PineScript 5策略延迟K线入场后仓位立即平仓问题求助

问题根源 #

你的代码中，startprice、pm3、takeprofit、stoploss的计算依赖当前K线的goLong/goShort状态。当延迟1根K线入场时，触发信号的K线已经过去，当前K线的goLong/goShort为false，导致：

• pm3计算错误（会取1 / math.sign(strategy.position_size)，但刚入场时仓位还没更新，可能得到错误的方向）
• startprice取到的是更早的信号价格，和入场价格不匹配，导致止损止盈价格直接触发平仓

解决方案 #

需要在触发信号时，把对应的入场基准价和交易方向存储下来，延迟入场时调用这些存储的值来计算止损止盈，而不是依赖当前K线的信号状态。

修正后的MACD策略代码 #

//@version=5
strategy('Moving Regression Band Breakout strategy', shorttitle='MRBS', overlay=true, default_qty_type=strategy.percent_of_equity, default_qty_value=3)

issl = input.bool(title='SL', inline='linesl1', group='Stop Loss / Take Profit:', defval=true)
slpercent = input.float(title=', %', inline='linesl1', group='Stop Loss / Take Profit:', defval=5, minval=0.0)
istrailing = input.bool(title='Trailing', inline='linesl1', group='Stop Loss / Take Profit:', defval=false)
istp = input.bool(title='TP', inline='linetp1', group='Stop Loss / Take Profit:', defval=true)
tppercent = input.float(title=', %', inline='linetp1', group='Stop Loss / Take Profit:', defval=2)
//
fastLength = input.int(title='Fast Length', defval=12)
slowLength = input.int(title='Slow Length', defval=26)
signalLength = input.int(title='Signal Length', defval=9)
[macdLine, signalLine, _] = ta.macd(close, fastLength, slowLength, signalLength)

longCondition = ta.crossover(macdLine, signalLine)
shortCondition = ta.crossunder(macdLine, signalLine)

// D-RSI signals
goLong = longCondition
goShort = shortCondition

// 存储触发信号时的基准价和方向
var float stored_startprice = na
var int stored_dir = 0 // 1=多, -1=空

var float position_size = 0.0
if strategy.position_size != 0
    position_size := strategy.position_size

var nb = 0

//Entrys

if goLong
    if position_size < 0
        strategy.close('Short', comment='Exit S', qty=math.abs(position_size))
    nb := 1
    // 记录多单信号的基准价和方向
    stored_startprice := close
    stored_dir := 1
    label.new(bar_index, low, 'C', xloc=xloc.bar_index, yloc=yloc.price, color=color.rgb(200, 250, 64), style=label.style_label_up, size=size.small)

if nb[1] == 1
    nb := 0
    alert(syminfo.tickerid + ' Long Signal', alert.freq_once_per_bar_close)
    label.new(bar_index, low, 'L', xloc=xloc.bar_index, yloc=yloc.price, color=color.green, style=label.style_label_up, size=size.small)
    strategy.entry('Long', strategy.long)
    // 用存储的值计算多单止损止盈
    long_tp = stored_startprice * (1 + stored_dir * tppercent * 0.01)
    long_sl = stored_startprice * (1 - stored_dir * slpercent * 0.01)
    strategy.exit(id='s/t L', from_entry='Long', stop=issl ? long_sl : na, limit=istp ? long_tp : na, alert_message='Close L')

if goShort
    if position_size > 0
        strategy.close('Long', comment='Exit L', qty=math.abs(position_size))
    nb := 2
    // 记录空单信号的基准价和方向
    stored_startprice := close
    stored_dir := -1
    label.new(bar_index, low, 'C', xloc=xloc.bar_index, yloc=yloc.price, color=color.rgb(209, 255, 3), style=label.style_label_up, size=size.small)
  
if nb[1] == 2
    nb := 0
    alert(syminfo.tickerid + ' Short Signal', alert.freq_once_per_bar_close)
    label.new(bar_index, high, 'S', xloc=xloc.bar_index, yloc=yloc.price, color=color.rgb(255, 0, 0), style=label.style_label_down, size=size.small)
    strategy.entry('Short', strategy.short)
    // 用存储的值计算空单止损止盈
    short_tp = stored_startprice * (1 + stored_dir * tppercent * 0.01)
    short_sl = stored_startprice * (1 - stored_dir * slpercent * 0.01)
    strategy.exit(id='s/t S', from_entry='Short', stop=issl ? short_sl : na, limit=istp ? short_tp : na, alert_message='Stop Loss S')

关键修改说明 #

1. 新增stored_startprice和stored_dir变量，在触发goLong/goShort信号时，记录当时的收盘价和交易方向。
2. 延迟入场时，调用存储的基准价和方向计算止损止盈，避免依赖当前K线的信号状态。
3. 将strategy.exit移至延迟入场的代码块中，确保每次入场都使用对应信号的止损止盈参数。

内容的提问来源于stack exchange，提问作者redbluefan