Bronco Capital's ABCD Harmonic Pattern Journal | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Gartley/bat/crab/butterfly/abcd/three Drives Pattern Indicator Help 50 replies

Harmonic trading-harmonic patterns 527 replies

The Bucking Bronco 6,547 replies

Harmonic Pattern Journal 79 replies

Harmonic Pattern Journal 8 replies

Attachments: Bronco Capital's ABCD Harmonic Pattern Journal
Exit Attachments

Bronco Capital's ABCD Harmonic Pattern Journal

  • #41
  • Feb 25, 2013 4:36pm Feb 25, 2013 4:36pm
Continued...

Magenta EURUSD 1min--
http://farm9.staticflickr.com/8518/8...b060558b_b.jpg

SkyBlue EURUSD 1min--
http://farm9.staticflickr.com/8096/8...62787b4c_b.jpg

I know some will be asking about why did I take this trade
and not that one...Answer is, my trades are nearly random,
although the market action must fulfill the wave + 61.8% retrace,
but I am not cherry picking or filtering trades. I am looking
for waves and trading them, no thought involved. Yes, I could
likely do better with some discretion, but I could also do worse,
and I could drive myself bat-crap crazy in the process. Right
now it is just about one trade and the next.
  • #42
  • Feb 26, 2013 4:57pm Feb 26, 2013 4:57pm
Grind, grind, grind, but also getting geared up to go live.
Again traded the 78.6% AB stop and reverse pattern with
both going to full 100% AB targets. Nets out 4 wins out of
6 total trades, although I did f'up an entry costing me a
chunk of demo profits (best place/time to make these errors).

Ledger--
http://farm9.staticflickr.com/8378/8...4c5a238c_b.jpg

Yellow EURUSD 1min--
http://farm9.staticflickr.com/8529/8...9e4076bb_b.jpg

SkyBlue EURUSD 1min--
http://farm9.staticflickr.com/8514/8...7ee97981_b.jpg

Aqua5 EURUSD 5min--
http://farm9.staticflickr.com/8236/8...f9e39c59_b.jpg

Pink EURUSD 1min--
http://farm9.staticflickr.com/8088/8...cf7aac50_b.jpg

6 trades away from 100, although it does not really matter,
I still have plenty of work before going live. In the
meantime I will continue trading and collecting data.
  • #43
  • Feb 27, 2013 3:37pm Feb 27, 2013 3:37pm
Still waiting for 1 more trade to hit 786BC or ABfromC, but better
to get results posted while I have a second. Another great day, but
still having order entry mistakes. Simultaneous entry into two platforms
is a pain and shows me just how important it is to get automated.
Luckily, I know how to do it. In the meantime, an unfilled order at
EURUSD 4.31102 cost me about 1R and roughly 1/3 of the days profits.

Ledger--
http://farm9.staticflickr.com/8226/8...a86e4d7d_b.jpg

Plum EURUSD 1min--
http://farm9.staticflickr.com/8370/8...613feaa5_b.jpg

Pink EURUSD 1min--
http://farm9.staticflickr.com/8231/8...a4e62917_b.jpg

White EURUSD 1min--
http://farm9.staticflickr.com/8369/8...c66366f5_b.jpg

Green5 EURUSD 5min--
http://farm9.staticflickr.com/8372/8...38cfd4a6_b.jpg

Cannot complain. Usually get a incredibly good or incredibly bad run
before going live. But I am still 2 weeks out, at least before thinking
about putting on a live trade.

I will probably update full system stats and updates over the weekend.
  • #44
  • Feb 28, 2013 3:57pm Feb 28, 2013 3:57pm
Ughhh, mental note, late night of beer and basket ball =
multiple mistakes the next, hungover morning. Cost myself
1R, again, when I entered a wrong order (stop vs. limit). I
have to get programming and make this automated.

I also got to experience the worst case scenario on the 78.6AB
stop and reverse trade. See the Blue ABCD where the market
retraced through 786AB only to return to 0AB. That will be a
-2R loss. Although, there is a possible trade to counter-act that,
but with only 6 instances out of 100, I am not willing to implement
it yet. Still need more data.

Crossed 100 trades today, also noted I have made 4 trades every
day this week and all have been 2 winner, 2 losers. 50% is great
for this system.

Ledger--
http://farm9.staticflickr.com/8237/8...49f45802_b.jpg

Aqua5 EURUSD 5min (this was ABCD mentioned here)--
http://farm9.staticflickr.com/8369/8...8e376eee_b.jpg

Green EURUSD 1min--
http://farm9.staticflickr.com/8233/8...f7e03cb9_b.jpg

Blue EURUSD 1min--
http://farm9.staticflickr.com/8389/8...57e95738_b.jpg

Silver EURUSD 1min--
http://farm9.staticflickr.com/8379/8...9070af61_b.jpg

Will trade tomorrow then look at the statistics. This week only improves them
I believe. Still have programming to do and then I can start thinking about
trading live with 0.5% risk / trade or something similarly safe.

More to come...
  • #45
  • Mar 1, 2013 12:44pm Mar 1, 2013 12:44pm
Here we go. No trades today, rather I dove into the data to see what
100 trades tells me.

Here are all the trades--
http://farm9.staticflickr.com/8378/8...24cbf114_b.jpg

Basic System--
The basic system was to enter a trade at 61.8% retracement of a manually
identified wave. The target was a full move back to point B with a stop
at point A. If this does not make sense, go review the ABCD pattern.

The target gives a return of 0.618*AB and the stop is -0.382*AB. Just to be super
clear, if the length of AB was 10 pips, a win would nominally be worth 6.18 pips and a
loss -3.82 pips. That gives a reward:risk of 6.18/3.82 or 1.6178.

So, I completed 102 trades with these specifications. There were 40 winners and
62 losers, giving a win rate of 39.22%. Given nominal 1.6178 return, I *would have*
gained +2.712R. 'R' is how much I risk, which is nominally 1% of my account balance.
So I would have returned roughly 2.7% this month.

Here are simulated equity curves with these parameters (39.22% and 1.6178 R:R)--
http://farm9.staticflickr.com/8248/8...f12d26e7_b.jpg

Those results are weak at best. But with 102 data points, we can see what worked
and what didn't. As I saw when I analyzed 75 trades, the 78.6% level basically
divides winners and losers. If that level is breached, the trade will turn into a winner
only 14% of the time, the other 86% of the time, the market will continue to the stop.

So, if I move my stop from point A (100% AB) to 78.6% AB, I greatly shrink my losers
while keeping my winners at 0.618*AB. Losers are now 0.168*AB (0.786-0.618). The
reward to risk shifts to 3.679 (0.618/0.168). In this case, there were 10 trades that hit
78.6% AB then hit target, so those become losers. There would be 30 winners and 72
losers with these parameters (29.41% win rate). Had I traded this, I would have gained
a staggering +38.37R or ~38% return based on 1% risk.

Here is the simulated equity curve with the updated stop (29.41% and 3.679 R:R)--
http://farm9.staticflickr.com/8390/8...9bf41e60_b.jpg

The final tweak offered was actually to stop and reverse at 78.6% AB. Given 62 of 72
trades that hit 78.6% AB ultimately went to A (100% AB), it seems I could flip my position
and catch the ride thereby overcoming the original loss from 61.8% to 78.6%.

BUT, setting the stop for this trade is tricky. Really, it must go at point B, but that means
my reward is 0.214*AB but my risk is 0.786*AB giving 0.2723. The inverted reward:risk is
problematic, especially since the 61.8% trade is risking 1%. In order to make 1% on the
stop and reverse, I must risk 3.673% (1/0.2723). None the less, with a winning % of 86.11%
the stop and reverse is profitable. Had I traded the stop and reverse with these parameters,
I would have gained +25.27R or ~25% using the 3.673% risk / trade.

Here is that equity curve (86.11% and 0.2723 R:R)--
http://farm9.staticflickr.com/8527/8...d9fbac89_b.jpg

Now there is a final update when the stop and reverse is in play, but there are only 6
instances so far (6% of the trades). Given that, I am not going to spell out the details
and will just keep tracking it. If I get 20-30 instances, I will see if it is profitable. Of
course it wins 100% of the time so far! And I challenge you to find this instance in the
trade log above.

As a final treat for you, here are simulated equity curves implementing both updates,
the 78.6% stop and reverse. The simulation first sees if the 61.8% trade is a win or
not. If not, it closes for a loss then sees if the 78.6% stop and reverse is a winner. It
then sums up the two outcomes.

Here are the curves--
http://farm9.staticflickr.com/8239/8...306a34d9_b.jpg

So there you have it. Those equity curves are sick, I am not going to go live yet as I still
need to program some automation, but I will continue to report demo results following
these updates.

There are caveats--

(1) My current automation, when looking at 78.6% level was simulating a
limit order fill, not a stop and reverse type fill. As a result, there are a few times where
a stop may have been triggered. In the end, this could work either way, but suffice to
say, the 102 trades do have a slight bias that I will watch carefully.

(2) The identified waves were manually determined. I may get better or worse at picking
out waves. The markets change, my abilities change. Only by tracking performance can
we see if this system continues to perform going forward.

(3) The system was tested solely on the EURUSD market using 1min to 30min charts.
Step outside these parameters at your own risk.

(4) All the "could have made" values assume perfect entries, 0 friction, and no mistakes.
I made plenty of mistakes during these trades, so I would be very happy with 1/2 the
report R totals.

I will open this thread to comments, criticisms, questions. I have presented a supremely
simple and profitable system. Bastardize it, load it with indicators, and otherwise make
it more complicated at your own risk.

As always, more to come...
  • #46
  • Mar 1, 2013 1:56pm Mar 1, 2013 1:56pm
Very interesting results with a good number of trades done.

Interesting tweaks, and as you say, they have their pros and cons. But the most important thing is what you get at the end, the equity curve, and it seems like it would dramatically increase your performance.

What about those times when you hit te 786 but then the market retrace back, hit the 618, and continue downwards to the A point (Here for example: http://farm9.staticflickr.com/8233/8...f7e03cb9_b.jpg). I think that it could be included as another tweak. You are having an opportunity there to at least leave at the level you entered. But that could happen less often than hitting the A point. If that's the case, disregard this idea.

Have you considered trailing stop your trades once they get to certain targets?
Best time to plant a tree was 20 years ago, second best is now
  • #47
  • Mar 1, 2013 3:53pm Mar 1, 2013 3:53pm
Quoting horuxito
Very interesting results with a good number of trades done.

Interesting tweaks, and as you say, they have their pros and cons. But the most important thing is what you get at the end, the equity curve, and it seems like it would dramatically increase your performance.

What about those times when you hit te 786 but then the market retrace back, hit the 618, and continue downwards to the A point (Here for example: http://farm9.staticflickr.com/8233/8...f7e03cb9_b.jpg). I think that it could be included as another tweak. You...
Ha! I'll make you do the leg work, but the answer is in my data. I should post the xls file so it is easier to search and categorize the trades. As a matter of fact, it will be attached.

The main issue with the "x then y followed by z" type scenarios is that the statistics get dicey due to small sample size. As I say, there is one of those progressions that currently has 100% win rate, but only 6 trades out of 102 followed the pattern. Unless I see 20-30 good examples, I would not trade it. Plus, it is much easier to automate an "x then y" than "x then y, z then a followed by b". Just think of this, if I only traded 618AB entries with a 786AB stop and 0AB target, I would be easily profitable. The stop and reverse at 786AB is just icing on the cake.

Good that you are thinking about possibilities. I would suggest you also think about what the market is telling you as these levels come and go. Visually, what does the progression look like? With respect to highs and lows, what is going on? What would Mr. Pip say about order flow? It is one thing to look at stats, I think it is another to think about possibly why the market is doing what it does. I would never go solely on what I 'think' the market is doing, but having an idea what needs to occur for a pattern to succeed (or fail) is important.

Use the attached spreadsheet (or, better yet, trade and create your own), and share your results!

Bronco

Ah, something else I forgot to add. Going forward, I am going to change how I define point C. The 102 trades used a dynamic C point that moved with the market. Going forward, I am going to assign C to the 786AB level. This creates a constant swing size from which I can size positions and track levels systematically. Similarly, for trades that do not hit 786AB, the 618AB level will be C after 0AB is hit. This is necessary to track the post-target extension or retrace of the market from 0AB. As I mentioned in the post before, the automation does bias the results a bit based on whether bid/ask is used and also due to dynamic C level. Something to think about...never stop thinking!
Attached File(s)
File Type: xls 100 Trades.xls   44 KB | 400 downloads
  • #48
  • Mar 2, 2013 7:11am Mar 2, 2013 7:11am
Quoting BroncoCap
75 Trades and time for some analysis--

So, there is the first analysis of the 75 trades collected so far. I had 6 trades today
which will be posted shortly. 1 was a winner in which the Bid hit 78.6AB, but the ask
did not (= MONSTER PROBLEM), 4 hit 78.6AB and continued to 100AB, and 1 is still
in play w/o reaching 0AB nor 78.6AB.

More to come as always!
Very very very nice work bronco... I love your method of analysys and how you present the results.
How have (if at all) solved the MONSTER PROBLEM?
The results with your last method (with the SL at 78,6 and the stop and reverse) are obtained considering this MONSTER PROBLEM or not?
  • #49
  • Mar 2, 2013 8:42am Mar 2, 2013 8:42am
Quoting lucariga
Very very very nice work bronco... I love your method of analysys and how you present the results.
How have (if at all) solved the MONSTER PROBLEM?
The results with your last method (with the SL at 78,6 and the stop and reverse) are obtained considering this MONSTER PROBLEM or not?
Ah, no, these do not account for the "Monster Problem". That problem is related to the difference in bid-ask along with stop vs limit order functionality. My very first idea was to add a position at 78.6%, therefore I wrote my script to look for the limit-style price (ask for long/bid for short) behavior. Now that I believe stop orders belong at 78.6%, I need to recode the script to handle stop-style price behavior (bid for long stop/ask for short stop). You can see how it was possible to have the price hit 78.6% without my script recording it, if and only if, one of the two prices hit the level.

Anytime I code levels, I try to make it hardest for limits to fill and easiest for stops to fill, as the market always makes it this way. So targets are limit fills at the bid/ask, stops use stop style fills. But to do that, I must pre-define targets and stops. When just 'tracking' or watching a level, it is hard to know which price to use. I could use bid+ask/2 average, but that could complicate matters.

So, short answer, not solved, but will be when I implement the system updates.

Cheers!
  • #50
  • Mar 4, 2013 4:06pm Mar 4, 2013 4:06pm
It is going to be a chaotic week to say the least, too
much on my plate. May not have many trades nor posts in
my journal, but things will be working in the background.

Officially trading the 78.6% Stop and Reverse Update, meaning
both I stop out my 61.8% entry at 78.6% and I also take
the trade expecting the move to continue to 100%.

Worked well this AM as the few minutes I had to view the market
resulted in 1 61.8% win (White), 1 61.8% loss (Silver), and a 78.6%
StopNRev win (Silver). Net is +1.5R. I am still debating on position sizing.
Also, slippage around the 78.6% level resulted in slightly larger loss,
but that is life.

No true ledger, although I still am keeping track, just busy and
exhausted. Charts are below.

White EURUSD 1min--
http://farm9.staticflickr.com/8507/8...b06310c2_b.jpg

Silver EURUSD 1min--
http://farm9.staticflickr.com/8249/8...acc46dc5_b.jpg

More to come, albeit, little by little.
  • #51
  • Mar 4, 2013 6:20pm Mar 4, 2013 6:20pm
I hope it doesn't get that chaotic, but I must say it's an amazing job what you are doing. I wanted to analyze more deeply last weekend the xls and stuff but I couldn't or/and was lazy lol.

There they come a lot of variants to my head, but as you say, it's better having a yes or no scenario and not too much "if, else if, else if, else".

Keep it brutal!!!

Quoting BroncoCap
[i]It is going to be a chaotic week to say the least, too
much on my plate. May not have many trades nor posts in
my journal, but things will be working in the background.

Officially trading the 78.6% Stop and Reverse Update, meaning
both I stop out my 61.8% entry at 78.6% and I also take
the trade expecting the move to continue to 100%.

Worked well this AM as the few minutes I had to view the market
resulted in 1 61.8% win (White), 1 61.8% loss (Silver), and a 78.6%
StopNRev win (Silver). Net is +1.5R. I am still debating on position sizing....
Best time to plant a tree was 20 years ago, second best is now
  • #52
  • Mar 5, 2013 9:41am Mar 5, 2013 9:41am
Quoting BroncoCap
[i]It is going to be a chaotic week to say the least, too
much on my plate. May not have many trades nor posts in
my journal, but things will be working in the background.

Officially trading the 78.6% Stop and Reverse Update, meaning
both I stop out my 61.8% entry at 78.6% and I also take
the trade expecting the move to continue to 100%.

Worked well this AM as the few minutes I had to view the market
resulted in 1 61.8% win (White), 1 61.8% loss (Silver), and a 78.6%
StopNRev win (Silver). Net is +1.5R. I am still debating on position...
Hey Bronco, What you think about implement the SL teached by MR Pip (example 10 pips above/below the A point for a H1 chart), instead of place it directly at A point?
Maybe you can avoid the S&R order at 78.6?

I don't remember exactely the pips of buffer used in a 1 minute chart but i think that your "silver" chart could be a win without the risky Stop and Reverse....
  • #53
  • Mar 5, 2013 9:50am Mar 5, 2013 9:50am
Quoting 4xf
Hi,

Concerning the exit strategy, it seems to me that there is a relationship between the win percentage and the strategy to apply...

I have tried to explain this and it maybe should help you to improve the performance of your EA.

http://izi4x.wordpress.com/2013/02/28/trade-management/

Thanks for sharing this interesting study !
4xf,
Interesting, I will have to take a closer look and see if I can rework what you said into something I could implement. The key issue I have is keeping it simple. Since I trade while working and code everything myself, the easier the system, the faster I can trade it live and the more comfortable I am trading it.

I am still not 100% in favor of the stop and reverse, I think the final tweak to the system lies in that trade.

Cheers and keep in touch!
  • #54
  • Mar 5, 2013 9:53am Mar 5, 2013 9:53am
Quoting horuxito
I hope it doesn't get that chaotic, but I must say it's an amazing job what you are doing. I wanted to analyze more deeply last weekend the xls and stuff but I couldn't or/and was lazy lol.

There they come a lot of variants to my head, but as you say, it's better having a yes or no scenario and not too much "if, else if, else if, else".

Keep it brutal!!!
Ah, chaos is good, to a point. More time to trade today, along with better market action. Will update the journal when I have time. As for analyzing and working over the weekend, it was a lazy weekend for me as well. I should have had the complete system coded and testing, but nope, so I feel your pain (and mental holiday).

Yes, the variants are many, but without solid numbers, I wish to keep it simple. The key to finalizing the system is going to be the stop and reverse. It has a positive expectancy, but the .214AB win vs .786AB loss is horrific. Problem is with a 85%+ win rate, it is hard to know exactly how the loser's price action unfolds. Darn winners! So, much work to be done there.

Keep in touch and keep thinking. Amazing that this has worked as well as it has. With a few final tweaks, I will be ready to rock!
  • #55
  • Mar 5, 2013 9:57am Mar 5, 2013 9:57am
Quoting lucariga
Hey Bronco, What you think about implement the SL teached by MR Pip (example 10 pips above/below the A point for a H1 chart), instead of place it directly at A point?
Maybe you can avoid the S&R order at 78.6?

I don't remember exactely the pips of buffer used in a 1 minute chart but i think that your "silver" chart could be a win without the risky Stop and Reverse....
Lucariga,

You bring up and interesting and important point. First, my original system did place a stop 1 pip beyond point A to give the market some room and there were select trades that needed that room. But, to simply coding I moved it to point A. But, more importantly, if I give the losers an extra pip, I must give the winners an extra pip as well or I diminish my Reward:Risk. I think keeping the R:R is critical.

Just as important, should I use 1 pip, 10 pip, all that requires more testing either live or backtesting. I would probably find that X pip buffer is timeframe or market action dependent. I would rather keep it simple and just use A. As you can see in the post-exit tracking, some blast to 127AB or ABfromC, whereas others return to the 786AB/786BC area. I think it is roughly 50:50, therefore, adding a pip on either side is also likely 50:50.

Keep thinking and following, I definitely welcome the posts!
  • #56
  • Mar 5, 2013 3:30pm Mar 5, 2013 3:30pm
Good start, poor finish--Hard day where I just was not getting the
waves entered at the right time. Missed at least one good entry
while another wave took me for a double 61.8+78.6 StopNRev loss.
More stats though! And I will get a ledger posted sooner than later.
The end of the week looks more promising for these delayed tasks.

Green EURUSD 1min (+1.4R)--
http://farm9.staticflickr.com/8101/8...9bfb41d5_b.jpg

White EURUSD 1min (+1.5R)--
http://farm9.staticflickr.com/8388/8...63b6d643_b.jpg

Yellow EURUSD 1min (-2.0R)--
http://farm9.staticflickr.com/8242/8...a1019cc3_b.jpg

Green and White started the day off great. Yellow had potential but whipped
around and kicked me in the butt. Also in a position that has gone through the
78.6% stop (-0.4R), waiting to see if the StopNRev pays off or not. Also have 4
pending orders for longer timeframe ABCDs. Curious to see how those turn out.

Currently sitting at +0.5R, we shall see what rollover and the later sessions bring.

Cheers.
  • #57
  • Mar 6, 2013 12:42pm Mar 6, 2013 12:42pm
Hey Bronco,
I'm running out the 100 trades, actually I'm at 83 trades. For now the percentage of win rate is stunning: 47%!! I trade during London and London/NewYork Session. Take Profit at 0AB and Stop Loss 1-2 Pip above 786 level, because I've noticed that many trades just hit 786 level and turn back. I just try to follow the micro-trend of the day by using a 100 EMA (very often this MA coincides with the 0.618 level of a swing), however for higher swings it's useless. I'll share the excel file at 100 trades.

Cheers
  • #58
  • Mar 6, 2013 12:49pm Mar 6, 2013 12:49pm
Quoting dibo90
Hey Bronco,
I'm running out the 100 trades, actually I'm at 83 trades. For now the percentage of win rate is stunning: 47%!! I trade during London and London/NewYork Session. Take Profit at 0AB and Stop Loss 1-2 Pip above 786 level, because I've noticed that many trades just hit 786 level and turn back. I just try to follow the micro-trend of the day by using a 100 EMA (very often this MA coincides with the 0.618 level of a swing), however for higher swings it's useless. I'll share the excel file at 100 trades.

Cheers
Spectacular, keep it up. This week's price action is notably different from the last few. Lots of 78.6% retraces that do not go to 100%. On the other hand, trends are not allowing 61.8% retraces. So either you get in during chop or cannot get in during range. I have traded more long term ABCD's this week as well (15min, 30min charts). The key is to just keep grinding and ultimately build a sizable database of trades to determine patterns, statistics, and an edge.

It will be great to see your spreadsheet and how many trades we have in common (although our screen time is different as I am always post-NY open, usually).

Cheers and keep it up!
  • #59
  • Mar 8, 2013 10:40am Mar 8, 2013 10:40am
Some thoughts about this week

For me it has been quite good. Only 12 trades (of which 7 on monday): 5 win, 7 loss based on rules = 41,7 % win rate, very good. Altough I didn't follow the rules for one trade, due to very good expectation for $ on Non Farm Playrolls . You can see in the figure, I didn't close the position, so for me it was a 50% win rate. I suggest you to take a deep look on news, because I think that a confluence of Aggressive C Zone + a good news (for the currency we're taking position) has a very high probability of success (even 90-100% win rate). It's just a thought, only statistic will tell us if I'm right.

Good weekend!
Attached Image (click to enlarge)
Click to Enlarge Name: Chart_EUR_USD_Hourly_snapshot.png Size: 30 KB
  • #60
  • Mar 8, 2013 12:29pm Mar 8, 2013 12:29pm
Quoting dibo90
Some thoughts about this week

For me it has been quite good. Only 12 trades (of which 7 on monday): 5 win, 7 loss based on rules = 41,7 % win rate, very good. Altough I didn't follow the rules for one trade, due to very good expectation for $ on Non Farm Playrolls . You can see in the figure, I didn't close the position, so for me it was a 50% win rate. I suggest you to take a deep look on news, because I think that a confluence of Aggressive C Zone + a good news (for the currency we're taking position) has a very high probability of success...
Ha! Yeah, I was in that trade, full loss as the 78.6% stop and rev got me. Full synopsis of the week coming up. Lots to show, lots to learn, markets change, being able to see, adjust, and overcome is key. Even still, I had a win rate of only 30.8%. One trade, a short, had the bid go to target, but the ask did not and the fill did not occur, became a full loss. Stuff happens!

Hope to get the summary up today, it was chaos week, to say the least!
1 trader viewing now