When 337 pips isn’t enough

The last few days I’ve been watching an uptrend in GBPJPY. My interest was sparked when my Shock and Retrace strategy, on the 1 minute timeframe was triggered. If you’ll recall, FXCM asked me to run the strategy at 1m timeframes to gather some data, which isn’t what the strategy was designed for but I managed to backtest some good scenarios.

So here’s the email alert automatically sent by Shock and Retrace strategy on trade:

Signal: SHOCK_RETRACE(GBP/JPY.m1,(3, 1, 48, 25, true, true))
Symbol: GBP/JPY
Time: 2013/11/12 20:33:02
Time (UTC): 2013/11/12 09:33:02
Message:
SHOCK_RETRACE:2013/11/12 20:33:02 Enter BUY 1000@158.20300 Bid: 158.16700 Ask: 158.20300 TakeProfit: 161.57200(3.40500) Stop: 157.48600(0.68100)

That’s a stop loss of 68 pips and a profit target of 340 pips. That’s huge, and highlights one of the flaws in backtesting. Backtesting can very quickly scan over hundreds or thousands of trades and it is all too easy to look at the bottom line and go “Look, it made money, must be good”… Deeper analysis should be required, and a profit target of 340 pips is rather unrealistic.

Good thing I’m still forward testing, hence the 1000 amount (1 mini-lot). Regardless, I’m now in it, and I’m curious to where it will go. Is there a good reason behind this large profit target or not that is multiples greater than the size of the original spike we detected?

Well in the end the trade did very well and the profit target was reached last night. Here is the email alert from The Watch add-on, minus my account and trade number, when my trade was automatically closed.

Signal: THE_WATCH
Symbol: GBP/JPY
Time: 2013/11/16 02:02:43
Time (UTC): 2013/11/15 15:02:43
Message:
The Watch:Trade Closed ([TRADE_NUMBER])
Instrument: GBP/JPY
PnL: 337.00
Gross: 35.89
Open: 158.20300
Close: 161.57300
Amount(k): 1k
QTXT: SHOCK_RETRACE(GBP/JPY.m1,(3, 1, 48, 25, true, true))
Account: [ACCOUNT_NUMBER]
Close Time: 2013/11/16 02:02:44
Close Time (UTC): 2013/11/15 15:02:44

And here’s a screenshot (ignore the other close and reopen in the middle, that’s different strategies running on the same pair)

337 Pips on GBPJPY

337 Pips on GBPJPY

It turns out that often a huge spike in the price can be the start of a trend. I guess, if I think about it, this may make sense if some economic news comes that immediately shifts the fair value of the currency. That will of course lead to a gap or spike in pricing, which could be read as the starting line to a new phase in the economy’s strength.

If this is the case, then it is fine to not only expect the spike to retrace, but to extend further. Furthermore, this is actually something I had observed before when playing around with trailing stops on this strategy. So it does seem there is something at play here that we should look to take advantage of. However, overall, this trade didn’t sit ok for me for the following reasons

  • You don’t get many 300 pip opportunities very often. Ideally, on a run like this the strategy should have scaled in to take further advantage of it.
  • Perhaps the trend will continue but we are already on the sidelines. Perhaps a trailing stop would be better to keep our winning trade going.
  • What if it had got to 250 pips and then turned around and then made a loss? What a waste of pips! Some kind of partial profit or at least moving the stop to break-even would be good.

I’ll think about some general improvements for the strategies. If anyone has any suggestions or comments, please let me know!

Posted in Backtesting, Research, Retracement, Scaling, Strategies, Trend and tagged , , , , , , , , .

One Comment

  1. Pingback: General Strategy Improvements | MooMoo Forex Trading

Leave a Reply

Your email address will not be published. Required fields are marked *