Ratios & Harmonics: a Different Way to Trade | Page 2379 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

"Ratios & Harmonics: a Different Way to Trade." Summary 37 replies

Learning to trade with harmonics 11 replies

Time for Change - Pivots & Harmonics to freedom 67 replies

180 reasons to trade Harmonics and Patterns 13 replies

Group Trading based on Elliott wave and Harmonics 0 replies

Attachments: Ratios & Harmonics: a Different Way to Trade
Exit Attachments

Ratios & Harmonics: a Different Way to Trade

  • #47,561
  • Aug 25, 2019 6:41pm Aug 25, 2019 6:41pm
Quoting S41nt
{quote} I guess I am just thinking out loud and trying to take the subjectivity out of the pattern so that I can solidify something in my trade plan that I can strictly follow. I think we are talking about the same thing but I am trying to work out when to invoke which set of rules. I think the first picture shows both 2 and 3 but please correct me if I am wrong. The first picture A would be like your example 2, trading the 1:1 as a correction, with the BE at NS* and the TP at the 1:1/127.2 extension. Picture B, if i am understanding you correctly,...
Let’s keep it simple.

1. Against order flow, follow pattern BE and TP rules

2. With order flow, follow 1 to 1 or 127 fib rules. You can follow pattern rules on this if you want also
  • #47,562
  • Aug 25, 2019 6:46pm Aug 25, 2019 6:46pm
Quoting standrdtrde
{quote} Let’s keep it simple. 1. Against order flow, follow pattern BE and TP rules 2. With order flow, follow 1 to 1 or 127 fib rules. You can follow pattern rules on this if you want also
Hmmm I like that, thank you!
  • #47,563
  • Aug 25, 2019 9:05pm Aug 25, 2019 9:05pm
starting the week with this short
Attached Image (click to enlarge)
Click to Enlarge Name: NC_S_20190825-2103.jpg Size: 58 KB


I took wrong swing as my move. I think this is the correct one: The stop is bit much for me..but now I am committed:
Attached Image (click to enlarge)
Click to Enlarge Name: NC_S_20190825-2110.jpg Size: 60 KB


Lets see how this pans out.
  • #47,564
  • Aug 25, 2019 9:44pm Aug 25, 2019 9:44pm
Quoting standrdtrde
{quote} Let’s keep it simple. 1. Against order flow, follow pattern BE and TP rules 2. With order flow, follow 1 to 1 or 127 fib rules. You can follow pattern rules on this if you want also
great advice
  • #47,565
  • Edited 10:19pm Aug 25, 2019 9:58pm | Edited 10:19pm
I've spent some time recently putting together automated backtest models for this system on the YM and I want to share some of those results with the group. First of all let me state from the outset that I'm not here to release an algo nor will I be doing so at anytime in the near future, but I am more than happy to use the algo that I've developed to highlight some statistical expectancies of this system and to help answer any questions you might have concerning how "your plan" affects the expectancy. Note of course that the results posted here represent my ability to code Damian's system and are not meant to represent what Damian or others are capable of producing manually.

1. I have defined swings as closely as is programatically possible to the way Damian defines swings. That being said logic must be used when swings become extended and we must requalify order flow. I'll be happy to discuss how I did that later. No consideration was given to "retests". Swings either projected or failed. A minimum of 40 points was required to define a swing.

2. No consideration whatsoever was given to the entry. For me to program an entry that only a computer is capable of picking up on would be a discredit to the results. For the purpose of these results, entries were made on any 144T bar that closed below the .382 retracement of a swing. Every single swing was traded.

3. The stop was placed 3 points above/below the swing on every trade.

4. No attempt was made to move to BE or to rescue any trade.

5. The profit target was the 1.27 extension or the one to one move whichever came first. Trades were closed before the session ended if they had not been stopped out or reached their targets.

6. Trailing was implemented at worst trailing each candle by 2 points and at best trailing decision candles by 1 point. I'll be happy to talk about programmed decision candles later as well.

7. The results are for the last year (8-25-2018 to 8-25-19) for trades taken between 9:30am and Noon EST

Note that the first screenshot is an example of what the algo looks like on a chart. (the little red dots are trailing stops)
Special thanks to @spetscom and @cstern for their help...
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 116 KB
Click to Enlarge Name: Screenshot2.png Size: 46 KB
Click to Enlarge Name: Screenshot3.png Size: 137 KB
  • #47,566
  • Aug 25, 2019 10:04pm Aug 25, 2019 10:04pm
A few statistics that I want to point out:

1. There was a drawdown period of 47 days - requires patience doesn't it.
2. Even a dumb system that blindly enters in the direction of the expectation is profitable if you take every trade.
3. The average # of trades per day for these trading hours was 3.59. How does your frequency stack up?
  • #47,567
  • Aug 25, 2019 10:14pm Aug 25, 2019 10:14pm
Now, just for the fun of it, here's the same system, with slightly more optimized entries / exits, traded 24 hours/5 days a week excluding lunch and midnight.
Every single trade was taken with no exceptions. 1 contract was traded for every $15,000 in an account starting with $15,000 (min 1 contract, max 10 contracts).

The simple point I hope to make is, trade the expectation, drop your emotionally driven trade management. And if you think you have a good trade management idea, I'll be happy to put it to the automated test for you to the best of my abilities. Thank you to Damian for giving us the keys to the gold machine.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 132 KB
  • #47,568
  • Aug 25, 2019 10:17pm Aug 25, 2019 10:17pm
Quoting bmdenton23
A few statistics that I want to point out: 1. There was a drawdown period of 47 days - requires patience doesn't it. 2. Even a dumb system that blindly enters in the direction of the expectation is profitable if you take every trade. 3. The average # of trades per day for these trading hours was 3.59. How does your frequency stack up?
Hey, I know you put together something on MES for us on discord for the periods that instrument began in May-present but is it possible for you to use the same swing stats you used for the ym one you just made and make it for ES for the year time period just so the people that are on ES could get a statistical map to see where we are heading in terms of points to expect and drawdowns etc
for 9:30 to 12 EST
  • #47,569
  • Aug 25, 2019 10:21pm Aug 25, 2019 10:21pm
Quoting bmdenton23
Now, just for the fun of it, here's the same system, with slightly more optimized entries / exits, traded 24 hours/5 days a week excluding lunch and midnight. Every single trade was taken with no exceptions. 1 contract was traded for every $15,000 in an account starting with $15,000 (min 1 contract, max 10 contracts). The simple point I hope to make is, trade the expectation, drop your emotionally driven trade management. And if you think you have a good trade management idea, I'll be happy to put it to the automated test for you to the best of...
Holy shit I just realized how much more efficiently aggressive Damian’s money management system spreadsheet is compared to this... Mr. Pip knew all along about the profitability of this swing system....
  • #47,570
  • Aug 25, 2019 11:23pm Aug 25, 2019 11:23pm
Quoting bmdenton23
Now, just for the fun of it, here's the same system, with slightly more optimized entries / exits, traded 24 hours/5 days a week excluding lunch and midnight. Every single trade was taken with no exceptions. 1 contract was traded for every $15,000 in an account starting with $15,000 (min 1 contract, max 10 contracts). The simple point I hope to make is, trade the expectation, drop your emotionally driven trade management. And if you think you have a good trade management idea, I'll be happy to put it to the automated test for you to the best of...
This information is awesome and thank you for sharing. I have been going back and testing weeks at a time on the replay and experiencing very good results. The issue I have is the expectancy - for whatever reason. Did not reach my AOI, did not have good rotation, strength was coming into my AOI, etc. After you look back at the day, you realize out of the 7-8 swings made in the AM session, I only participated in 2-3 and this is mostly driven by emotion.

With that in mind, I have been testing the expectancy and thought what would happen if I made the commitment to enter once a proper correction has been made. For the week of July 7, which I just completed, I did this for the intraday order flow taking 1/2 contract (or 5 micros) at each fib levels on MYM - taking in no regard to rotation, structure, etc. 1/2 contract at each level would equal the 2 contracts on YM. It was traded 24 hours for the week versus just the NY AM session. Trade management was the same as my plan, moving to BE, trailing at 127/1-1, etc.

Results are below:

Attached Image


I am hoping to complete more weeks as the expectation failures just fell really well this week. I was able to rescue one, small loss on one and the other 3 were ones where I moved to BE and were winners in some form. However, expectation completions versus expectation failures were right in line at 74%.

The other interesting piece you shared was the draw down for 47 days. Our position size is important in relation to our account value. We have to be able to handle the storm to live to trade another day. Damian's results this year were powerful, which he shared last week. At some point during 2019, it showed he lost 77 consecutive trades and he also had a streak of over 200 winners. If we take an average of 50 points a loss x 77 = 3,850 points. Now, I am no where near Damian, so I would be in the triple digits for losses MM is really important to maintain our accounts. It is kind of like the roulette table. If you are playing black / red, sometimes there is a streak of 10 or 15 red and you are thinking how on earth can this be? In that small window of time it looks very unbalanced, but over the course of a couple days, it is closer to 49% (keeping the 0 and 00 for the house in mind).

There are going to be peaks and valleys, which we are in a bit of a valley right now. Your analysis confirms what Damian is telling us with the small account and the market. Things are going to turn and this streak of consecutive winners is coming if you keep the frequency.

I have moved to the MYM to try and remove the emotional component. I don't like the chart, as it is not as liquid as the YM, but the 55 tick compares reasonably with the 144 / 233 on YM. I am trying to get to the position of "It does not matter if this loses" - NEXT, so I can enter without hesitation to my plan.

Thanks for your analysis and just gives us more confirmation to stick to the plan.
Good work is the key to good fortune
  • #47,571
  • Aug 25, 2019 11:39pm Aug 25, 2019 11:39pm
Quoting Rush2112
{quote} This information is awesome and thank you for sharing. I have been going back and testing weeks at a time on the replay and experiencing very good results. The issue I have is the expectancy - for whatever reason. Did not reach my AOI, did not have good rotation, strength was coming into my AOI, etc. After you look back at the day, you realize out of the 7-8 swings made in the AM session, I only participated in 2-3 and this is mostly driven by emotion. With that in mind, I have been testing the expectancy and thought what would happen if...
There was a comment that said that ninjatrader counts consecutive wins and consecutive losses in a weird way. Mr Pip did not have 77 losses in a row nor did he have a streak of over 200 according to the comment.

It would be good to get clarification from him on this
  • #47,572
  • Aug 26, 2019 1:07am Aug 26, 2019 1:07am
Quoting bmdenton23
I've spent some time recently putting together automated backtest models for this system on the YM and I want to share some of those results with the group. First of all let me state from the outset that I'm not here to release an algo nor will I be doing so at anytime in the near future, but I am more than happy to use the algo that I've developed to highlight some statistical expectancies of this system and to help answer any questions you might have concerning how "your plan" affects the expectancy. Note of course that the results posted here...



One question about entries: the entry to all trades was on the first candle that closed above/below 0.382. Basically this means that there was never an entry in the golden zone, except for the cases when the closing bar was big enough to reach that, and that most of the entries were made between 0.382 and 0.5?
  • #47,573
  • Edited 8:19am Aug 26, 2019 3:02am | Edited 8:19am
Quoting S41nt
{quote} ...Reading your explanation then, for the pattern to be viable, the 61.8 percent cant break past the C point? So does that mean, a 1:1 move with a weaker correction with the C point could be hinting at a potential expectation failure to come?...

No sorry if it confused you. I was trying to give some extra argumentation but it may have been too much detail.
Forget the last part and just keep the first part of the explanation

1) The pattern exists in all timeframes
2) The pattern works as a correction of the orderflow you're trading, in that case your entry is in line with the "expectation".
3) But it also works when the pattern is in line with the trend/order flow you're trading, in that case you're trading a correction of the orderflow you're trading.

Obviously, your target is achieved more easily when you're trading situation 2), as the entry is in line with the "expectation".
In 2) you can have the 61.8 as a target for start of trailing, or you can decide to BE at the 61.8 and look for extended targets. Because after all, you're expecting a NSH or NSL with a 1:1 harmonic projection, which always by far exceeds the 61.8 profit target of the pattern.

In 3) you're trading a correction against the main orderflow you're trading. So it is not reasonable to expect extended targets. With situation 3) it is better to stick with the 38.2 = BE and 61.8 = trailing method.

Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 69 KB
  • #47,574
  • Aug 26, 2019 3:41am Aug 26, 2019 3:41am
This might be pretty much a basic question for those who have been using ninjatrader. On the ninjatrader website it talks about a Free Live Trading License...

This left me confused, has anyone here tried the Free license for actual live trading? where you have to open a ninjatrader brokerage account and (though unconfirmed) you pay for using ninjatrader via higher commissions.

As it stands I can't afford to purchase a license now.
  • #47,575
  • Aug 26, 2019 3:46am Aug 26, 2019 3:46am
Quoting bmdenton23
I've spent some time recently putting together automated backtest models for this system on the YM and I want to share some of those results with the group. First of all let me state from the outset that I'm not here to release an algo nor will I be doing so at anytime in the near future, but I am more than happy to use the algo that I've developed to highlight some statistical expectancies of this system and to help answer any questions you might have concerning how "your plan" affects the expectancy. Note of course that the results posted here...
Hello bmdenton23, it would be interesting to see how the stats differ if you enter at the 0.618 retracement of a swing instead of the 0.382. Thank you.
  • #47,576
  • Aug 26, 2019 4:08am Aug 26, 2019 4:08am
Quoting BlacMoni
This might be pretty much a basic question for those who have been using ninjatrader. On the ninjatrader website it talks about a Free Live Trading License... This left me confused, has anyone here tried the Free license for actual live trading? where you have to open a ninjatrader brokerage account and (though unconfirmed) you pay for using ninjatrader via higher commissions. As it stands I can't afford to purchase a license now.
Free version of NT is BS in my opinion, for live trading. You wont be able to use their chart trader for example which is essential and not included. Basically you can execute but order management will be frustrating since you won't even see them on the chart.

Had this problem for my micro account. Switched to Sierra Chart instead since I was already using it for the YM. SC is $18 - $26 a month with a lot of functionality and better performance than NT. I prefer SC over NT.
  • #47,577
  • Aug 26, 2019 4:32am Aug 26, 2019 4:32am
Quoting Hansl
{quote} Free version of NT is BS in my opinion, for live trading. You wont be able to use their chart trader for example which is essential and not included. Basically you can execute but order management will be frustrating since you won't even see them on the chart. Had this problem for my micro account. Switched to Sierra Chart instead since I was already using it for the YM. SC is $18 - $26 a month with a lot of functionality and better performance than NT. I prefer SC over NT.
Thanks for the feedback
  • #47,578
  • Aug 26, 2019 6:02am Aug 26, 2019 6:02am
Quoting brrand
starting the week with this short{image} I took wrong swing as my move. I think this is the correct one: The stop is bit much for me..but now I am committed: {image} Lets see how this pans out.
Out at BE
  • #47,579
  • Aug 26, 2019 7:30am Aug 26, 2019 7:30am
Quoting standrdtrde
{quote} There was a comment that said that ninjatrader counts consecutive wins and consecutive losses in a weird way. Mr Pip did not have 77 losses in a row nor did he have a streak of over 200 according to the comment. It would be good to get clarification from him on this
This one I can answer. Say he buys 100 contracts. There is probably not 100 available at the same price on the bid so he gets 20 at a price, 20 at another price, and so on. Each of these buys is a separate trade to ninjatrader. You will see it on a chart as multiple entry markers. I've seen it on playback even using 2 contracts in a fast moving candle. So if his 100 contract buy is actually 6 entries, and they all lost, you have 6 losing trades.

Nevertheless, this applies to his winners as well. You can see he has more consecutive winners than losers so still going strong.
  • #47,580
  • Aug 26, 2019 7:33am Aug 26, 2019 7:33am
Quoting fxboss69
{quote} Hello bmdenton23, it would be interesting to see how the stats differ if you enter at the 0.618 retracement of a swing instead of the 0.382. Thank you.
Its worse. You miss out on winners that offset your losers.
15 traders viewing now