Trading with altered price chart: P&F, Range Bar, Renko, Kagi, etc. | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

New Renko Bar and Reversal Renko Bar Indicator Alert Help Please 5 replies

Range bar/Renko bar chart not working 5 replies

Kagi Chart Question 8 replies

Correct Kagi settings on Daily Chart? 2 replies

Kagi settings on the one day chart? 1 reply

Attachments: Trading with altered price chart: P&F, Range Bar, Renko, Kagi, etc.
Exit Attachments

Trading with altered price chart: P&F, Range Bar, Renko, Kagi, etc.

  • #1
  • First Post: Edited 11:17am Mar 14, 2013 5:03am | Edited 11:17am
I haven't yet found a single place to discuss thoroughly what the pros and cons of trading with altered price chart are. I say we make an extensive discussion on it here. I think this is a topic worth discussing about. I'll delve into the details later on. But how's the experience for you guys trading on altered price charts? Anyone here trade exclusively and successfully with altered charts?

General Pro: Seeing the same thing with a new set of eyes. What is not visible on the normal chart often can be more visible on the altered chart.

General Con: The word altered in this case means there are some aspects that are taken off the normal chart, mainly time dimension.
  • #2
  • Mar 14, 2013 5:13am Mar 14, 2013 5:13am
I see following Pros and Cons


Pros

* Sometimes can see the especial patterns/things, etc. with the altered chart that normal price chart don't
* Developing TA skills
* Big change especialy for eyes
* Eliminate the boredom


Cons

*Need to spend additional few hours on the screens
  • #3
  • Mar 14, 2013 5:25am Mar 14, 2013 5:25am
Quoting Ishtana
The word altered in this case means there are some aspects that are taken off the normal chart, mainly time dimension.
An important aspect that is taken off the standard charts is the noise ... or at least part of the noise.
MT4/MT5 EAs/Indicators/Alerts coder
  • #4
  • Mar 14, 2013 5:32am Mar 14, 2013 5:32am
Im a great fan of renko.

Pros:

- Clean and tidy.
- If using indies, they paint the picture with higher efficiency.
- Swings are clearly defined.
- Breaks down nicely big moves which are harder to on time bars.
- Shows quiet periods clearly and eliminates a lot of random noise.

Cons:

- Time isnt shown, but having said that they are expressed via multiple consecutive moves (momentum).
- ...

Unless one is aiming to gather statistical patterns based on time, then i cannot personally see many negatives.

My 2 euros worth.-
  • #5
  • Mar 15, 2013 10:05am Mar 15, 2013 10:05am
Cons
- It is very difficult to check the patterns ( reversal / or continuous )
- Not use to it
- difficult to learn


--------
trading regression channel
  • #6
  • Mar 17, 2013 10:35pm Mar 17, 2013 10:35pm
I think at some point many of us go through the learning process and start to question the basic beliefs that are passed around as truth. Such as, do historical prices really contain information that can look forward into the future?

The type of data aggregation that we use is essentially our lens through which our strategies are filtered. In an engineering book on digital signal processing (DSP), altered price chart would be referred to as a sampling problem. I think there are some insights that can be gained from the topic of sampling as they apply to financial time series.

Sampling converts the independent variable (time) from continuous to discrete. The effect is to ignore price changes between sampling periods. In the case of altered price charts the independent variable is altered from time to some type of range. In the case of a range bar this is an equal range for each bar. For a Renko, the bricks are developed according to an algorithm where the reversal bars are double the size of the continuation bars but where tails are also allowed. These properties of Renko make it a fairly decent input for indicators that rely on stationary price movement. While Renko bars don't make price stationary, there is a higher probability of continuation than reversal due to the way the algorithm works, and thus your wiggly moving average acts less wiggly on a Renko than on just about any other type of chart. This is the reason many system developers favor Renko.

What are the implications of this in terms of sampling? I have no idea other than you give up some of the data. If the data you give up is less valuable than the data you receive through this type of Renko sampling then on balance it is a good trade off.

By way of information, there was a thread on a different forum a ways back that may be relevant to this topic Price Bars
  • #7
  • Mar 17, 2013 10:55pm Mar 17, 2013 10:55pm
More on sampling taken from Chapter 3 of The Scientist and Engineer's Guide to Digital Signal Processing By Steven W. Smith, Ph.D. page 5 from the downloadable PDF. By the way, market data may considered a continuous signal.

The Sampling Theorem
Quote
The definition of proper sampling is quite simple. Suppose you sample a continuous signal in some manner. If you can exactly reconstruct the analog signal from the samples, you must have done the sampling properly. Even if the sampled data appears confusing or incomplete, the key information has been captured if you can reverse the process.

I don't know for sure if any of the bar types typically used in charting packages can meet the requirement of the sampling theorem. But possibly time based charts meet the requirement.
  • #8
  • Mar 20, 2013 8:12am Mar 20, 2013 8:12am
I'm more and more interested in the range bars. Not to remove the noise but to break down the strong moves. But I'm facing some issues with the erasure of the time. A trend line in a regular chart represents the increase of the price per unit of time. What does a trend line mean on a range bar chart? Certainly nothing at all. If I use the range bar chart to spot S/R levels I get the same ones as usual. But a weekly or daily S/R is normally stronger than a hourly level. If I remove the time I can no more make this distinction. I still need the time dimension. I consider using a hybrid approach. I will see where this leads me.
No greed. No fear. Just maths.
  • #9
  • Mar 20, 2013 6:23pm Mar 20, 2013 6:23pm
Quoting PipMeUp
I'm more and more interested in the range bars. Not to remove the noise but to break down the strong moves. But I'm facing some issues with the erasure of the time. A trend line in a regular chart represents the increase of the price per unit of time. What does a trend line mean on a range bar chart? Certainly nothing at all. If I use the range bar chart to spot S/R levels I get the same ones as usual. But a weekly or daily S/R is normally stronger than a hourly level. If I remove the time I can no more make this distinction. I still need the time...
When range bars are used you are correct that the time element is erased. But the x axis doesn't represent nothing, but rather 10 pip ranges for each bar, if that is the size of bar you choose. So to draw an analogy between weekly / daily S/R being stronger than hourly, so thus 40 pip S/R and 80 pip S/R will be more significant than 10 pip S/R in terms of the larger moves. And so a trend line on a 10 pip range bar would represent an increase in the price per 10 pip range.

But I think you make a good realization about the trend line possibly meaning nothing at all on a range bar chart. I've taken that one further and extrapolated that the same is true for trend lines as created via typical TA methods applied to any chart. There is a stark difference between the definition and application of 'trend lines' between TA and the scientific research field. This started my turn from TA based methods to quantitative based methods. PipMeUp, you seem like someone who would naturally embrace quantitative methods over TA methods.
  • #10
  • Mar 21, 2013 4:32am Mar 21, 2013 4:32am
Quoting PipMeUp
I'm more and more interested in the range bars. Not to remove the noise but to break down the strong moves. But I'm facing some issues with the erasure of the time. A trend line in a regular chart represents the increase of the price per unit of time. What does a trend line mean on a range bar chart? Certainly nothing at all. If I use the range bar chart to spot S/R levels I get the same ones as usual. But a weekly or daily S/R is normally stronger than a hourly level. If I remove the time I can no more make this distinction. I still need the time...
If you know how to code, then why not create a time chart overlay over the range chart?

10pt range bars with an overlay of daily timeframe on it perhaps....
  • #11
  • Mar 21, 2013 5:25am Mar 21, 2013 5:25am
Quoting Pipalicious
10pt range bars with an overlay of daily timeframe on it perhaps....
The problem is that if the pair is ranging within the 10pt range, you might have a single candle for few days.

This is just an example ...
MT4/MT5 EAs/Indicators/Alerts coder
  • #12
  • Mar 21, 2013 5:47am Mar 21, 2013 5:47am
IMO, more accurate historical chart, the better. Charts are a good visual aid to help with assessing value based on past price action. Not so good price prediction though.
  • #13
  • Mar 21, 2013 6:41am Mar 21, 2013 6:41am
Quoting EmeraldEyes
IMO, more accurate historical chart, the better. Charts are a good visual aid to help with assessing value based on past price action. Not so good price prediction though.
What im amused to find on this forum is the huge lack of risk management being discussed...its always better charts, better indies, better predictions; when in reality it all boils down to risk management
  • #14
  • Mar 21, 2013 7:39am Mar 21, 2013 7:39am
True Pipalicious,

But I believe this conversation is specifically about charts. I think it would be safe to assume that you already do that on your own. Particularly for me, this charts help me see better. I have gone thru countless attempts at PA on regular charts and every guru I've seen draws a different line in the chart. Many of them do so at the same time and come up with completely different perspectives. I don't care how much you lean on high probability if you are creating your own reference system by drawing the lines that go according to your own view of things.
Please pull my head out of my ass, I am running out of oxygen...
  • #15
  • Mar 21, 2013 7:46am Mar 21, 2013 7:46am
Let me just remark that with that I don't deny the key role of risk management and that no amount of charting and TA would ever overcome your need for risk management. In the end, TA itself should be part of your risk management strategy.
Please pull my head out of my ass, I am running out of oxygen...
  • #16
  • Mar 21, 2013 8:13am Mar 21, 2013 8:13am
Quoting FXEZ
What are the implications of this in terms of sampling? I have no idea other than you give up some of the data. If the data you give up is less valuable than the data you receive through this type of Renko sampling then on balance it is a good trade off.
that's actually a pretty good bottom line.....

renko charts have a definite appeal to me....h
to trade and code, keep both simple... no call to impress....h
  • #17
  • Mar 21, 2013 8:55am Mar 21, 2013 8:55am
Quoting FXEZ
40 pip S/R and 80 pip S/R will be more significant than 10 pip S/R in terms of the larger moves
You are right. I didn't think about it. I didn't plan to use them to find S/R.

Quoting FXEZ
There is a stark difference between the definition and application of 'trend lines' between TA and the scientific research field. This started my turn from TA based methods to quantitative based methods. PipMeUp, you seem like someone who would naturally embrace quantitative methods over TA methods.
What about somewhat quantifying the TA methods? It's uncanny how TA works while objectively I see no reason why... Yet it works! On the other hand have a look at daily trends but on H4, near the end. On some of them you will see the oscillations on the chart increase in amplitude while the period decreases. Compare to a Von Karman alley. As if energy conservation applied to the market.

That's why I try to see the price through different prisms. Altered price chart is such a prism to me. You get another point of view.

Quoting Pipalicious
What im amused to find on this forum is the huge lack of risk management being discussed...its always better charts, better indies, better predictions; when in reality it all boils down to risk management
I tried. I failed. http://www.forexfactory.com/showthread.php?t=411923
Looks like people prefer believing in the reward rather than mesuring the risk...

Quoting Pipalicious
If you know how to code, then why not create a time chart overlay over the range chart?
10pt range bars with an overlay of daily timeframe on it perhaps....
Here are H1 and D1 overlaid with range 20 pips. The red/green is the range chart, the blue/yellow is the time chart.
Attached Image (click to enlarge)
Click to Enlarge Name: H1_over_RG20.png Size: 116 KB

Attached Image (click to enlarge)
Click to Enlarge Name: D1_over_RG20.png Size: 146 KB
No greed. No fear. Just maths.
  • #18
  • Edited 11:57pm Mar 21, 2013 11:40pm | Edited 11:57pm
Quoting Ishtana
But how's the experience for you guys trading on altered price charts? Anyone here trade exclusively and successfully with altered charts?
I love range bars (well, my own slightly tweaked version)! For almost 3 years now I have traded EURUSD using 7-pip range bars. My system is automated, so the visual aspects of my "altered" charts are irrelevant to me. The range bars encapsulate the moves in a "cleaner" fashion and make reversals and continuations more obvious (at least in a mathematical way).

Quote
General Pro: Seeing the same thing with a new set of eyes. What is not visible on the normal chart often can be more visible on the altered chart.

General Con: The word altered in this case means there are some aspects that are taken off the normal chart, mainly time dimension.

There is certainly still a time dimension to the altered candles. It's just that the time dimension is varied instead of uniform.

I can't say I've used P&F, Renko, or Kagi for my trading. I'm sure they can be also be useful, depending on the application. I'd highly recommend thinking outside-the-box when considering the type of charts to use. There are so many different ways to be successful in trading and I personally think range bars are a perfect fit for my way of trading.

Regardless what the gurus say, you don't have to trade H4 and above...you don't have to use trend lines and support/resistance...you don't have to throw all of the indicators out the window. You CAN be successful trading smaller, variable, or (dare I say) infinite timeframes. Just make sure you have good entry/exit criteria and you'll enjoy the ride.
Attached Image (click to enlarge)
Click to Enlarge Name: eurusdm907.png Size: 77 KB
  • #19
  • Mar 29, 2013 4:35am Mar 29, 2013 4:35am
I've been meaning to make a thorough research on this topic, but just haven't had a chance just yet. But I'm just gonna put my thoughts here before it goes away.

I find all of them to be very interesting. Looking at the altered chart alone, I can easily see why some people use them. And I realized that even the default price chart IS a filtered price chart. Filtered by the broker. Real price chart is what the banks see on their chart. Someone with an API access is the closest thing they can get to the real thing. Anyway, I have 4 objections to price altered chart that stop me from trading them.

 

  1. The subjective choice of box size. Why 8? Why 10? Why any other number? What's the basis of that choice? Personal? Doesn't it mean you curve-fitting your altered chart?
  2. Most indicators either do no work or their actions are altered. This can be an advantage rather than a disadvantage. But a simple 200 EMA will cross the price at different time and levels on different altered chart.
  3. Candlestick patterns are invisible from these charts. That means price action study is not accessible, and I've always been a believer that price is king.
  4. Time is just a minor objection. I notice in range bar, if you put your cursor to the bottom window where the time/date display is at, you can tell when the new range bar is opened. Besides you can just watch a clock to get your time bearing, especially in expectant of news releases. Nevertheless, time is not continuous on these charts.

  • #20
  • Mar 29, 2013 4:43am Mar 29, 2013 4:43am
Quoting Ishtana
The subjective choice of box size. Why 8? Why 10? Why any other number? What's the basis of that choice? Personal? Doesn't it mean you curve-fitting your altered chart?
This is the same question you have to ask yourself when you choose to trade M5 rather than H4.

Quoting Ishtana
Most indicators either do no work or their actions are altered. This can be an advantage rather than a disadvantage. But a simple 200 EMA will cross the price at different time and levels on different altered chart.
Exactly like a 200 EMA on different standard timeframes.

Quoting Ishtana
Candlestick patterns are invisible from these charts. That means price action study is not accessible, and I've always been a believer that price is king.
They are not the same, but patterns are present on offline charts. Activate the wicks (at least on renko) and you will even see patterns better.
MT4/MT5 EAs/Indicators/Alerts coder
0 traders viewing now