long Entry Id", strategy.long)
    
    shortCondition = crossunder(output_signal, output_signal[1])
    if (shortCondition)
        strategy.entry("My Short Entry Id", strategy.short)

// DCA Strat
i_percent_exit = input(2.0,'percent exit in profit')/100
i_percent_drop = input(2.0,'percent drop before each entry')/100

var entryPrice = 0.0
var exitPrice = 0.0


var inTrade = false
var tradeCount = 0
var moneyInTrade = 0.0

if(output_signal > output_signal[1] and output_signal[1]<=output_signal[2] and i_strategy=='dca long')
//if (true)    
    if (inTrade==false)
        strategy.entry('Long',long=true)
        entryPrice:=close
        moneyInTrade:=close
        exitPrice:=entryPrice + (entryPrice*(i_percent_exit))
        inTrade:=true
        tradeCount := 1
        
    if (inTrade==true and close <= (entryPrice-(entryPrice*(i_percent_drop) )))
        //calculate DCA //math is incorrect!!!
        if (tradeCount <= i_pyramid)
            tradeCount := tradeCount+1
            entryPrice:=close
            moneyInTrade := moneyInTrade+close
            exitPrice2 = moneyInTrade / tradeCount
            exitPrice := exitPrice2 + (exitPrice2 *(i_percent_exit)) 
           
            strategy.entry('Long',long=true)
            

if(close >= exitPrice and inTrade==true and output_signal <= output_signal[1] and output_signal[1]>=output_signal[2] and i_strategy=='dca long')
    inTrade:=false
   
    strategy.close('Long')
    
    


// Dashboard -------------------------------------------------------------------


//deltav
deltav = slope

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-25. Collected in the Quant Trading Vault. Unvalidated research: re-test with your own data and costs before trusting it.

Build this with AgenKit

A spec is not a trading system. Paste this into your coding agent and AgenKit runs a specialist team — architecture, test-first build, risk gates, review — with paper trading on by default.

in Claude Code, Cursor, Codex & 4 more

/agenkit build a production-grade trading bot with a backtest engine, risk controls and paper trading by default from strategies/Orion-Trading-Strategy.md

npx agenkit install engineering-kit in your project
Paste the command into your agent and approve each phase
Get tests, backtest, risk controls and a paper-trading switch

Get AgenKit How /agenkit works

Understand it first

Before you run this, read the lesson behind the idea and how to backtest it honestly. Specs like this one are the survivors of an unknown number of trials; deflate their results accordingly.