Reg: GBPJPY, I was long in fact yesterday (along with GBPAUD) but then exited late night after noticing that the daily candle was forming to be a dark cloud cover pattern (indicating short side bias). Taking this into account and the larger JPY position (ex: AJ/NJ/CJ, all heading lower), I just reversed my position.
In effect with GJ & Philip’s method, the short at the first stoch cross would have been a valid entry anyways.
So right now, i’m short in GBPJPY and exited my GA longs as I expect a retracement at a better level than my previous long entry.
Sorry for the bother, but my modeling #'s are not coming out as I expected, but I fear that I’m not doing something right. Assuming that the Fib extensions are not getting you out of a trade (and I know that you personally ignore Initial Stop if stochastics are working in your favor), algorithmically, how can a trade be exited?:
If opposite EMA crossover happens during trade by itself?
If Stochastic crossover at opposite end of entry happens during trade by itself?
If # 1 happens first, followed by # 2?
If #2 happens first, followed by # 1?
How many of these 4 scenarios above alert us to exit trade?
If we neglect fib extensions. We exit a long position when the following happens:
The 20 Ema crosses below the 50 EMA.
Stochastic moves to overbought area (80+)
Stochastic crosses below its moving average.
As I said since you are testing the system on the stock market, I’d only trade long entries. The idea here is the trend in any stockmarket is always up and this is a trend following system.
Not trading GBPCAD
GBPCAD, could be short as failure to rise above previous support at 1.91351 shows a temporary resistance being formed. The short term trend line should be breaking soon.
Thanks for all of your help here. I just finished automating the Fib Extension Trailing Stop logic, and this is quite powerful, and I also implemented your the 3 criteria Exit strategy noted above. You are totally on the mark with this being a LONG system with stocks - preliminary results indicate Long Trades at 60.13% winnable, and Short Trades at only 38.37%, using a 10% of equity position sizing strategy. Equity curve quintuples over 23 years on Long Side; roughly breaks even on Short Side.
I need to proof all my logic, comment it up a bit, and I hope (TBD) to post my code and results sometime this weekend on this board.
One thing I need to still look at (and possibly address) is that every trade over the 23 years of my study (Dow-30 Daily) have the Initial Stop Loss being hit on the first bar after position entry, which seems like I have some sort of logic flaw. I know you never let your Initial Stop Loss be hit (unless trade goes against you), but does this seem viable for a trending system like this?
Also, on these Initial Stop Losses being hit, I notice that the predominant characteristic of these trades is that the respective LOW value (on a Buy) and HIGH value (on a Short) are very tight relative to the trade point of entry. I verified this correctness from my charts. Given this, is there some sort of optimization on entry that you might think is plausible to prevent these kinds of trades from happening (maybe the delta between HIGH/LOW and price of entry has to be > than some percentage of price?).
Anyway, any of your thoughts on these would be greatly appreciated. This system looks really cool.
I’m afraid I will need to see a chart as I don’t understand how my initial stop can be hit the very next candle on every possible trade. I looked at Dow 30 just to see what you mean but there was no sign of what you said.
[QUOTE=“PhilipPirrip;688054”] I’m afraid I will need to see a chart as I don’t understand how my initial stop can be hit the very next candle on every possible trade. I looked at Dow 30 just to see what you mean but there was no sign of what you said.[/QUOTE]
Philip, I’m not explaining it correct. what I mean, is that for all the trades that were stopped out based on the Initial Stop (LOW for Buys; HIGh for Shorts)’ in those cases the Initial Stop was hit on the 1st candle after entry. I never got stopped out on the Initial Stop on 2nd or later candles (after entry). I had Trailing Stops hit on many trades, based on Fib Extension algorithm, on later candles. Do you find this odd or flawed, that Initial Stops were never hit on 2nd (or subsequent) candles?
Also, on those 1st candle Stops, the High/low was very close to price of entry; hence, the likelihood of those stops being hit. So, as a 2nd question, should we not take trades when the High/low is real close to entry price?
We only enter at the close of the 4hr candle don’t we? Looks like at the close of this next 4H candle on MT4 (in 3 hours and 35 minutes) we could have entries for AUDNZD long, GBPNZD long and NZDCAD short.
AUDNZD very close to signaling long
NZDCAD, could take another 2 - 3 candles (thats about 8 - 12 hours approx)
GBPNZD, waiting for signal (should be within the next 4 hours)
NZDUSD, waiting for sell 2 - 3 candles maybe or more
dubadat, yes all H4 TF only.
Philip & the rest… I noticed that when using the Fib tools, signals occur at 38.2% retracement and doesn’t breach 23.6%.
Those using this method could apply this and notice. It might help in ‘anticipating’ a reversal.
Here’s an example for you guys. What do you think?
