What Really Turned My Trading Around

[rooi]

@ Relativity

Well done on the forum/thread.

Nearly half way through the thread.will try and get through the rest after the London break before NY open tomorrow .

Looking forward to seeing you become a $billionaire$ as it seems you have combined all the right PA stuffs into your thinking process a.I have never come across the pip wick thing.Well spotted!

Hope all goes well

ROOI:22:

[rooi]

Hi Rel

Would you mind sharing the indi that gives you the news events on the chart.

Still going through he stuff.

Thanks so much for your time and effort!!!

ROOI:22:

[rooi]

Thats ok.

Keep up the good work.

ROOI:22:

[ProEvo]

Hi Rel,

Been reading through your thread for sometime, to get a good overview start to finish.

Im working on improving my excel skills since programming is something im not acquainted with.

Im curious to ask you something with regards to you gathering statistics on time-bars…

What are your fundamental reasons for using time-bars as a means to gather stats?

Have you considered using other types of bars, such as renko-bars (with wicks) or range-bars to acquire stats? Any thoughts on this?

Good thread you have here - very humble and transparent.

Regards,

Evo

[ProEvo]

The moment I got WA right, I archived all my other tools into a corner.

Are you refering to HH/HL & LL/LH? (barros) or something else altogether?

(note not predict! to be desperate to predict the market is the wrong approach).

What would you consider to be the key differences between predicting (attempting to) and not predicting?

---

On a (random) sidenote…have you considered creating a synthetic index where it contains on one chart alone (!) the eur/xxx and the usd/xxx?

As oposed to 2 charts - 1) eur/xxx and 2) usd/xxx, simply unifying them into one. Any thoughts on this?

[ProEvo]

Attempting to predict / high probability = 99.99999%

Statistics -> Observation -> Exploitability/Frequency -> Entry+Exit. (my opinion of course)

But, statistics do change with time, so, identifying its cut-off points across different market ‘conditions’ is key imo…

Question is: what do you look for to know the difference between a good losing trade and a bad losing trade?

As for predicting i also agree, you either predict a bounce or a break…magnitude of such chosen move is the difference, along with its MAE / MFE. Again, down to statistics.

Hmm…so it turns out i cant post pictures until 5 posts…i guess i have 2 to go :20: Then i can show you what i mean along with some more questions regarding your previous post.

No, referring to the recent completion of SB waves. Barros and SB waves are both HH/HL & LL/LH, but Barros uses time momentum. SB waves uses time periods/intervals.

Your going to have to excuse me here since i dont truly read many trading related books. Ive found that a lot of them do in fact contain regurgitated fluff. However (!) i saw you mentioned Ray Barros (Nature of trends) and i will give it an detailed overview.

On a sidenote, who (or what) is SB waves? Im still reading through your thread Excuse me but i havent read or seen this part yet.

[Jeffreyvnlk]

Do you guy notice the shift to Asian session? I usually get in to London but recently i see EUR already moved 60 pips so we dont have many pips remaind for both London and NY.

[Jeffreyvnlk]

This is why I’ve always insist on the weekly ‘trend’ so much.

I think , on the last column, second row, should be MondayHigh to FridayLow

[Silver1surfer]

Fixing minor bugs in SB waves and making the waves join up better

Looks good. Certainly maps out price movement well.

Every little bit of work / study / research on a daily basis helps. Don’t look down on small progress like this! It does add up.

Thats good and on the 25 Sep 12 I noticed that the M pattern was not drawn, is this the right way it should be or do you think the fix above has sorted it.

• Silver

[Silver1surfer]

Gathering statistics off those SR lines plotted by the highest volume candle close for the day.

I like this code template a lot, as I can use it to test other types of SR lines.

fantastic… so you can now see the strongest SR’s … brilliant

And what are you observing… if its acted as a Resistance & Support and then broken once… is the barried removed or is it still valid later ?!

• Silver

[Silver1surfer]

This is what printed out for 25th sept. Don’t quite get u about where’s the M. The fix might had sorted it out.

There’s also the thing about semi established waves getting redrawn into another waveform if certain price levels (high / low) get beaten. To me this is actually a good thing; it shows how price is trying to evolve itself.

For ytd’s case as an example, a D1 wave failed to continue down. When price began to reverse and go up strongly as a reaction, it changed into a U4 wave.

Recently I’ve taken down the SB wave indicator. I suppose I should put it back again soon, especially right now after several good fixes / tunings. I will only post the basic version where the swings are drawn, not the full version with statistical analysis. So that whoever who is using it can see what I really mean regarding ‘price evolving’.

What I meant was the double top reversal but again this was LT not on the Weekly so that was fine.

[Jeffreyvnlk]

Have i missed anything ? What is SB ?
Give me a day to read back this topic to grasp what everyone talk about lol

[ProEvo]

If you are talking about combining the values back into one, then you will end up getting the standard EURUSD chart anyway.

Hmm, didnt think of that :33:. Not quite sure i explained myself correctly.
Gathering each currencies individual ‘tick’, adding them up together, then comparing them to see which one is moving more. For example:

EUR/USD = +1tick
EUR/GBP = +2tick
EUR/CHF = -4tick
EUR/JPY = +1tick
EUR/CAD = -1tick
EUR/AUD = +3tick
EUR/NZD =0tick

EURO TOTAL = +2tick

Then do the same for the xxx/USD charts, get the total:

EUR = +2tick
USD = -8tick

TOTAL = -6tick.

EUR/USD = up 6ticks.

So in a sense its a synthetic chart and the fluctuations are the difference between total sum of euro and usd.

Then gather statistics on this synthetic chart…meh…thinking aloud here. Must admit 80%-90% of my ideas end up in the trash can.

[njokie]

Unbelievable results.

Lower result = retain trend on the inside, less price ticks than seconds. ‘Holds’ well.
Higher result = show trend on the outside, more price ticks than seconds. ‘Move’ well.

Observations and thoughts :

• M60 is the most balanced between the 2 worlds, in terms of position. M30 comes next along with M5.
• M1 and M1440/D1 (!!! surprised) show the trend. The higher the value, the more pronounced PA is in these timeframes? This means reading PA off directly the candles is worth it. This further proves 1-reading D1 candles is important, 2-M1 does show micro level PA well.
• M10080/W1 obviously retains the trend, so one can’t really read off these candles. Have to go into the W1 candle and see what’s inside. But knowing this time period has its own level of importance matters.
• M15 is the next best of in terms of ‘holding’. Because of this, I was thinking of using this as a base alternative to tick data.
• Timeframes that hold well will have key price levels that don’t move too often. But bad for reading live PA.
• Timeframes that move well will be good for reading live PA, but have unstable/always moving/fluid key price levels.

Hi rel i like this statement

• M1 and M1440/D1 (!!! surprised) show the trend. The higher the value, the more pronounced PA is in these timeframes? This means reading PA off directly the candles is worth it. This further proves 1-reading D1 candles is important, 2-M1 does show micro level PA well.

also having learnt how to use volume in hourly charts from petes thread its really helping me with my trading

[ProEvo]

Hey lol don’t say that. This is a sound idea worth testing. I only need time to code it and see what I can find lol.

The only hurdle is that tick data on MT4 can be a problem. There is tick data available for MT4, but I really don’t want to split hairs… yet!

I agree with your approach, as my charts do tell me there is a difference in tick volume when comparing say EUR vs USD. So if tick volume is saying something, then its worth checking the ticks.

Yes tick data (high quality) is key to implement this concept, nonetheless it is of interest to pursue.

Im learning about statistics et all, and im very confused about the purpose of standard deviation/bell curves, any way you could explain in layman terms to me, or, direct me to a website that speaks in layman terms please?

Thanks.

[ProEvo]

My thinking process of this particular situation. Going to map this process into the engine.

Hi Relativity,

Its funny im seeing this here, because, this is precisely what im working on to build into my conceptual trading model. But (!) im doing it in a slightly different way:

Market movement can occur at any given point in space and time and it can still go either up or down.

These individual movements when combined give birth to all types of patterns (like yours here) and swings.

So price movement evolves from a spec into a pattern and into swings.

Curious to see your evolution here Rel, your onto something very very powerful (!!!) :53:

[njokie]

I normally trade from daily charts (daily chart) i entered the eur/jpy from the pin bar shown but i wanted to show how true volume is in 1h
am combining it with price action (hourly chart) the pin in e/j 1h with high volume showed there were more buyers and has gone up some pips since the pin with high volume
rel u made me reveiw my classes on volume i had abandoned them thenx

[ProEvo]

Nice. What kind of working model do you have on your own currently? Maybe I can help / exchange notes with you. IMO the hardest part is really about coming out with a stable (!!) model; one that maps out relative patterns so that its much easier for the trader to digest.

Hey Rel,

I dont have any working model as of yet, all i have is a conceptual model im putting together (plausible [?] theory).

Once organized my aim is to gather statistics, but for this i need to learn it properly (hence the comment on learning excel) and then probably build the adequate tools to highlight the stats.

I dont think i have anything of real value (proven) that i can share with you, just being honest and humble.

But to give you a general overview of what im doing to see if it is of interest to you (or similar/different):

Market movements look the same on a visual, but the underlying properties of each movement all differ.

• These differing properties can be isolated.

• Market phases/conditions can be isolated.

A trend with low volatility will produce signals that are different to a trend with medium (or high) volatility.

A trend with low volatility will produce false signals (these can be isolated) and real signals (these can be isolated).

Same goes for low, med, high ranging conditions, exhaustion phases…

So in essence im looking to dissect how each market phase/condition produces false/real signals and the underlying elements that highlight its differences.

Its a big task, but it is possible with the right mental approach/thought process, especially when one knows how to code and gather the important stats from the above.

Dissect - Classify - Gather Stats - Identify entry/exit - Implement into trading model - Rinse and Repeat.

Im very edgy sharing this in the open, yet not many will understand what ive written either

[colmex]

@ProEvo

Once organized my aim is to gather statistics, but for this i need to learn it properly (hence the comment on learning excel) and then probably build the adequate tools to highlight the stats.

hey ProEvo,
if you’re looking to learn statistics look into coursera.org
They’re free online university lectures and they have a huge array of subjects. Right now I’m doing the statistics one which is where I got the idea to make a mysql database of ticks and market data (futures & options open volumes, country macro-indicator news, etc.) and then use R (the program they teach to do statistical analysis in the coursera statistics course) to analyze the data from the database.

They’re pretty good courses, the way it works is they post around 2-3 hours of video lecture per week and then have a few homework questions each week (which get marked) and then they have a midterm and a final. And there’s a forum so you can ask other people in the class and the teacher questions if you didn’t understand stuff from the videos.

@Relativity

The major hurdle I would see you go through, at least according to my experience, is getting tools up that will help you gather price action / quickly map out historical market movement easily and in a very organized manner. This is really the hardest part as one has to creatively come up with something, but not too overboard until PA gets distorted… Traditional tools don’t make the cut : this is why you see me working hard on the aspect of tool creation a lot.

I agree 100%. Also considering tools that work for one person aren’t necessarily the best fit for other people. Its good to see other people’s ways of tools, but to be 100% honest I stopped using the code that relativity posts a few weeks ago and I just make my own things in either Mt4 or excel because you can make it show exactly what you need, nothing more nothing less. And of course you learn more about the market.

Just posting for two reasons

-add my input to thread (see above)

-update on what I’m working on;
Learning more statistics with the coursera course, learning unix/linux, learning mysql, learning R.
so far have a debian computer running with mysql and R, this week my task is to get the data from mysql into R, also halfway done coding to get tick data into mysql.

I’ve put learning more about the history of countries in depth off for now as I think I might take a part time course in university next semester or during the summer and I feel I have enough on my plate for now, though I am casually reading a history book about europe right now.

---

also relativity I have a few questions about the swing pattern analysis you’ve been doing that arose out of me also trying to find patterns in swings and wondering a few things about how you were doing it, but I’ll email or pm you about it so check expect a pretty long email with quite a bit of questions about it in the next 48 hours haha

[Kyubi]

Hi Rela,

Good to see this thread coming back to life again after you announced that you’ll retreat from the thread. It’s amazing how much new interesting concepts this thread provides! Unfortunately I can only follow a fractual of information provided here.
Since you are working so much with volume, I was wondering how reliable volume is in the forex market? The volume data you work with is based on your broker and since forex is not centralised, the volume you get represents only part of the whole market volume. I am sure you are aware of this.