I have a new system but I need to openly ask for help in testing this one. I have employed numerous testing methods, but I don't have the data necessary to do a full-bore backtest with this.
My system is a simple breakout strategy. It works like this:
1) Use a daily timeframe for your chart - candles, bars, whatever you're comfortable with. I will use GBPUSD for this illustration.
2) Set a 1000-period SMA on the chart (yes, I said a THOUSAND). If you don't have a 1000 day's worth of data in your system, then set the MA to the highest value possible that will still give you a value on screen.
3) Look at the chart at the very beginning of a new candle/bar. You will enter ALL of your orders for the day at this time.
4) The 1000 SMA shows you the LONG term trend. If the previous day closed above the SMA, you are looking to buy. If the previous day closed below the SMA, you are looking to sell. In other words, you are looking to trade with the (very) long-term trend. GBPUSD is currently above the 1000 SMA, so we will be placing buy orders.
5) Write down the day's opening price. For illustration, we will use 1.9984 for GBPUSD.
6) Open 5 stop orders like so:
Buy 0.1 GBPUSD @ 2.0039, SL @ 1.9939
Buy 0.1 GBPUSD @ 2.0049, SL @ 1.9969
Buy 0.1 GBPUSD @ 2.0059, SL @ 1.9999
Buy 0.1 GBPUSD @ 2.0069, SL @ 2.0029
Buy 0.1 GBPUSD @ 2.0079, SL @ 2.0059
7) At the very end of the day (just prior to the close of the candle/bar), you close any open positions, regardless of whether they are positive or negative.
Notice that this is a feathering technique. It feathers into the position from 0 to .5 lots in increments of 0.1, and it feathers back out of the position in an inverse manner.
The first stop order is placed 55 pips above the opening price. The other orders are triggered on each successive rise of 10 pips, until you have a total of .5 lots open when the price reaches 2.0079.
If the price then reverses, you will be gradually stopped out of the entire position, with the first lot dying at 2.0059 and each successive lot dying when the price retreats by another 30 pips.
"55" was not picked as an arbitrary number. Almost all daily bars/candles experience some retracement before reversing course and heading in their "true" direction for the day. 80% of all GBPUSD daily candles/bars since 1998 have experienced a daily retracement of 55 pips or less before heading in the day's true direction. This means that once the day's price has travelled 55 pips, there is an 80% chance that this represents the true direction for the day. This method can be applied to any pair, but the value of 55 must be tweaked for each pair. The exact values are as follows:
USDJPY: 43
EURUSD: 45
USDCHF: 62
USDCAD: 41
AUDUSD: 29
EURGBP: 23
EURJPY: 55
GBPJPY: 78
EURCHF: 27
CHFJPY: 28
GBPCHF: 81
EURAUD: 57
EURCAD: 49
AUDCAD: 33
AUDJPY: 31
NZDUSD: 28
AUDNZD: 39
CADJPY: 30
Under this strategy, results will typically fall under four categories:
1) The price moves against you at, or shortly after, the start of the day. This is a great scenario. It means that none of your orders were ever triggered, so your exposure to this pair is nil.
2) The pair moves strongly in your direction, and then strongly reverses. This is the nightmare scenario. It means that the price moved far enough to trigger some or all of your orders, and then reversed against you. Although this is painful, at least you are protected by the series of progressive SLs.
3) The pair moves strongly in your direction and closes strongly in your direction. This is big-win scenario. It means that all of your orders were triggered and will all be in the money by the time that you close the positions at the end of the day.
4) The pair moves strongly enough to trigger some or all of your orders, but then wobbles around for the rest of the day. Maybe the price closes lower and you take a small-to-midsize loss. Maybe it still closes higher and you realize some profit. Depending upon the price action of the day, one or two of your orders can get stopped out but you can still end up positive for the day.
This is where testing gets tricky. You can't just look at daily bars to figure out what would have happened to your positions. You really need to be going through the data minute-by-minute or, preferably, tick-by-tick, to know for certain which orders would have been opened, which ones would have been stopped out, and which ones would have closed positively.
Can someone here please help?
My system is a simple breakout strategy. It works like this:
1) Use a daily timeframe for your chart - candles, bars, whatever you're comfortable with. I will use GBPUSD for this illustration.
2) Set a 1000-period SMA on the chart (yes, I said a THOUSAND). If you don't have a 1000 day's worth of data in your system, then set the MA to the highest value possible that will still give you a value on screen.
3) Look at the chart at the very beginning of a new candle/bar. You will enter ALL of your orders for the day at this time.
4) The 1000 SMA shows you the LONG term trend. If the previous day closed above the SMA, you are looking to buy. If the previous day closed below the SMA, you are looking to sell. In other words, you are looking to trade with the (very) long-term trend. GBPUSD is currently above the 1000 SMA, so we will be placing buy orders.
5) Write down the day's opening price. For illustration, we will use 1.9984 for GBPUSD.
6) Open 5 stop orders like so:
Buy 0.1 GBPUSD @ 2.0039, SL @ 1.9939
Buy 0.1 GBPUSD @ 2.0049, SL @ 1.9969
Buy 0.1 GBPUSD @ 2.0059, SL @ 1.9999
Buy 0.1 GBPUSD @ 2.0069, SL @ 2.0029
Buy 0.1 GBPUSD @ 2.0079, SL @ 2.0059
7) At the very end of the day (just prior to the close of the candle/bar), you close any open positions, regardless of whether they are positive or negative.
Notice that this is a feathering technique. It feathers into the position from 0 to .5 lots in increments of 0.1, and it feathers back out of the position in an inverse manner.
The first stop order is placed 55 pips above the opening price. The other orders are triggered on each successive rise of 10 pips, until you have a total of .5 lots open when the price reaches 2.0079.
If the price then reverses, you will be gradually stopped out of the entire position, with the first lot dying at 2.0059 and each successive lot dying when the price retreats by another 30 pips.
"55" was not picked as an arbitrary number. Almost all daily bars/candles experience some retracement before reversing course and heading in their "true" direction for the day. 80% of all GBPUSD daily candles/bars since 1998 have experienced a daily retracement of 55 pips or less before heading in the day's true direction. This means that once the day's price has travelled 55 pips, there is an 80% chance that this represents the true direction for the day. This method can be applied to any pair, but the value of 55 must be tweaked for each pair. The exact values are as follows:
USDJPY: 43
EURUSD: 45
USDCHF: 62
USDCAD: 41
AUDUSD: 29
EURGBP: 23
EURJPY: 55
GBPJPY: 78
EURCHF: 27
CHFJPY: 28
GBPCHF: 81
EURAUD: 57
EURCAD: 49
AUDCAD: 33
AUDJPY: 31
NZDUSD: 28
AUDNZD: 39
CADJPY: 30
Under this strategy, results will typically fall under four categories:
1) The price moves against you at, or shortly after, the start of the day. This is a great scenario. It means that none of your orders were ever triggered, so your exposure to this pair is nil.
2) The pair moves strongly in your direction, and then strongly reverses. This is the nightmare scenario. It means that the price moved far enough to trigger some or all of your orders, and then reversed against you. Although this is painful, at least you are protected by the series of progressive SLs.
3) The pair moves strongly in your direction and closes strongly in your direction. This is big-win scenario. It means that all of your orders were triggered and will all be in the money by the time that you close the positions at the end of the day.
4) The pair moves strongly enough to trigger some or all of your orders, but then wobbles around for the rest of the day. Maybe the price closes lower and you take a small-to-midsize loss. Maybe it still closes higher and you realize some profit. Depending upon the price action of the day, one or two of your orders can get stopped out but you can still end up positive for the day.
This is where testing gets tricky. You can't just look at daily bars to figure out what would have happened to your positions. You really need to be going through the data minute-by-minute or, preferably, tick-by-tick, to know for certain which orders would have been opened, which ones would have been stopped out, and which ones would have closed positively.
Can someone here please help?