Iterative Refinement

Discussion in 'Journals' started by Spydertrader, Jan 3, 2008.

  1. Let me guess ....

    You don't see a 'Traverse Level' signal for change. As such, you can't see how one would know to reverse long at 11:55.

    This is a calibration issue. Your brain simply hasn't hard wired itself yet to 'see' these things as they unfold when market pace drops off significantly. In other words, you haven't created a rule based on the current context, either because you can't 'see' the correct context, or you have yet to correctly differentiate what you see he from something else.

    No worries. The market provides for this situation as well.

    Please note the attached snippet. Follow the 'faster fractal' traverse (Orange) - noting the gaussian changes ('skinny' red and black lines) within these Orange trend lines. Note how the gaussians work on the fractal itself (pink and brown).

    See if this orientation provides for a scenario where you can locate the signal for change under your current calibration mode. :) Please note how during periods of depressed pace the market often provides signals which appear to represent two different paths or two different 'views' to the market. Understand, only one view represents the 'correct' view. The market always provides the signals required to know which represents the accurate rendering. However, the reality is in this specific example both views place the trader back onto the right side of the market.

    The annotations to which you refer do not hold the same meaning that you have attributed to them. The annotations represent something close to your conclusion, but different enough, that your interpretation moves you off track.

    HTH.

    - Spydertrader

    <img src=http://www.elitetrader.com/vb/attachment.php?s=&postid=2173983>
     
    #9091     Nov 12, 2008
  2. You simply started your lateral in the incorrect location. Without a doubt, you believed the market had signaled something you have seen many times before. However, what the market actually provided was only slightly different, but different enough to provide an entirely different thing.

    Start by 'double checking' all your annotations. You'd be shocked at how easy it is to accidently place an incorrectly drawn line on a chart, and by doing so, shift one's mindset a country mile.

    - Spydertrader
     
    #9092     Nov 12, 2008
  3. ehorn

    ehorn

    Hi Spyder, Thanks for responding to my question earlier. As I debrief tonight and compare my annotations to yours I notice you have annotated a case of 2 bars 14:10-14:15 as an internal (Equal High/Lower Low). Do you treat this case (addition to the 7 cases) as an internal?

    Thanks for any insight.
     
    #9093     Nov 12, 2008
  4. As always, the answer here depends on the context.

    - Spydertrader
     
    #9094     Nov 12, 2008
  5. Thank you very much for taking the time to respond.

    Yes, I was able to recognize in real time the market signalling the Point 1 in the Green accelerated up traverse. (Blue p3)

    There is absolutely no doubt about that in my mind. The blue trendlines are there, whether I like it or not. ( An indirect proof of that is how the last pink down traverse development, while interacting with blue trendlines, was affected by the blue up traverse.)


    I believe your use of the word ‘calibration’ in this context is different from how Jack Hershey and Mak wrote about calibration referring to bar spacing, volume levels, chart scaling and how volatility affects the width of channels.

    If you could expand on what you mean by calibration, I would greatly appreciate it. Specifically, how do I fix this calibration issue you are referring to?


    In real time, seeing the 11:35 OB arrive, I made a mistaken conclusion that the sequence for the pink traverse was completed and permission now exists to seek change. Even the OB bar itself signals ‘Long’. Two more bars of decreasing volume appear to provide the Jokari Window change, but the spike (11:50) and more increasing red volume is not something one would see if any of the previous bars had signalled change. Therefore I incorrectly dismissed the possibility of pink traverse being true. However in this case I suspect the issue here is not one of the completion of sequence or change signal but something else entirely. As you said it’s close to my conclusion, yet different enough.

    Ever since I started paying attention to Bar dominance and Pace (at least where pace acceleration exists) along with monitoring drills – I have found that my ability to understand what the market does and why it does it improved tremendously along with my ability to see change signals (although not always on the same fractal).

    But in situations like this one, it feels as if I am missing some key variables (aside from sequence completion, dominance, and pace acceleration).

    What are the other variables that would allow one to see those four bars for what they are (from pt3 into traverse) in real time?
     
    #9095     Nov 12, 2008
  6. charts

    charts

    There are other possibilities ...
    ... the change comes 25 minutes later
    ... or pt 1 ... 35 minutes earlier
    They may not be when the market said the loudest "change" ... but would've worked fine ... :)
     
    #9096     Nov 12, 2008
  7. I don't agree. The market speaks loud on every single bar. I just like to learn to understand it's language, if you will, that's all.
     
    #9097     Nov 12, 2008
  8. In this specific context, my use of the word 'calibration' refers to your brain, and not as Jack and Mak have often inferred, to bar spacing scaling or Pace Levels. You (often quite clearly) 'see' things in a single fractal, yet you often miss that which to others, appears quite clear. You can 'see' the sequences on a ES 5 minute Traverse level, but the sub-fractal movements (or 'faster fractal' traverses) sometimes pass you by. In other words, if the market 'jumps a fractal' you have not anticipated the event.

    In addition, you have not yet fully differentiated exactly that which signals change. Make no mistake here. You have the vast majority down cold (and no doubt can pull a link to support this knowledge within seconds if asked), but again, your knowledge, while close, isn't quite complete. It's this last little step which currently eludes you, and which causes these areas of confusion for you currently.

    By example, you (most likely) still have yet to fully understand why the 10:15 bar (recently linked by dkm) isn't screaming change, even though, you know such a thing to represent change during most every other context. Something in this specific example is different from those which do represent change. You must clear this confusion from your mind. Until you do, the confidence required for you to 'know that you know' will remain just out of reach.

    Now, you are not alone in this boat. Whether or not others realize they are sitting next to you, represents another matter entirely. :)

    Everything begins with monitoring. Take a look at the chart you posted earlier today. You have increasing volume across your Gaussians annotated across an area represented by a lateral. Do laterals move in the dominant or non-dominant direction? Gaussians must match the trend lines. Do yours?

    Compare your chart to mine. Between the two of us, who has more annotations? For nearly two years, I have repeatedly placed an enormous amount of emphasis on correct and thorough annotation. Perhaps, a reason exists behind creating a chart which contains, full, complete, correct and thorough annotations. :D

    Ya' think? :D

    - Spydertrader
     
    #9098     Nov 12, 2008
  9. Yes, thanks, that clears it up for me. I think you really have to see the "point 3-ness" of the market at all fractals before it stops being theoretical. And distinguishing between different fractals, while often very clear, can sometimes be very tricky.

    On yesterday's chart, for example, the 1:05 far was a change signal - but did it mean resumption of the lat toward a new point 3, retrace down to the blue RTL, or the end of the up retrace and beginning of the down traverse. In context, with sequences complete, the latter seemed most likely, but I wasn't certain .

    And the subsequent brown down traverse on your chart I read as a completed down traverse (built of tapes, with R2R happening with the BO of the blue traverse), so that I read point 2 (your chart) as change, leading either to a widening of the traverse or the start of an up channel. That was jumping fractals, but even with extensive review I find it hard to be certain of the fractal in real-time.

    Anyway, thanks for taking the time to respond; that did help.

    -palinuro
     
    #9099     Nov 13, 2008
  10. I too am in this boat.

    Understanding each bar and its unique role in the overall context of the trend is a difficult challenge and one that [for me at least] has taken an awful lot of time and focus to clear up. Spyder refers to areas of confusion, to me these are "confusions of the correct context" that each bar represents. I am generally at the point where I may not know the reason why something is about to happen, but I know it is about to happen. Which is a substantial improvement for me. There are still periods where I believe we are at a certain point ,but the market has thrown a couple of curve balls to shake my confidence. 14:15 to 14:55 was such a time as a result I made poor decisions. These are becoming more rare {thankfully}.

    The point is the reason that everyone hasn't already mastered the method is because we haven't cleared up the areas of "confusion in context" which provides the confidence that MADA was correctly done.

    -thoughts from the guava
     
    #9100     Nov 13, 2008