Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
This was posted to my private journal for trade occurring on March 28.
The system entered another short this week on GBPJPY at 20:01 chart time (on March 28) at 138.422 and exited by way of normal/primary exit logic of the system on the open of the very next bar at 00:00 (on 3/29) at 138.123 for a +299 pts gain. The SL was set to 139.147 (-725pts loss), and TP was set to 137.213 (+1,209pts gain). From the screenshot, you can see that the price continued in a positive direction even after the trade exited, and then price bounced almost exactly off the bottom line of the contracting pennant trend line I had previously drawn on the chart. This trade was the largest winner (point-wise) other than the two trades that hit their TP.
This trade had opened and closed without me seeing it and I formed no conclusion about whether I should close it. However, if it had still been open when I observed the trade, it would have been around +635pts and everything in the world would have told me to exit if it had not already exited....because of the contracting pennant formation (or whatever you call it).
I am also curious about this trade opening and then closing through the normal trade logic on the very next bar. That is an unusual occurrence for this system, and I don’t believe I have ever seen that in any permutation of parameters. However, it yet again shows this system loves when there is sustained trends formed through reasonably-sized uniform'ish bars. If we have two or three bars in a direction that fits with the longer term trend (downward in this case), then the system will open a trade. This system is not the overall most efficient trend system but the equity curves seems to be relatively flat so far without large DD. This is probably the point where things should blowup.
Can you help answer these questions from other members on NexusFi?
Ok, so I have gone back and posted all my old journal entries as of now. Anything from here on will be new stuff having occurred after I started this FIO journal.
As the first new post, I wanted to post a broader picture of the GBPJPY and the consolidating trend lines. I drew these trend line after the "first trade" on GBPJPY but before the "second trade" on GBPJPY. You can go back and look at the two GBPJPY trades and see the screenshots I took around the time of those trades. You should see these same red trend lines in the photos. But this gives the bigger picture. According to the levels of the trend lines, it should break south when it finally breaks through.
I really liked how the price action bounced off those lines. One of two things will likely occur from here, on GBPJPY. Either we won't get any trades at all due to the rangebound-ness (which would be good) or we will get loser trades that may go so bad as to hit the SL. I will be keeping an eye to see how this currency pair plays out over the next two weeks.
We've had no other trades since the last GBPJPY trade, which closed on the very opening bar of March 29 (midnight GMT time). The system is within expectations so far. If I ran this system "hard", meaning with a poor-decision-making level of leverage, it could easily be up 10% account value over the last 2.5 weeks. And of course I know counting my chickens before they hatch will cause me utter ruin, not to mention that level of leverage and risk-ratio is obscene. However, on that point, if I have done my math right in my head just now, a moderate level of risk would have netted a 2% gain in account value. Again, this is not nearly long enough or enough trades to draw conclusions. So I should just be quiet now.
Yes yes, I wholeheartedly agree. A pin bar is the exact pattern I was thinking in my head, without calling it correctly. And I agree with your thought about putting it in context, but doing so would require a more discretionary approach rather than trying to code it in a script...at least for now. I could see this being a key pattern to look for when considering a discretionary exit on a system-entered trade.
That is definitely the advice I would give. But, of course I want to believe I'm different and my system is different...and the rules don't apply to me. Garbage. This is where psychology comes in to automated system development (at least one of the many). I will continue to let the trades run over the weekend while I am testing it. And if I ever allowed this system to open late Friday trades and hold over weekend on a real money account, I would have to manually check the system on late Friday afternoons and on Sunday evenings...to make discretionary exits as needed. Allowing the script to do that requires direct oversight. ..like I said, I want to believe all my systems are "special"...I need to realize they aren't and never have been.
So this system has not had any additional trades during last week, but I wanted to make a couple posts regarding currency pairs this system is trading.
First, is the EURUSD. If you look at the 4-hour chart, you may ask yourself why a short trade wasn't entered this week. In fact, it would have been a pretty decent trade the script had. However, the parameters of my system--in essence--just barely missed the trade. This script was very...very...close to having entered a short trade this week, that is, where I have the blue arrow pointing on the attached chart. But the system did not enter a trade at that particular spot due to the parameters I use for the EURUSD pair.
The conundrum with which I am faced is whether this is a good thing or bad thing. Given that my goals for this system are to flatten the equity curve, I must conclude the failure to enter a short trade is likely good in the overall picture. But, of course, I want to believe this right now. The reason I want to believe this is that, arguably, if the system would have entered a trade where the blue arrow is pointing, the system likewise would have entered similar setups that would not have panned out. In other words, allowing this good trade would likewise (arguably) allow more bad trades that would be overall negative for the production of this system. I don't know that this is true, but I believe it to be true. I could be wrong, but I see nothing to indicate that I am wrong. ...lol jeez.
Btw, I should add that...if the sideways move preceding the blue arrow had lasted just a little bit longer, then the script would have entered a short trade. The longer sideways movement creates a greater likelihood of success. This sideways move wasn't long enough, which is also what she said about that guy over there.
If you recall, I was discussing this consolidating pattern with the prediction that it would bust south...not north. Well, my systems do not depend on me correctly predicting the direction of a breakout but let's take a look at what has transpired.
I have attached a screen shot that shows the price has closed and proceeded just above the north end trend line. But it has not actually "broken out." So what does this mean? I still think the south move is the overall most likely result. But I have no money betting on that, and unless you have money on a particular direction...you're opinion is worthless. And if you do have money on a particular position, then your opinion is tainted. Opinions are like a--holes, everybody's got one. Including me.
This is AUFCHF, which is another trade that the script *almost* entered. It would have been profitable if the script had entered a long trade where the arrow is pointing. I am not sure whether this is overall good or bad, but I stick with me overall good belief, for now. If we have a large drawdown over the next two weeks, then I might admit I was wrong and this trade (along with the missed EURUSD trade) would helped to smooth out the equity curve.
I think there was no way to predict this bounce/breakout based solely on price action, although hindsight is 20/20.
I think there appears to be a lot of directional shifts in the currency market as of this past week. I am currently working on a mean-reversion system for equity-index futures because I believe the indexes will move fairly sideways over the course of the summer until we get a better idea of what tax reform will look like in the US. The summer is always fairly sideways, but unless we know what tax reform will look like, we have no way of knowing whether current pricing of equities is valid...as many have already said. Anyway, my observations regarding this particular script tend to agree with that overall sentiment...sideways until we know what's up with tax reform.
I think the 'failure' to enter the trade was neither a good or bad thing. It just didnt enter in that particular set of circumstances and hindsight curve fitting is not going to help you (arguably). You dont want to become a victim of overfitting your strategy to accommodate historical data. Read this article on Wikipedia, I reckon it help address some of the uncertainties in your mind wrt your algo.
Oh I hear ya, I am well aware of the dangers over fitting a system to historical data. That's what I was alluding to. For me, though, the problem is not simply knowing (or not knowing) the dangers of curve fitting, but is continuously fighting against those tendencies towards "bad behavior"....which is what I keep going back to: the psychology of algorithmic trading.
Over fitting is just one example of what I will call "bad behavior" in system development. Unrecognized (and unaccounted-for) bias in our beliefs about the markets and what the markets "should" or "will" do (i.e., what is a sound edge) is another example of mine. And going about the optimization process incorrectly, which is largely the same as curve fitting, but kind of different.
The psychological hurdles of system development, I think, are mostly the same hurdles that discretionary traders face, at least at the underlying level. Hell, all psychological hurdles that all humans face in every endeavor in mankind's history are arguably the same at the underlying level. But the hurdles manifest themselves in slightly different forms in each endeavor. In other words, I think the battles are the same but the battle grounds are different...with system development v. discretionary trading.
I still think the "failure" to enter a trade at that particular instance (in my prior post) was a good thing precisely because I believe it tends to indicate the system is not overly fit to historical data...at least that is possibly what it indicates but I could be wrong and I really cannot know whether I am truly wrong or not. There is no way to know, except by looking at your long term performance. This belief could represent an unknown bias I am not aware of. But, at some point, I just gotta say enough overthinking is enough.
....pie-eyed soapbox rant of the day: a good man knows where to draw the line, which necessarily means he also knows where not to draw the line. As applied to trading, this means a good man is the one who knows exactly what to do and what not to do, at the correct time. But I think a wise man knows that the answers, as to when to draw the line and when not to draw the line, only come from within. ...okay wow, I am done, I need to go to sleep now.
Peace man, thanks for the post and keeping me on my toes!
Btw, still no trades with this system since the last one I posted above...and that's running it on 13 different currency pairs still. Serious no trend time.
But I did want to post two more shots of the GBPJPY chart, with the consolidation pattern. Basically, we had a fakeout to the north. I had predicted the breakout would ultimately occur in a southerly direction, and that still seems probable to me. I feel this pattern is fairly text book, and we will likely see a lot of consolidation in all markets in general as summer is approaching and people are trying to see what's going on in the US (tax reform) and UK (Brexit)..yadda yadda yadda.
And I think the fakeout essentially broadens the contraction zone, which will prolong the time for which the contraction will occur before the ultimate breakout occurs. I wonder whether we will see one more fakeout (false breakout) before the ultimate breakout. I don't know, but I'm still sticking with my original call of a southerly breakout. If I was setting a discretionary entry, I would set a sell stop entry at around 137.200 or 137.000 maybe. But I am not actually putting money on that bet so my opinion is worthless (unless one of my systems pounces on it).
And I have two photos, both of the same chart but the second photo shows what I am referring to with the broadened contraction zone...i.e., just a differently angled trend line.
