English EN
View on GitHub

SwingTrader Strategy

Overview

The SwingTrader Strategy is a StockSharp port of the MetaTrader 4 expert advisor SwingTrader.mq4. The original EA looks for Bollinger Band reversals: when price bounces from the outer band and the next bar crosses the middle line, the advisor opens a position and starts building a martingale-style averaging grid. The translated strategy reproduces the same high-level behaviour using StockSharp candles, Bollinger Bands from StockSharp.Algo.Indicators, and the framework's order helpers (BuyMarket, SellMarket). Volume scaling, the width of the grid and the liquidation rules mirror the MT4 code while respecting the exchange limits provided by the Security metadata.

Trading logic

  1. Subscribe to the configured timeframe (CandleType) and feed a Bollinger Bands indicator with BollingerPeriod length and a fixed standard deviation multiplier of 2.
  2. Work only with finished candles; the indicator callback ignores partially formed bars to replicate the MT4 IsNewCandle() guard.
  3. Track whether the previous candle touched the upper or lower band. The boolean pair _upTouch / _downTouch follows the original toggling logic that keeps only one side active until the opposite band is touched.
  4. When no basket is open:
    • open a long position if the last completed bar crossed above the middle band after previously touching the lower band;
    • open a short position if the bar crossed below the middle band after touching the upper band. The first order volume equals InitialVolume (after exchange rounding) and the initial grid width equals the latest distance between the upper and lower Bollinger bands.
  5. When a basket exists, watch for adverse movement of one full band width from the very first fill:
    • for longs, if the candle's low is at least one band width below the anchor price, buy another slice whose size is multiplied by Multiplier with each new level;
    • for shorts, if the candle's high is one band width above the anchor price, sell an additional slice using the same multiplier logic.
  6. Keep aggregating new orders until either the profit or the maximum tolerated loss target is hit.

Money management and exits

  • The helper CalculateUnrealizedProfit reproduces the MT4 floating PnL calculation by converting price differences to price steps (Security.PriceStep) and step value (Security.StepPrice).
  • The invested capital proxy uses the original formula Lots * Price / TickSize * TickValue / 30, where Lots becomes the sum of grid volumes and the tick parameters are sourced from Security.
  • Close the entire basket once the floating profit exceeds TakeProfitFactor * invested capital.
  • Force an emergency liquidation when the floating loss reaches 10 * TakeProfitFactor * invested capital (same ratio as the MT4 code).
  • All exits are executed with market orders in the opposite direction; once flat, the grid state is reset and new touches must be detected before another entry can trigger.

Parameters

Name Type Default Description
TakeProfitFactor decimal 0.05 Multiplier applied to invested capital to define the profit target.
Multiplier decimal 1.5 Volume multiplier for every additional averaging order.
BollingerPeriod int 20 Number of candles used by the Bollinger Bands indicator.
InitialVolume decimal 1 Base volume of the first trade in a new basket (rounded to venue limits).
CandleType DataType 15-minute timeframe Timeframe used for signal generation.

Differences from the original EA

  • StockSharp works with net positions; the strategy maintains explicit lists of grid entries to emulate MT4's ticket-based order handling.
  • Exchange volume filters (Security.MinVolume, Security.VolumeStep, Security.MaxVolume) are applied automatically instead of manually calling CheckVolumeValue.
  • Signals are evaluated on closed candles; intrabar triggers from the MT4 version are approximated by using candle highs and lows for averaging decisions.
  • Orders are always sent as market instructions, whereas MT4 used OrderSend with explicit bid/ask parameters.

Usage notes

  • Provide realistic metadata for the traded instrument: PriceStep, StepPrice, MinVolume, VolumeStep and MaxVolume must be populated for the profit, loss and volume calculations to match the MT4 behaviour.
  • Because the averaging grid scales geometrically, test the configuration on historical data and consider broker margin requirements before running it live.
  • The grid width equals the current Bollinger Band width; changing BollingerPeriod directly affects both entry timing and grid spacing. Validate the sensitivity during optimisation.
using System;

using Ecng.Common;

using StockSharp.Algo.Indicators;
using StockSharp.Algo.Strategies;
using StockSharp.BusinessEntities;
using StockSharp.Messages;

namespace StockSharp.Samples.Strategies;

/// <summary>
/// Simplified from "SwingTrader" MetaTrader expert.
/// Uses Bollinger Band touches to detect swing direction, then enters on middle-band cross.
/// </summary>
public class SwingTraderStrategy : Strategy
{
	private readonly StrategyParam<DataType> _candleType;
	private readonly StrategyParam<int> _bollingerPeriod;
	private readonly StrategyParam<decimal> _bollingerWidth;

	private BollingerBands _bollinger;
	private bool _upTouch;
	private bool _downTouch;
	private decimal? _prevClose;
	private decimal? _prevMiddle;

	public DataType CandleType
	{
		get => _candleType.Value;
		set => _candleType.Value = value;
	}

	public int BollingerPeriod
	{
		get => _bollingerPeriod.Value;
		set => _bollingerPeriod.Value = value;
	}

	public decimal BollingerWidth
	{
		get => _bollingerWidth.Value;
		set => _bollingerWidth.Value = value;
	}

	public SwingTraderStrategy()
	{
		_candleType = Param(nameof(CandleType), TimeSpan.FromMinutes(5).TimeFrame())
			.SetDisplay("Candle Type", "Timeframe for signals", "General");

		_bollingerPeriod = Param(nameof(BollingerPeriod), 20)
			.SetGreaterThanZero()
			.SetDisplay("BB Period", "Bollinger Bands period", "Indicators");

		_bollingerWidth = Param(nameof(BollingerWidth), 2m)
			.SetGreaterThanZero()
			.SetDisplay("BB Width", "Bollinger Bands deviation", "Indicators");
	}

	protected override void OnStarted2(DateTime time)
	{
		base.OnStarted2(time);

		_bollinger = new BollingerBands { Length = BollingerPeriod, Width = BollingerWidth };
		_upTouch = false;
		_downTouch = false;
		_prevClose = null;
		_prevMiddle = null;

		var subscription = SubscribeCandles(CandleType);
		subscription
			.BindEx(_bollinger, ProcessCandle)
			.Start();

		var area = CreateChartArea();
		if (area != null)
		{
			DrawCandles(area, subscription);
			DrawIndicator(area, _bollinger);
			DrawOwnTrades(area);
		}
	}

	private void ProcessCandle(ICandleMessage candle, IIndicatorValue bbValue)
	{
		if (candle.State != CandleStates.Finished)
			return;

		if (!bbValue.IsFinal)
			return;

		if (bbValue is not BollingerBandsValue bbVal)
			return;

		if (bbVal.UpBand is not decimal upper || bbVal.LowBand is not decimal lower || bbVal.MovingAverage is not decimal middle)
			return;

		if (!_bollinger.IsFormed)
		{
			_prevClose = candle.ClosePrice;
			_prevMiddle = middle;
			return;
		}

		var close = candle.ClosePrice;

		// Track Bollinger touches
		if (candle.HighPrice > upper)
		{
			_upTouch = true;
			_downTouch = false;
		}
		if (candle.LowPrice < lower)
		{
			_downTouch = true;
			_upTouch = false;
		}

		if (_prevClose is null || _prevMiddle is null)
		{
			_prevClose = close;
			_prevMiddle = middle;
			return;
		}

		var volume = Volume;
		if (volume <= 0)
			volume = 1;

		// Buy: had a lower band touch, now price crosses above middle
		var buySignal = _downTouch && _prevClose.Value < _prevMiddle.Value && close > middle;
		// Sell: had an upper band touch, now price crosses below middle
		var sellSignal = _upTouch && _prevClose.Value > _prevMiddle.Value && close < middle;

		if (buySignal)
		{
			if (Position < 0)
				BuyMarket(Math.Abs(Position));

			if (Position <= 0)
				BuyMarket(volume);

			_downTouch = false;
		}
		else if (sellSignal)
		{
			if (Position > 0)
				SellMarket(Position);

			if (Position >= 0)
				SellMarket(volume);

			_upTouch = false;
		}

		_prevClose = close;
		_prevMiddle = middle;
	}

	/// <inheritdoc />
	protected override void OnReseted()
	{
		_bollinger = null;
		_upTouch = false;
		_downTouch = false;
		_prevClose = null;
		_prevMiddle = null;

		base.OnReseted();
	}
}